Early Drainage of Macula-Involving Massive Suprachoroidal Haemorrhage Assisted with Recombinant Tissue Plasminogen Activator May Lead to Better Visual Prognosis Open Access

Suprachoroidal haemorrhage (SCH) is a sight-threatening complication which can occur secondary to intraocular surgery [1], melanoma [2], trauma [3], or very rarely spontaneously [4]. It can cause significant morbidity due to pain and visual impairment-associated hyphaema, vitreous haemorrhage, retinal detachment, and ocular hypertension [5]. A SCH is termed “massive” when it is large enough to expel the intraocular contents through the open wound or bring the retinal tissues into apposition (“kissing” SCH) [1] and carries a poorer visual prognosis [6]. The management of a SCH depends upon its size and symptoms. A localised asymptomatic SCH is likely to be spontaneously absorbed [7]. A massive SCH (MSCH) has a high risk of loss of vision and developing phthisis [8, 9]. Surgical drainage of MSCH may produce better anatomical outcomes, particularly when associated with retinal detachments, retinal or vitreous incarceration, persistent flat anterior chamber, or haemorrhage [3, 10]. Unfortunately, the visual function of patients treated with surgical drainage tends to remain low [6, 11]. A 7–14-day delay is typically imposed to allow the clot to liquefy and thus aid surgical drainage [5]. We hypothesise this may be partially responsible for the poor visual recovery because of the irreversible changes occurring in the retina during this time, including retinal necrosis and atrophy [7].

Tissue plasminogen activator (tPA) is a natural serine protease that functions by selectively activating fibrin-bound plasminogen to form plasmin. Plasmin, in turn, degrades fibrin to help liquefy blood clots [12]. Recombinant DNA techniques have been used to produce tPA for medicinal use. The practice of using recombinant tPA (r-tPA, alteplase) to accelerate SCH drainage has not reached widespread use due to unknown risks and side effects. Table 1 summarises cases reports in the literature where recombinant r-tPA has been used as an adjunct to hasten clot liquefaction [13‒18].

Here we present a series of cases where r-tPA has been injected into the suprachoroidal space prior to surgery to assist early drainage of macula-involving MSCH. To the best of our knowledge, this is the first case series treated with a standardised approach early drainage of SCH.

We performed retrospective review of cases with macula-involving MSCH who were treated with r-tPA prior to surgical drainage (Table 2). In each case, SCH was diagnosed based on clinical examination and confirmed by ultrasound. Procedures were performed by two vitreoretinal surgeon (S.S. and E.J.H.) in NNUH and ZNA Middelheim Hospital, respectively.

The position of SCH was carefully documented during ultrasound examination. This was then used as a guide to inject trans-scleral r-tPA into the SCH using a 25- or 27-gauge needle in aseptic setting. 24 h were allowed before starting the first stage of treatment. This was to allow maximum haemolysis through direct contact of r-tPA molecules with the blood clot and minimise the risk of rebleed. Injection of r-tPA was placed in 2 quadrants in the clock hour of the highest point of SCH as marked by ultrasound, 4–5 mm from the limbus, at 2–3 mm deep perpendicular to the sclera.

r-TPA acts by direct contact; therefore, it was divided into two injection sites 50/0.2 mL at each site to facilitate diffusion. Induced ocular hypertension was treated with topical antiglaucoma drops and systemic acetazolamide when required.

A further interval (24–48 h) was arranged before the second stage of surgery was allowed to commence. Drainage of SCH was performed via standard scleral incisions for 3 port pars plana vitrectomy (PPV) using 23-gauge MVR blade, while the intraocular infusion was placed in the anterior chamber. Following this, PPV with patient specific additional procedures and internal silicon oil tamponade was performed at the discretion of the operating surgeon.

Written informed consent was obtained from all the patients for publication of the details of their medical case and any accompanying images. Ethics approval was not required by the West of Scotland Ethics Committee as it was a retrospective study. The principles of the Declaration of Helsinki were maintained throughout the audit process.

The CARE Checklist has been completed by the authors for this case report, attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000545290).

A 55-year-old pseudophakic female with a background of aniridia and with previous (prior to trauma) best corrected visual acuity (BCVA) of 6/18 presented following repaired globe rupture from 12 to 3 o’clock and total hyphaema following blunt trauma was referred to us due to macula-involving SCH confirmed on ultrasound demonstrating large macular haemorrhage, choroidal effusions of kissing configuration (Fig. 1). Her BCVA at presentation was hand movements. Injection of r-tPA 100 μg/0.4 mL into the suprachoroidal space was performed 24 h after the initial haemorrhage. Drainage of the SCH was performed 48 h later, combined with PPV, localised retinectomy of incarcerated superior retina, and a continuous 10-0 prolene retention suture in the iris plane for the oil tamponade. Postoperatively, the retina appeared flat with no residual SCH (Fig. 2). 1 year postoperatively, BCVA was counting fingers with and intraocular pressure (IOP) of 14 mm Hg.

