Article Text
PDF
Abstract
Background/aims To investigate the clinical effectiveness of adjunctive triamcinolone acetonide (TA) given at the time of vitreoretinal surgery following open globe trauma (OGT).
Methods A phase 3, multicentre, double-masked randomised controlled trial of patients undergoing vitrectomy following OGT comparing adjunctive TA (intravitreal and subtenons) against standard care (2014–2020). The primary outcome was the proportion of patients with at least 10 Early Treatment Diabetic Retinopathy Study (ETDRS) letter improvement in corrected visual acuity (VA) at 6 months. Secondary outcomes included: change in ETDRS, retinal detachment (RD) secondary to PVR, retinal reattachment, macular reattachment, tractional RD, number of operations, hypotony, elevated intraocular pressure and quality of life.
Results 280 patients were randomised over 75 months, of which 259 completed the study. 46.9% (n=61/130) of patients in the treatment group had a 10-letter improvement in VA compared with 43.4% (n=56/129) of the control group (difference 3.5% (95% CI −8.6% to 15.6%), OR=1.03 (95% CI 0.61 to 1.75), p=0.908)). Secondary outcome measures also failed to show any treatment benefit. For two of the secondary outcome measures, stable complete retinal and macular reattachment, outcomes were worse in the treatment group compared with controls, respectively, 51.6% (n=65/126) vs 64.2% (n=79/123), OR=0.59 (95% CI 0.36 to 0.99), and 54.0% (n=68/126) vs 66.7% (n=82/123), OR=0.59 (95% CI 0.35 to 0.98), for TA vs control.
Conclusion The use of combined intraocular and sub-Tenons capsule TA is not recommended as an adjunct to vitrectomy surgery following OGT.
Trial registration number NCT02873026.
- Retina
- Trauma
- Vitreous
Data availability statement
Data are available upon reasonable request. Data requests would need to be assessed by the Moorfields Eye Hospital Research Managers to ensure any data shared complies with Data Protection laws.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THE TOPIC
Proliferative vitreoretinopathy (PVR) is the most common cause of recurrent retinal detachment in eyes following open globe trauma.
WHAT THIS STUDY ADDS
Adjuvant triamcinolone did not improve visual outcomes at the time of vitrectomy following open globe trauma.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The study adds to the evidence surrounding the treatment of PVR but was limited perhaps by its broad inclusion criteria. Future PVR trial designs will need to have a clear focus on disease categorisation and the timing of intervention.
Introduction
Ocular trauma is a leading cause of visual loss worldwide and is the most common cause of unilateral visual loss, with significant socioeconomic implications.1 In eyes that suffer open globe trauma (OGT), retinal detachment is a frequent complication and often requires multiple surgical procedures.2 3 Proliferative vitreoretinopathy (PVR) is the most common cause of recurrent retinal detachment in these eyes and is associated with a worse visual outcome.4
Corticosteroid treatment has the potential to reduce the inflammatory and proliferative components of PVR; laboratory studies have suggested that corticosteroids can downregulate the pathobiological processes of PVR development.5–7 A previous pilot randomised controlled study suggested that the adjuvant use of corticosteroid at the time of vitrectomy in eyes that had suffered OGT was associated with an improvement in visual outcomes.8 A randomised controlled trial (RCT) on slow-release dexamethasone in vitreoretinal surgery also suggested a positive effect from steroid adjuncts in PVR cases,9 although other studies using oral steroids have not shown an effect.10
The aim of this study was to test the hypothesis that adjunctive triamcinolone acetonide (TA), given at the time of surgery, can improve the outcome of vitreoretinal surgery following open globe trauma. The study analysed its effect on VA and the incidence of re-detachment.
