A mortar fragment catches a gunner in the face during a halt. The platoon's Combat Lifesaver finds a cut through the left upper eyelid and blood over the eye, presses a folded gauze pad onto it, and ties a cravat tight around the head to stop the ooze. At the casualty collection point the gunner is talking, his radial pulse is strong, and he has no other bleeding. The medic takes the cravat off. The left pupil is peaked toward a small wound at the edge of the cornea, and that eye can only tell that a hand is moving in front of it.
The cravat is the error. A suspected penetrating eye injury gets a vision check, a rigid shield that puts no pressure on the eye, and one of the 2026 antibiotics, then a fast move toward an eye surgeon.
The 1 May 2026 TCCC Guidelines give penetrating eye trauma three steps in Tactical Field Care [1]. If a penetrating eye injury is noted or suspected:
Only the third step is new. TCCC Change 25-1 replaced moxifloxacin by mouth and ertapenem by injection with cefadroxil (preferred) or cephalexin (alternative) by mouth and ceftriaxone by injection, after a panel review of spectrum, side effects, stability, dosing and cost [2]. The general wound antibiotic section gives ceftriaxone IV, IO or IM for the casualty who cannot take oral drugs [1]. The eye section names IV or IM only [1]. The FY26 Joint En Route Care Guidelines call ceftriaxone 2 g IV or IM the recommended eye antibiotic and list cephalexin 500 mg or cefadroxil 1 g by mouth as the oral options [3]. Our summary of the 2026 TCCC update covers the rest of the antibiotic change.
The same guideline settles a common question about pain control: TBI or eye injury does not preclude ketamine [1]. The dosing paths are in our review of the 2026 analgesia update.
The Joint Trauma System (JTS) Eye Trauma: Initial Care guideline lists the findings of an open globe: a collapsed or badly distorted eye, a full-thickness corneal or scleral laceration, a shallow anterior chamber, a peaked or irregular pupil, prolapse of eye contents, and subconjunctival hemorrhage, especially all the way around the eye [4]. Intraocular foreign body findings can be subtle, so the guideline asks for a high index of suspicion [4]. A small fragment wound near the eye after a blast is enough to suspect one.
TCCC training uses a quick four-level field test instead of a Snellen chart [5]:
Record the best level for each eye on the DD Form 1380 [5]. "Left: hand motion. Right: reads print" gives the next provider a baseline. "Eye injury" does not. Do not force open a swollen eye to test it [5]. The military ophthalmologists who adapted the ABCs mnemonic for non-ophthalmologists put acuity first and ask for the best exam of both eyes, starting with the uninjured eye [6].
Two more don'ts from the JTS guideline: no pressure on the eye, and no ultrasound of the injured eye, because the probe presses on the globe [4].
The JTS guideline gives the shield rules plainly [4]:
TCCC training adds two details: tape at 45-degree angles across the forehead and cheek, and if no rigid shield is available, use an improvised shield or the casualty's tactical eyewear [5]. Do not try to remove an intraocular foreign body [4].
For the gunner, the gauze comes off, nothing replaces it, and the shield sits on the brow and cheekbone. He gets his antibiotic and a time on the card.
The JTS guideline asks for aggressive treatment of nausea and vomiting, nothing to eat or drink, and the head raised 30 degrees when possible [4]. The 2026 TCCC antiemetic is ondansetron 4 mg ODT, IV, IO or IM every 8 hours as needed [1]. In the field, raise the head only when the casualty is not in shock and tactics allow, the same limits TCCC sets for head and torso elevation in TBI [1].
The infection to prevent is post-traumatic endophthalmitis, infection inside the eye. In a civilian series of 558 open globe injuries, the rate was 0.9% under a protocol of 48 hours of IV vancomycin and ceftazidime and repair by a dedicated eye trauma service [7]. An intraocular foreign body was a risk factor (odds ratio 7.52) [7]. At Walter Reed, 79 combat eyes with intraocular foreign bodies had delayed removal (median 21 days) under systemic and topical antibiotics, with no endophthalmitis [8]. The most common systemic drug was levofloxacin, and the most common topical drug was moxifloxacin [8].
