Author: Claude (AI), from PubMed abstracts · Reviewed by: Dr Mina Abdelmseih · Reviewed: 2026-11-15 · Review due: 2027-11-15

Educational content only. Not personalised medical advice. Clinical decisions remain the responsibility of the treating clinician.

Phacoemulsification: outcomes and techniques

Phacoemulsification is the main surgical procedure used to treat age-related cataract [S3]. This article summarises what systematic reviews and trials say about incision size, pupil dilation, postoperative anti-inflammatory treatment, higher-risk eyes and surgical training [S2] [S3] [S5] [S6] [S10].

Key points - Smaller coaxial incisions (2.2 or 1.8 mm) may reduce surgically induced astigmatism by only about 0.2 D compared with about 3.0 mm, with very low-certainty evidence and similar best-corrected acuity [S3]. - Pseudoexfoliation roughly doubles the odds of posterior capsule rupture or zonular dialysis (pooled OR 2.1363) [S1], yet uncomplicated pseudoexfoliation eyes had similar 10-year decentration and capsulotomy rates in a large Indian trial cohort [S11]. - NSAIDs, alone or added to corticosteroids, may lower the risk of cystoid macular oedema (CME), with low-certainty evidence [S2]; in diabetes, topical steroids plus NSAIDs prevented 75.8% of pseudophakic CME events compared with steroids alone [S4]. - Uncomplicated phacoemulsification is followed by posterior vitreous detachment (PVD) in 33.3% of eyes over a mean 28.3 months [S9]. - Residents' complication risk rises with patient factors (pseudoexfoliation, small pupils, prior vitrectomy, zonular pathology) and with absent supervision [S10].

Evidence last checked: 2026-10-17 (PubMed abstracts).

Red flags

• Pseudoexfoliation, zonular pathology, small pupil, prior vitrectomy: recognised risk factors for complications [S1] [S10].
• Diabetes: endothelial damage is greater and corneal recovery is delayed after phacoemulsification [S8]; diabetic eyes also need CME prophylaxis planning [S4].
• New floaters or flashes after surgery, especially without a pre-existing PVD: phacoemulsification accelerates PVD, and PVD carries a risk of retinal tears and detachment [S9].
• Pain, redness and reduced vision after surgery: possible endophthalmitis; see Refer when.

Recognise it: patient selection and risk

• Patient-related risk factors for complications in resident-performed surgery include older age, hypertension, prior vitrectomy, zonular pathology, pseudoexfoliation, poor preoperative acuity, small pupils and selected cataract types [S10].
• Surgeon-related risk factors include postgraduate year; procedure factors include long phacoemulsification time and high power and torsion [S10].
• In open-angle glaucoma, RCTs show a consistent IOP reduction of 4.1 to 8.5 mmHg after cataract surgery and 0.2 to 1.0 fewer medications [S7].

Confirm it: preoperative assessment

• Evaluate the vitreoretinal interface before surgery, and follow up carefully those without a pre-existing PVD [S9].
• Note diabetes status: diabetic eyes had lower endothelial cell density at all postoperative time points up to 6 months and thicker corneas at 1 and 3 months [S8].
• Check pupil size and signs of pseudoexfoliation, which change the risk profile [S1] [S10].

Manage it: technique and perioperative care

Before acting on this section, check the national and local guidelines, product information and regulatory approvals that apply where you practise. See the medical disclaimer.

Tier A (primary care / no specialist)

• Identify patients with cataract, diabetes or pseudoexfoliation and refer to a surgical service with the history noted [S1] [S8].
• After surgery, escalate pain, worsening vision or new floaters and flashes as in Refer when [S9].

Tier B (general eye clinic)

1. Incision size. A Cochrane review (2017) of 26 RCTs (3120 eyes) found 2.2 mm coaxial incisions gave less astigmatism than about 3.0 mm (MD -0.19 D, very low certainty) and similar BCVA; 1.8 mm gave MD -0.23 D. Smaller incisions were not consistently associated with less astigmatism, and safety outcomes were poorly reported [S3].
2. Pupil dilation. A Cochrane review (2021) of 14 RCTs could not establish the superiority of topical, intracameral or depot mydriatics. Topical mydriatics may give a larger pupil at capsulorhexis (MD 1.06 mm, low certainty); ocular discomfort may be greater with topical than intracameral mydriatics (moderate certainty) [S5].
3. Postoperative inflammation. A Cochrane review (2017) of 48 RCTs found insufficient evidence to establish the equivalence or superiority of NSAIDs with or without steroids versus steroids alone. NSAID alone vs steroid alone: lower CME risk at one month (RR 0.26, low certainty) [S2]. Drug choice and dosing are not covered by these abstracts: follow local formularies.
4. Diabetic patients. A meta-analysis of RCTs found that topical steroids plus NSAIDs prevented 75.8% of pseudophakic CME events vs steroids alone in patients without preoperative macular oedema; depot plus topical steroids were superior but raised IOP more often; anti-VEGF added no benefit [S4].
5. Glaucoma with cataract. Consider a medication reduction in well-controlled glaucoma after phacoemulsification [S7].

Tier C (specialist centre)

• Pseudoexfoliation. A meta-analysis of 22 studies found a pooled OR of 2.1363 for capsule rupture or zonular dialysis [S1]. In APEX (930 pseudoexfoliation eyes, 476 controls, randomised to 1- or 3-piece IOLs and to capsular tension ring or not), 10-year IOL decentration (2.4% vs 1.7%) and Nd:YAG capsulotomy (5.7% vs 5.67%) were similar; capsular phimosis and a capsulorhexis not overlapping the optic predicted both [S11]. APEX excluded phacodonesis, shallow chambers and pupils under 4 mm [S11].
• Training. Assign cases to residents with these risk factors in mind and ensure supervision [S10]. A Cochrane review (2021) of 6 RCTs (151 trainees) suggests virtual reality training may improve performance compared with no supplementary training, but all evidence was very low certainty [S6].

Refer when

Pitfalls

• Overselling micro-incisions: the astigmatism benefit is about 0.2 D and uncertain [S3].
• Assuming one dilation method is clearly best: the trials cannot show it [S5].
• Forgetting CME prophylaxis planning in diabetic patients [S4].
• Not warning patients without a PVD about retinal tear symptoms [S9].
• Giving trainees high-risk cases without supervision [S10].

Evidence notes

• Most comparisons in [S2] [S3] [S5] [S6] are low or very low certainty; the Cochrane authors call for larger, better-reported trials.
• Pseudoexfoliation evidence conflicts on the surface: [S1] pools all pseudoexfoliation eyes, whereas [S11] excluded eyes with zonular weakness and small pupils and used experienced surgeons.
• Two of the VR training studies were industry-funded [S6].
• Not covered by these sources: nucleus disassembly techniques, fluidics settings, ophthalmic viscosurgical device choice, anaesthesia, endophthalmitis prophylaxis and drug doses.

Related articles

• Manual small-incision cataract surgery
• Femtosecond laser-assisted cataract surgery
• Biometry and IOL power calculation

References

See the numbered source list below. Each marker [S#] links to its PubMed and DOI record.