Methods
A phase 3 multicentre double-masked RCT was carried out between November 2014 and September 2020. The trial was registered on the Clinicaltrials.gov database on 19 August 2016. The study complied at all times with the tenets of the Declaration of Helsinki. Patients provided written informed consent before enrolment. An independent Data Monitoring Committee and Trial Steering Committee provided oversight throughout the trial. The full study protocol is detailed elsewhere.11 During the trial, the protocol was amended to change the primary outcome from a dual outcome to a single outcome (proportion of patients with a greater than 10 letter Early Treatment Diabetic Retinopathy Study (ETDRS) letter score improvement). The co-primary outcome, ETDRS letter score at 6 months, was converted to a principle secondary outcome. This was because it was felt that ETDRS letter score would have an unusual distribution, therefore requiring a complex statistical model which may be more difficult to communicate and that a binary outcome was more clinically meaningful to clinicians and patients.
Participants
Eligible patients were those over 18 years old who had suffered full-thickness OGT and were undergoing vitrectomy, able to give written consent, willing to accept randomisation and able to attend 6-month follow-up. Exclusion criteria were (1) age under 18 years old; (2) pre-existing uncontrolled uveitis; (3) previous diagnosis of steroid-induced glaucoma; (4) pregnant or breast-feeding women; (5) allergy or previous reaction to TA; (6) inability to attend follow-up; (7) inability to give written consent; (8) current or planned systemic corticosteroid use of a dose above physiological levels (>10 mg prednisolone). The indication for vitrectomy following OGT was at the discretion of the operating surgeon.
Recruitment
The multicentre study was planned across 20 UK sites. Recruitment was monitored and seven more recruiting sites were added during the trial.
Intervention
Both groups received standard surgical treatment and routine preoperative and postoperative treatment and care. Standard care involved vitrectomy, treatment of retinal detachment if present and the surgeon’s choice of intraocular tamponade agent. The Adjunct Group received additional postoperative steroid combination (triamcinolone acetonide, Kenalog, E.R. Squibb & Sons, New York, USA) consisting of 4 mg/0.1 mL into the vitreous cavity and 40 mg/1 mL subtenons. The control group received standard care only.
Randomisation, masking and assessments
Eligible participants were randomised (1:1) using a telephone service to the Emergency Scientific Medical Services global service hosted at the King’s College Clinical Trials Unit. The randomisation assignments were created using permuted blocks of varying sizes with stratification by trial centre. Randomisation and treatment allocation were performed intraoperatively once the operating surgeon had confirmed that the retina was attached. Operating surgeons were masked until the end of surgery; participants and study investigators were masked to treatment allocation throughout. The trial statistician remained subgroup masked throughout the trial and analysis.
Baseline assessments were performed within 14 days prior to the study vitrectomy. Participants’ study visits were at 3 and 6 months postoperatively. Adverse events were reported to the sponsor as per the study protocol. Elevated intraocular pressure was defined as >25 mm Hg.
Outcomes
The primary outcome was the proportion of patients with an improvement from baseline to 6 months of at least 10 on the corrected VA (ETDRS letter score at a starting distance of 4 m) in the study eye. The secondary outcomes were (1) change in corrected VA score (ETDRS letter score) at 6 months after initial study surgery; (2) retinal detachment with PVR at any timepoint within 6 months; (3) stable complete retinal reattachment (without internal tamponade present) at 6 months; (4) stable macular retinal reattachment (without internal tamponade present) at 6 months; (5) tractional retinal detachment at any timepoint within 6 months; (6) the number of operations to achieve stable retinal reattachment (either complete or macula) at 6 months; (7) hypotony (<6 mm Hg) at any timepoint within 6 months; (8) raised intraocular pressure (>25 mm Hg) at any timepoint within 6 months; (9) macular pucker by 3 and 6 months and/or require macular pucker surgery at any timepoint within 6 months; (10) quality of life measured using the VFQ25, CSRI and the EQ-5D-5L questionnaire.
