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.

Open-globe injury

Open-globe injury (OGI) is a vision-threatening injury with significant global morbidity [S12]. Surgery is the mainstay of management, and the timing of repair may affect the risk of blinding complications such as endophthalmitis [S8].

Key points - Primary repair within 24 hours of injury was associated with lower odds of endophthalmitis than later repair (low-certainty, retrospective evidence) [S8]. - A normal CT scan does not rule out an open globe or an intraocular foreign body (IOFB); CT cannot replace examination by an ophthalmologist [S10]. - When systemic antibiotic prophylaxis is used, oral antibiotics appear non-inferior to intravenous ones (moderate evidence) [S6]. - Intraocular antibiotics at primary repair reduced endophthalmitis, mainly in eyes with a retained IOFB [S1] [S3]. - Sympathetic ophthalmia after OGI is rare (about 0.19%), and there is no evidence that primary eye removal lowers its risk [S4] [S7].

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

Red flags

• Any suspected open globe after trauma, even with a reassuring scan: CT was only 74% sensitive for OGI [S10].
• A possible retained IOFB. In one RCT, 7 of 25 control eyes with an IOFB developed endophthalmitis, compared with 1 of 142 control eyes without one [S1]. A negative CT does not exclude an IOFB [S10].
• Delay to surgery. Repair later than 24 hours was associated with more endophthalmitis [S8].
• Blast or conflict injuries, often seen where specialist care is scarce [S11].
• No light perception (NLP). NLP predicts a poor prognosis, but outcomes of NLP eyes have improved with exploratory vitreoretinal surgery, so NLP alone should not end the assessment [S5].

Recognise it

• In a meta-analysis of 47 studies (12,515 patients), males made up 83% of paediatric and 69% of adult cohorts [S12].
• Globe rupture was the most common injury in children (23%), and IOFB-related injuries in adults (29%) [S12].
• Zone I injuries predominated (more than 50%), and traumatic cataract was the leading complication [S12].
• Penetrating and IOFB injuries were the most common injury types in the timing meta-analysis [S8].

Confirm it

• Clinical examination by an ophthalmologist is the reference standard. In one systematic review (11 studies, 1126 patients), CT was 74% sensitive and 93% specific for OGI compared with clinical diagnosis [S10]. A second meta-analysis (12 studies, 937 patients) found a pooled sensitivity of 75% and specificity of 83% [S9].
• CT signs suggesting OGI: change in globe contour, scleral deformity or injury, corneal injury, altered or shallow anterior chamber depth, lens abnormality or dislocation, and vitreous haemorrhage [S9] [S10]. Shallow anterior chamber and lens dislocation or destruction had the highest sensitivity; lens dislocation and IOFB had the highest specificity [S9].
• IOFB: CT was 69% sensitive for IOFB against intraoperative findings [S10]. B-scan ultrasound was 86% sensitive for IOFB against CT, but its safety with respect to pressure on the globe extruding intraocular contents was not studied [S10].
• Prognosis: the Ocular Trauma Score (OTS) and Paediatric OTS correlate with final vision, but agreement between predicted and observed categories was only moderate (Cohen's κ 0.48), so use them cautiously when counselling [S12].

Manage it

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)

• A resource-adapted guideline (EXTRACCT, 2026, expert consensus) gives frontline workers recommendations for assessing, classifying and managing major ocular injuries without specialist expertise or equipment, including dosing; consult the full guideline [S11].
• Protect the eye, avoid any pressure on the globe and arrange immediate transfer to an eye surgeon.
• Do not delay transfer to obtain imaging: CT cannot confirm or exclude an OGI on its own [S10].

Tier B (general eye clinic)

• Systemic prophylaxis. In a meta-analysis of 2 RCTs (1555 patients), endophthalmitis occurred in 2.26% with oral and 2.11% with IV (± oral) antibiotics, with no significant difference; high-risk cases were excluded [S6]. In a South African RCT, 3 days of IV cefazolin plus oral ciprofloxacin, or oral ciprofloxacin plus oral cefuroxime, from admission gave endophthalmitis rates of 2.0% and 2.7% (not significant), as reported in [S2]. The oral-only regimen shortened hospital stay and reduced costs [S2]. Doses are not stated in these abstracts; follow local formularies.
• Timing. Aim for primary repair as early as possible: repair within 24 hours was associated with an odds ratio of 0.39 for endophthalmitis versus later repair, with no significant difference in visual outcome [S8].

Tier C (specialist centre)

• Intraocular antibiotics. A meta-analysis of 3 trials found that intravitreal or intracameral antibiotics reduced endophthalmitis (RR 0.19) without improving visual acuity [S3]. In the largest RCT (346 eyes), 40 µg gentamicin and 45 µg clindamycin given after primary repair reduced endophthalmitis from 2.3% to 0.3%, as reported in [S1]. The benefit was concentrated in eyes with an IOFB, and intravitreal injection was superior to intracameral [S1].
• Primary eye removal. Pooled sympathetic ophthalmia rates were 0.05% after primary enucleation or evisceration and 0.15% after primary repair, a non-significant difference; there is no evidence that primary removal reduces the risk [S7].
• NLP eyes. A 2022 consensus guideline gives 16 recommendations on the determination, surgical indication and timing of vitrectomy in NLP eyes after mechanical trauma [S5].

Refer when

Pitfalls

• Treating a negative CT as proof that the globe is intact or that there is no IOFB [S10].
• Pressing on a possibly open globe for B-scan; this safety question was not studied [S10].
• Deferring repair for logistics; delay beyond 24 hours was associated with more endophthalmitis [S8].
• Assuming IV prophylaxis is required; oral regimens performed similarly in trials that excluded high-risk cases [S2] [S6].
• Removing an eye mainly to prevent sympathetic ophthalmia; evidence does not show this lowers the risk [S7].
• Over-relying on trauma scores for an individual patient [S12].

Evidence notes

• The timing evidence is retrospective and non-randomised, with low GRADE certainty and very high heterogeneity (I² = 95%) [S8].
• The antibiotic trials excluded high-risk cases, so their low endophthalmitis rates may not apply to contaminated or IOFB injuries [S2] [S6].
• The two CT meta-analyses differ in specificity (93% vs 83%) [S9] [S10].
• The sympathetic ophthalmia evidence is observational and low certainty [S7]. The intraocular antibiotic meta-analysis included only 3 trials [S3].
• The EXTRACCT guideline and the NLP guideline are expert consensus; their specific recommendations are not given in the abstracts [S5] [S11].
• Not covered by these sources: shield and analgesia protocols, tetanus prophylaxis, antiemetics, specific surgical techniques and doses of systemic antibiotics.

Related articles

• Intraocular foreign body
• Blunt eye injury and traumatic hyphaema
• Post-operative endophthalmitis
• Eyelid and Canalicular Lacerations

References

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