This pilot study suggests that intravitreal triamcinolone can be given safely and easily at the time of phacoemulsification surgery.
We took care to inject the triamcinolone inferotemporally away from the visual axis and anterior chamber to avoid visually troublesome floaters and transit of triamcinolone into the anterior chamber. The retinal view was unimpaired post-operatively allowing accurate retinal assessment and further laser treatment if needed. Similar to other studies [32, 33] we found only 22% of eyes developed increased intraocular pressure and all were treated successfully with topical ocular hypotensive agents with spontaneous improvement with time.
Intravitreal triamcinolone has been shown to lead to an improvement in macular oedema and visual improvement in diabetic patients not undergoing cataract surgery [24–26] but has not; to our knowledge; been previously studied in a series such as this.
Combining cataract surgery with triamcinolone rather than giving triamcinolone before surgery as a separate procedure avoided the potential for progression of lens opacities associated with intraocular steroids [34, 35] which could have further interfered with retinopathy assessment. We had no cases of endophthalmitis, which can occur with triamcinolone injections [36, 37]. Combining the two procedures reduces the patient's potential risk of endophthalmitis from two separate intraocular episodes to one, whilst at the same time offering improved patient convenience. The technique was simple adding very little time to the procedure and in this series there was no significant ocular morbidity associated with the triamcinolone. We choose to combine the procedure with our standard clear corneal temporal incision phacoemulsification technique. There is debate regarding the possibility of an increased risk of endophthalmitis with temporal clear corneal incisions . However this has not been our experience. We have had no increase in our rate of endophthalmitis over the last five years during which a clear corneal temporal approach has been adopted by all surgeons at our unit. Incidence of endophthalmitis in our unit for 1999 was 0.11% and for 2004 was 0.09%, based on approximately 6000 cases /year . It is possible that other factors such as wound construction and lid draping and preparation are more important than incision position itself . All the cases in this series were done in theatre with full asepsis, topical povidone pre-operatively, careful lid draping and carefully constructed wounds, which were watertight at the close of surgery. These are important factors in avoiding infective complications.
We had no cases of posterior capsule rupture in this series. Triamcinolone has been used to help visualise vitreous during posterior capsule rupture and anterior vitrectomy  and anecdotally in cases of phacoemulsification surgery with posterior capsule rupture to reduce the incidence of post operative cystoid macular oedema and postoperative inflammation . Potentially therefore intravitreal triamcinolone administration could be considered even if posterior capsule rupture was to occur although we have no experience of this.
The natural history of patients with foveal oedema at the time of cataract surgery is recognised as being poor and the patients in the study had a number of other features associated with a particularly poor prognosis after cataract surgery- increasing age, female sex, poor glycaemic control with high Haemoglobin A 1C (%) at the time of surgery and moderate to severe background retinopathy changes, have all been associated with a poor prognosis in other studies [12, 13, 16, 18]. All the patients in group A had chronic macular oedema prior to cataract surgery, which had been unresponsive to treatment. Indeed patients such as these with chronic unresponsive macular oedema, particularly if there is only a moderate degree of cataract, are often declined surgery on the basis that the maculopathy would limit the underlying visual acuity, which can also deteriorate with surgery. Despite this approximately 50% of these patients achieved 6/12 vision.
Patients who present with dense cataracts and significant retinopathy, especially maculopathy, which is untreatable pre-operatively because of the lens opacities, pose a difficult clinical scenario. Laser can be performed post-operatively but this can be difficult, for reasons previously stated and surgically induced inflammation. There were three patients in the study in this group – group B. Two of these patients had very severe non proliferative retinopathy, in addition to maculopathy, and were treated with intra-operative PRP which can also exacerbate macular oedema. Despite these difficulties and risk factors for maculopathy exacerbation all three patients had complete macular oedema resolution at the two week examination without any macular laser having been applied at that stage.
Overall fifteen (83%) of the patients had complete resolution of their macular oedema at two weeks follow up. Recurrence of macular oedema occurred in some patients in this study as would be expected from the known short term effect of intravitreal triamcinolone and other studies with intravitreal triamcinolone and diabetic macular oedema. However the triamcinolone clearly prevented the short term exacerbation of macular oedema that can be associated with blood ocular breakdown due to intraocular surgery and PRP [23, 29, 42]. It seems logical to use a drug, albeit with a known short-lived effect, in this way to potentially improve visual outcome until longer lasting alternatives are produced.
Retinopathy progression occurred during follow up in only six patients (none within 5 months of surgery and 4 out of 6 at more than 10 months following surgery) and it may be that triamcinolone has a role in reducing the deterioration of retinopathy that has been reported following cataract surgery especially in patients with more advanced retinopathy. It may also have a role in the inhibition of retinal neovascularisation [43, 44]. Numbers in this uncontrolled pilot study are too limited to draw any definitive conclusions.
Triamcinolone was only injected in those eyes with pre-existing macular oedema at the time of cataract surgery. It was not used in eyes thought to be at risk of developing macular oedema after surgery in those with clinically dry foveae at the time of surgery. The natural history of patients with dry maculae at the time of surgery who develop macular oedema following surgery is relatively good  and it was felt that the risks associated with triamcinolone in that group would outweigh the benefits. At present we are not recommending triamcinolone in that group of patients, although this may merit further investigation.