Sample size
The target sample size was 300 (150 per arm) over a 3-and-a-half-year recruitment period. This was based on an assumed proportion of individuals with clinically meaningful improvement in VA (>10 letters) of 55% in the standard care arm and a 19% increase in the adjunct group to 75%, with approximately 7% loss to follow-up, at least 90% power and two-sided 5% type 1 error. Following slower than anticipated recruitment, the recruitment period was extended to 75 months. Over the full recruitment period, 280 eligible patients were recruited and are included within this analysis. Based on the original sample size parameters outlined above, it was established that the trial would still be adequately powered: a sample size of 280, assuming loss to follow-up of 7%, that is, 260 completers at 6 months provided 89.7% power to detect a 19% increase (55%–74%) in meaningful improvement in VA (>10 letters).
Statistical analysis
Analysis was conducted subgroup masked (ie, group A vs B) following the ASCOT statistical analysis plan.12 The main analysis was based on the intention-to-treat principle and included all eligible participants with follow-up in their randomised group to estimate the effect of the treatment policy (see online supplemental file). The primary analysis model consisted of a mixed logistic model with change in VA (<10, ≥10 change in 6-month ETDRS score) as the outcome and treatment arm and baseline value of the ETDRS as covariates. Treatment centre was included as a random intercept. The estimated treatment effect is reported as a subject-specific OR (conditional on centre and baseline ETDRS) with 95% CI and corresponding p value. Planned sensitivity analyses for the primary outcome assessed the impact of missing outcome data and data collected out of window outcome. Linear (Gaussian) mixed regression models were used to analyse continuous secondary outcomes. Binary secondary outcomes were analysed using mixed logistic regression models. For count outcomes, a mixed effect negative binomial model was fitted, which allowed for overdispersion. Pre-planned subgroup analysis investigated whether the treatment effect on the primary outcome differed by retinal detachment, foveal involvement, presence of PVR or retinal incarceration and lens status. For additional details, see online supplemental file 1.
Supplemental material
Estimates are presented with 95% CIs and two-sided p values. Statistical analysis was performed using Stata/IC V.15.2 (StataCorp) and a two-sided p value<0.05 was considered statistically significant.
Results
Site recruitment is shown in online supplemental eTable 1. Seven hundred ninety-two patients were screened, 317 were assessed for eligibility and 280 patients were randomised over 75 months. The CONSORT diagram is shown in figure 1. Four patients were withdrawn on the randomisation date from the adjunct group (two patients taking steroids, two were ineligible at the end of surgery). The remaining withdrawals were due to loss to follow-up (seven in each group) or no longer wanting to continue (one from control and two from adjunct group).
CONSORT diagram. CONSORT diagram shown for the Ascot study. †Four participants who were randomised in error as ineligible and immediately withdrawn on date of randomisation. Numbers withdrawn are cumulative. ITT, intention on treat.
Baseline characteristics
Baseline demographics and ocular characteristics are shown in table 1.
Baseline demographics
Participants were predominantly young white males, with poor preoperative vision (ETDRS zero or worse: 72% control and 78% adjunct) and with just under one-third suffering a Zone 3 injury (21% control and 31% adjunct). Of note, 69% of the control and 77% of the adjunct group had already undergone a primary repair, 48% of the control and 54% of the adjunct group had a pre-existing retinal detachment and PVR was present in 21%–27%, respectively. The time of vitrectomy was a median of 28 (control) and 26 (adjunct) days following the original injury. Operation and follow-up details are shown in table 2.
Operation and follow-up details
The groups were similar in terms of surgical gauge, need for lensectomy and silicone oil tamponade. The adjunct group had a higher rate of retinectomy and the follow-up data suggests that there was a higher rate of oil tamponade at both the 3-month and 6-month follow-up visit in the adjunct group.