A 22-year-old male presented with a large limbus to limbus full-thickness central corneal wound dehiscence after a night of intoxication and vomiting. He had a previous history of surgical repair of a glass injury followed by lensectomy, PPV 4 months earlier. On examination, he had hand movement vision, flat anterior chamber, no red reflex, and MSCH on ultrasound (Fig. 3). Corneal wound closure and injection of r-tPA 100 μg/0.4 mL into the suprachoroidal space was performed 24 h after the injury by surgeon 1 (S.S.). Drainage of the SCH was carried out 48 h later and was combined with redo-PPV, silicon oil tamponade, and retinopexy. Postoperatively, the retina was flat under oil (Fig. 4). One year postoperatively, BCVA was 6/36, IOP was 17 mm Hg, and the SCH was drained completely.

An 89-year-old male was referred from private sector because of a MSCH diagnosed 1 day after a XEN gel stent (Allergan, Dublin, CA) implantation for chronic open angle glaucoma. He presented with a flat anterior chamber, IOP of 45 mm Hg, and a MSCH. Visual acuity was light perception. Ultrasound confirmed a MSCH involving the macula. Injection of r-tPA 100 μg/0.1 mL into the suprachoroidal space was performed 24 h after the initial haemorrhage. Six hours later, a trans-scleral drainage via pars plana was performed by surgeon 2 (E.F.). Glaucoma drainage site was not interfered with in hope that blood residues will control the causative over drainage. At day 3 (24 h later), the patient presented again with a flat anterior chamber, a recurrent subtotal MSCH, with IOP of 45 mm Hg. The ultrasound showed a central opening in the vitreous space with remaining SCH in the 4 quadrants, mainly in the temporal area (Fig. 5). On the same day, 100 μg/0.4 mL of r-tPA was injected again into two temporal quadrants of SCH. After 6 h, the SCH was surgically drained. Intravitreal perfluorodecalin was used as a temporary tamponade and was removed 14 days later and replaced with an air tamponade. The retina and choroid remained flat postoperatively, and BCVA recovered to 6/15 at 1 year post operation.

The results from our case series demonstrated that 100 μg/0.4 mL of r-tPA (alteplase) can be safely administered as early as 24 h following initial injury and early drainage with assisted PPV can be performed within 3 days of initial injury. This led to improved final visual acuity in all our patients. We believe this is the earliest drainage of MSCH with use of 2-step approach with use of r-tPA to our knowledge. We will discuss the learning points alongside a literature review below.

The most effective timing for surgical drainage of SCH has not yet been confirmed. In the acute setting, surgical drainage has been shown to extend the haemorrhage into the vitreous cavity in rabbit models [19]. Subacutely, after clot formation, drainage is also likely to be ineffective and potentially unsafe if the suprachoroidal space is probed to remove residual clots [11]. Previous studies have advocated waiting for clot lysis and have consistently reported a mean time of SCH drainage between 11 and 14 days [6, 9, 11]. This is supported by ultrasound evaluation, revealing natural SCH clot lysis over 6–25 days [11]. However, in some cases, atrophic changes have developed in the choroid and retina while awaiting clot lysis [11]. Throughout our case series, we aimed to maximise visual function by early intervention and were able to initiate surgical drainage within 3 days of onset in two cases and as early as 2 days in 1 case. Early drainage may offer a better prognosis [7] and r-tPA may offer a solution to achieving this.

In case 3, drainage of the MSCH was performed 30 h after the onset with apparent anatomical success. Unfortunately, the SCH recurred on the following day. This may have happened due to over drainage by glaucoma devise, leading to flattening of anterior chamber and hypotony. However, one must also consider r-tPA promoting lysis of the thrombosed vessel. It is likely that the use of internal tamponade after draining SCH is protective from recurrence. We propose that delaying the r-tPA-assisted drainage for 6–48 h may allow more complete haemolysis and decrease the chances of recurrence of the SCH while still allowing improved visual outcome.

In the systemic treatment of stroke, the r-tPA (alteplase) is administered as a thrombolytic at a dose of 900 μg/kg (max. per dose 90 mg) [20]. When used intraocularly, dosages are markedly reduced. Treatment for submacular haemorrhage using intravitreal r-tPA was first described using a dose of 12.5 μg[31]. Previous reports (Table 1) of treating SCH with r-tPA have utilised doses between 25 μg and 100 μg.

Doses above 50 μg of intravitreal r-tPA have been identified as inducing retinal toxicity in rabbit models [21]. This has also been reported to occur in humans following two 50 μg doses, 3 days apart [22]. These investigations both reported that toxicity may be related to the carrier (l-arginine) rather than r-tPA itself. The toxic reactions included pigmentary changes, retinal detachments, electroretinogram changes, and poor visual recovery. Intravitreal doses below 25 μg have not been shown to cause harm [21]. However, we hypothesise the suprachoroidal administration of r-tPA may prevent direct retinal toxicity due to presence of choroid as mechanical barrier and the blood retinal barrier, and so higher doses than 25 μg may be used. To our knowledge, there is no published literature on this.