Primary outcome measure
The proportion of participants with at least a 10-letter improvement in VA over the 6-month post-vitrectomy was 43.4% (n=56/129) in the control group and 46.9% (n=61/130) in the adjunct group (unadjusted difference 3.5% (95% CI −8.6% to 15.6%)) (table 3). The baseline-adjusted OR for clinically meaningful change in VA for the adjunct relative to the control group was 1.03 (95% CI 0.61 to 1.75, p=0.908) indicating no significant difference between the treatment groups. Primary estimand attributes are shown in online supplemental eTable 2. Sensitivity analyses were performed and did not influence the outcome (see online supplemental material 1). Online supplemental eFigure 1 shows the subgroup analyses which suggested that if the fovea was detached, the adjunct increased the odds of an improvement in BCVA (OR 3.46 (95CI 1.16 to 10.36)). However, the 95% CI for this effect overlaps the treatment effect for those with the fovea on, meaning no difference in the treatment effect by fovea status cannot be ruled out. There was no difference in outcome in any of the other subgroups including retinal status and fovea status (online supplemental eFigure 1). Sensitivity analyses are shown in online supplemental eTables 3–4 & eFigure 2.
Outcome data
Secondary outcomes analyses
Secondary outcomes are shown in table 3.
For the principle secondary outcome (change in ETDRS VA at 6 months), there was no significant difference between the groups (−2.65, 95% CI −9.22 to 3.92, p value=0.430). There was also no significant difference between the groups for PVR re-detachment, tractional retinal detachment, hypotony, number of operations and quality of life at 6 months. There was a significantly lower rate of stable complete retinal and macular retinal reattachment in the adjunct group compared with the control group (n=65/126, 51.6% adjust vs n=79/123, 64.2% control had complete reattachment and n=68/126, 54.0% adjunct vs n=82/123, 66.7% control had macula reattachment). Elevated IOP was more common in the adjunct group and there was a trend for increased macular pucker, also in the adjunct group.
Adverse events
The summary of the adverse events is shown in figure 2. There was a higher rate of elevated IOP events in the adjunct group compared with the control group (58 vs 45 events, respectively) and a similar rate of hypotony. There were two case of endophthalmitis in the adjunct group. There were six cases of uveitis in the in the adjunct group compared with two in the control group.
Adverse events. In the control group, other adverse events were: conjunctivitis, graft opacification, swollen disc, vitreous haemorrhage, pupillary membrane, macular oedema, diplopia, total funnel inoperable retinal detachment, focal keratitis, corneal graft failure, corneal oedema, irritation and cystoid macula oedema. In adjunct group, other adverse events were: corneal abrasion, oil in anterior chamber, epiretinal membrane, epiretinal membrane, central macular subretinal bleed, cystoid macular oedema, proliferative vitreo-retinopathy, cataract formation, macular hole, irritation (x2), macular oedema, photophobia (x2), foreign body sensation (x2), pain (x2), watering eye (x2), keratitis at central cornea, intermittent headaches and loose corneal suture.
Discussion
The ASCOT study is the first large-scale RCT to investigate the use of adjunctive medication to prevent PVR following surgery for penetrating OGT. TA was chosen on the basis of preclinical evidence of its efficacy13 and promising results from a pilot trial.8 The primary outcome (improvement in VA) and principal secondary outcome (change in VA) did not demonstrate a significant treatment benefit for TA. The secondary outcome measures failed to show any treatment benefit. The use of combined intraocular and sub-Tenons capsule TA is therefore not recommended as an adjunct to vitrectomy surgery following OGT.
It is notable that stable complete retinal reattachment (51.6% (65/126) TA vs 64.2% (79/123) control group) and stable macular retinal reattachment (54.0% (68/126) TA group vs 66.7% (82/123) control group) at 6 months were significantly worse for the treatment group compared with controls. These two outcomes are related and are clearly of clinical importance. Over a range of baseline parameters, the treatment group appeared to have more severe pathology on presentation. The treatment group, by chance, had a higher level of previous primary repair (77% vs 69%), previous eye surgery (57% vs 49%), zone 3 injuries (31 vs 21%), vitreous haemorrhage (69% vs 63%), retinal incarceration (27% vs 18%), pre-existing retinal detachment (tractional and rhegmatogenous) (54% vs 48%) and pre-existing PVR (27% vs 21%). Although none of these parameters demonstrate a marked difference, taken together they may account for a difference in outcome between the groups. Nevertheless, a negative effect of TA as an adjunct to vitrectomy surgery for OGT cannot be discounted although the pathobiological mechanism for this is unclear. These findings support the conclusion that TA should not be routinely recommended as an adjunct in OGT cases.
Experimental studies have demonstrated that TA has the potential to downregulate the retinal response to injury and reduce the incidence of PVR.14 The potential for TA to produce a beneficial clinical effect on PVR is supported by pilot and small scale clinical studies suggesting a reduction in the inflammatory response and PVR in retinal detachment and trauma cases.8 15–17 The reasons for the failure of TA to produce a treatment effect therefore need to be considered. It is possible that in OGT cases where there is extensive blood-ocular barrier breakdown and a markedly upregulated drive towards PVR, the pharmacological effect of TA (at the dosage used in the study) is insufficient to influence the PVR process. In this context, it is notable that a recent uncontrolled study using a stronger antiproliferative agent, mitomycin C at the time of vitrectomy in severe intraocular foreign body cases appeared to reduce the incidence of PVR.18 Timing of drug delivery will likely have also played a role. Most patients in the ASCOT study had already undergone primary repair of the penetrating injury (69% and 77% of control and adjunct patients) and time of vitrectomy was a median of 28 (control) and 26 (adjunct) days following the primary repair and exceeded 119 days in a quarter of patients (table 1). The use of an adjunctive agent in this subset of patients so long after the original trauma will likely have been too late to alter their outcome or implies that their vitrectomy may not have been directly related their original OGT. Delivery of a therapeutic adjunct at the time of injury, potentially combined with sustained delivery, may produce a greater effect in modifying the PVR process in OGT.
Case selection is an additional factor which may have influenced the results. ASCOT recruited a broad spectrum of open globe trauma cases. Overall, 40% of cases had a globe rupture, 37% penetrating injury, 19% had intraocular foreign bodies and 4% a perforating injury. Cardillo and co-workers4 documented that these varied injury types have differing incidences of PVR: perforating injuries 43%, globe rupture 21%, penetrating injuries 15% and intraocular foreign bodies 11%. It is likely that the injury subtypes will have differing responses to therapeutic agents and in future studies more focused case selection, potentially limited to only one injury subtype, could result in a positive therapeutic response. Subgroup analysis suggested a possible benefit in eyes in which there was a fovea-involving retinal detachment, although it should be noted that the data was also consistent with no difference and it is well established that subgroup analysis can often be challenging.19 20 Even so, this subgroup may be of interest in future work.
The results of surgery for PVR have remained unsatisfactory with often poor visual outcomes and a need for multiple procedures. This has led to both preclinical and clinical studies to identify adjunctive agents to modify the disease process and improve surgical results.21 22 To date, no adjunctive agents have gained widespread acceptance and PVR remains a surgical disease.14 Intraocular daunomycin23 and the combination of 5 fluorouracil (5FU) and low-molecular-weight heparin (LMWH)24–26 have been studied in a series of RCTs. Although these produced promising initial findings—daunomycin reduced re-operations and the 5FU/LMWH combination reduced PVR in high-risk cases—these studies have not resulted in the drugs having widespread use. Likewise, a study of slow-release dexamethasone in established PVR failed to improve anatomical results although macular oedema was reduced and there was a trend to improved VA.9 A previous study on TA in non-traumatic PVR also failed to show a benefit.17 The reasons for the failure of these treatments to improve outcomes for PVR surgery is likely multifactorial and appear to relate to a lack of understanding of the PVR disease process and inadequate case selection.
Limitations of the study were its broad inclusion criteria and that it did not specify the indication for vitrectomy. This likely contributed to the observed length of time overall between OGT and vitrectomy in the study (median 27 days (IQR 12–119)). Future PVR trial designs will therefore need to have a clear focus on disease categorisation and the timing of intervention. The use of subconjunctival steroid at the time of the initial primary repair was not recorded as part of the study design, which may have influenced results.