If recognised as a useful treatment, the possibility of retinal toxicity needs to be balanced with the dose/concentration of r-tPA required to enable safe early clot lysis. Our observations suggest that a suprachoroidal injection of 100 μg r-tPA is safe.

r-tPA has a short systemic half-life of 5–10 min [23]. Although animal models have found this to be 4–6 h intravitreally and 10–12 h in the presence of a fibrin clot [24]. In vitro analysis on human plasma demonstrates that fibrinolysis begins to occur around 20 min following fibrinogen/plasmin contact [25]. However, both plasmin generation and fibrinolysis are thought to be a dynamic process in vivo and clot structure is an important influence [26].

Animal models have shown that after administration of just 25 μg of tPA allowed SCH to be aspirated after just 32 min. The time was reduced to 24 min if 50 μg of tPA was administered [27]. However, the tPA was administered as depot on a sponge sewn underneath a scleral flap. Although we cannot directly compare this model, it is apparent that there is a dose dependant response of clot lysis within the suprachoroidal space. In this case series, surgical drainage of the SCH was performed between 6 and 48 h following administration of r-tPA. Complete drainage was only complete in cases 1 and 2. These two cases had the longer interval between r-tPA administration and surgical drainage (48 h). Conceivably, this effect may have occurred through the short-acting bolus administration of r-tPA initiating sufficient liquefaction for the core and distal areas of the clot to be broken up by endogenous procoagulant activity. We feel that a balanced approach would be to recommend surgical drainage in 24–48 h after the r-tPA administration.

In this series, complications occurred in only case 3. Recurrence of SCH in case 3 most likely to have happened due to over drainage by glaucoma devise leading to hypotony and may have benefitted from device removal.

Surgical drainage of macula-involving MSCH has been found to produce favourable anatomical outcomes [5‒7] which was confirmed in our case series. However, prognosis for functional recovery following the drainage of MSCH usually remains guarded [6]. Scott et al. [28] managed 51 patients with MSCH and found disappointing visual outcomes. 27.5% of patients have a final vision of NPL and 29.4% retained pre-haemorrhage acuity or vision greater than 20/200.

Qureshi et al. [29] from Manchester University Hospital also looked retrospectively over a 10-year period (2008–2018) at 20 patients with SCH who underwent external surgical drainage alone versus combined with PPV according to surgeon preference and clinical indication such as presence of retinal detachment or dense vitreous haemorrhage. The average time for drainage for all cases was 11 ± 6 days. Overall the mean BCVA marginally improved from 20/3,319 to 20/526. It was suggested that drainage with or without PPV was valuable in the management of these cases.

More recently, Chen et al. [30] managed 12 eyes with SCH with external drainage and PPV between 2010 and 2020. In this study, the average time for drainage was 17 days (range: 7–30 days). Once again, the mean visual outcome was poor improving from light perception to counting fingers. This further illustrates the need for early drainage to improve visual potential.

Despite the generally poor prognosis of macula-involving MSCH, 2 out of 3 patients in our cohort showed an improvement upon their presenting visual acuity following expedited drainage of the MSCH. We suggest this is due to early drainage expedited by use of r-tPA.

Our cohort is small with various aetiologies causing MSCH. This makes the data difficult to analyse. However, we standardised the two-step operation with the timing of administration of r-tPA, as well as surgical intervention combining surgical drainage with PPV and produced favourable outcomes in 2 out of 3 cases.

Our results have shown that two-step r-tPA-assisted early drainage of macula-involving MSCH has the potential to restore some clinically useful visual function in cases with otherwise poor prognosis. Our observations lead us to believe that an appropriate dose of r-tPA would be 100 μg administered 24 h of the MSCH developing, prior to surgical drainage 24–48 h later. Our promising case series lead us to believe that further large multicentre studies are essential to establish confirmation of safety and efficacy before it can be recommended as a standard procedure for treatment of MSCH.

Written informed consent was obtained from the patient for publication of the details of their medical case and any accompanying images. Ethics approval was not required by the West of Scotland Ethics Committee as it was a retrospective study. The principles of the Declaration of Helsinki were maintained throughout the audit process.

The authors declare that there were no conflicts of interest.

The authors declare that there was no funding for this retrospective case series.

M.K. (corresponding author), M.S., M.V., J.A.D., and E.F. were involved in the writing of the study. S.S. was involved in the writing of the study, revised the article, and provided senior support.

All data generated or analysed during this study are included in this article and its online supplementary material files. Further enquiries can be directed to the corresponding author.