The ASCOT study was designed to investigate TA as an adjunct to vitreoretinal surgery following OGT. It provided a clear answer, in that there was no benefit from TA in a broad mix of OGT cases and adds to the evidence surrounding the treatment of PVR.
Data availability statement
Data are available upon reasonable request. Data requests would need to be assessed by the Moorfields Eye Hospital Research Managers to ensure any data shared complies with Data Protection laws.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by NRES Committee London Central (14/LO/1428) on 14 September 2014. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @edwardcasswell, @Suzie_cro
Collaborators See Contributor Statement regarding ASCOT Investigator Study Group.
Contributors Design of work: SC, VRC, PJB, RTE, VE, CB, JK, CM, ER, DC; acquisition, analysis or interpretation of data for the work: EJC, SC, VRC, PJB, TMZ, RTE, VE, BA, SMS, CB, JK, CM, ER, DC. All authors were involved in the drafting and review of the manuscript as well as final approval. DC is the guarantor of the study and responsible for the overall content. The ASCOT Investigator Study Group served as non-author contributors/collaborators, who collected data and cared for study patients, namely: Mr Arijit Mitra (Birmingham and Midland Eye Centre (BMEC)), Prof Tim Jackson (Kings College Hospital London), Mr Luke Membrey (Maidstone and Tunbridge Wells NHS Trust), Mr Jonathan Smith (Sunderland Eye Infirmary), Ms Arabella Poulson (Addenbrookes Hospital, Cambridge), Prof Ian Pearce (St Pauls Eye Unit, Liverpool), Mr Felipe Dhawahir-Scala (Manchester Royal Eye Hospital), Mr Yash Ramkissoon (Royal Hallamshire Hospital, Sheffield), Mr Alistair Laidlaw (St Thomas’ Hospital London), Ms Daniela Vaideanu-Collins (South Tees Hospital NHS Trust), Mr Mark Costen (Hull University Teaching Hospitals), Ms Cordelia McKechnie (Whipps Cross Hospital London), Mr Richard Haynes (University Hospitals Bristol), Dr Jas Singh (Princess Alexandra Eye Pavilion, Edinburgh), Dr David Yorston (Tennent Institute of Ophthalmology, Glasgow), Ms Rahila Zakir (Western Eye Hospital London), Mr Fung Yang (Queen Alexandra Hospital Portsmouth), Prof Robert MacLaren (Oxford Eye Hospital), Mr Fred Frimpong-Ansah (Royal Eye Infirmary Plymouth), Mr Ravikiran Gandhewa (University Hospital, Derby), Mr Aman Chandra (Mid and South Essex NHS Foundation Trust), Mr Kam Balaggan (Royal Wolverhampton NHS Trust), Ms Roxane Hillier (Newcastle Upon Tyne NHS Foundation Trust), Mr David Schultz (East Kent Hospital NHS Foundation Trust), Mr Mandeep Bindra (Stoke Mandeville Hospital).
Funding This study was supported by the United Kingdom Clinical Research Collaboration-registered King's Clinical Trials Unit at King's Health Partners, which is in part funded by the NIHR Biomedical Research Centre for Mental Health at South London and Maudsley NHS Foundation Trust and King's College London and the NIHR Evaluation, Trials and Studies Coordinating Centre. EJC was supported by the Royal College of Surgeons in Edinburgh and the Special Trustees of Moorfields Eye Hospital. CB is in part funded/supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at the Royal Marsden NHS Foundation Trust and the Institute of Cancer Research, London. DC is supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology. These funding organisations had no role in the design or conduct of the study; the collection, management, analysis, and interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.
Disclaimer The views expressed in this publication are those of the author(s) and not necessarily those of the RCSEd, Special Trustees, NHS, NIHR or Department of Health.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Linked Articles
- Highlights from this issue