Those low rates came under broad-spectrum IV or fluoroquinolone regimens. The Change 25-1 panel recognized that its new choices have a narrower spectrum. It asked for that to be considered in Prolonged Casualty Care guidance, and for surveillance for unanticipated increases in post-traumatic endophthalmitis, open fractures or abdominal injuries [2]. The JTS eye guideline was last revised on 1 June 2021, and the JTS index still listed that edition as current on 30 September 2026 [4, 9]. It still names levofloxacin with vancomycin, or moxifloxacin, as prophylaxis [4]. At the point of injury, the 2026 TCCC choice is the current standard. Document the drug, dose, route and time. That record is the surveillance.
Time to surgery is the second lever. A 2025 meta-analysis of 15 retrospective studies (8,497 eyes) found that repair within 24 hours was associated with lower odds of endophthalmitis than later repair (odds ratio 0.39, 95% CI 0.19 to 0.79) [10]. Final vision did not differ, and the authors rated the certainty of evidence as low [10]. The JTS guideline asks for evacuation to surgical repair within 24 hours when possible [4]. Note the eye injury and the injury time in the MEDEVAC request and the handoff, so the receiving system can route the casualty to an eye surgeon.
Combat ocular trauma was the fourth most common injury among wounded US service members in Iraq and Afghanistan [11]. Blast caused 64% to 84% of it [11]. In the Walter Reed Ocular Trauma Database, 40% of patients with eye trauma also had a TBI [11]. The overall enucleation rate in the large databases was 12% to 17% [11]. The eye casualty is often a head-injured casualty, so serial mental status checks stay on the plan. Our TBI management review covers that lane.
Prevention works. In an Israel Defense Forces trauma registry study of 2013 to 2019, 113 soldiers had combat eye injuries [12]. Among those with eye injuries, open globe injuries occurred in 9.1% of those wearing protective eyewear and 39.5% of those who were not [12]. This was a registry study, and it does not prove cause. It still supports enforcing eyewear on every patrol.
Ceftriaxone 2 g IV or IM, or cefadroxil 1 g by mouth, as soon as possible [1]. These replaced moxifloxacin and ertapenem under TCCC Change 25-1 [2]. Give one as soon as the tactical situation allows and document the time on the casualty card.
A patch or pressure dressing pushes on the eye and may force intraocular contents out through the wound [4]. A rigid shield rests on the bony rim of the orbit and touches nothing else. If no shield is available, TCCC training allows an improvised shield or the casualty's tactical eyewear [5].
Use the TCCC rapid field test: read print, count fingers, see hand motion, tell light from dark [5]. Record the best level for each eye on the DD Form 1380. Do not force a swollen eye open to complete the test [5].
I have worked as an emergency physician for more than 25 years at a Level I academic trauma center, and I serve as State Surgeon for the Mississippi Army National Guard. Eye injuries are among the easiest wounds to make worse with good intentions. This article gets the order right: vision, shield, drug, clock. The argument holds. The cravat in the opening scene comes from a reflex, and the reflex is a good one almost everywhere else. Something bleeds, so we press on it. Every other wound in TCCC rewards that reflex. The eye punishes it. Knowing the rule is not enough. Training has to build a second reflex that fires when the wound is near the eye, and that takes repetitions, not a slide.
That being said, the antibiotic change is the part I would watch. The honest answer is that I do not know yet how the new cephalosporins will perform for a combat eye wound, and neither does anyone else. There is a second risk the article does not cover. An injured eye can get worse after the shield is on. A tense, bulging eye with falling vision after facial trauma points to orbital compartment syndrome. That casualty needs a lateral canthotomy, fast, from someone trained to do it, and a shield will not fix it. It is a different eye from the gunner's. A peaked pupil means a suspected open globe, and that eye gets the shield and no pressure, not the cut. The acuity you wrote down at the point of injury is how the next provider knows the eye is getting worse. Without that first number, a change has nothing to be measured against.
If you are the medic on the ground, open your aid bag tonight and find the rigid shield. If it is not there, fix that before the next movement, and teach your Combat Lifesavers where it lives in your kit. Repeat the acuity check at every handoff, write each result on the card with the time, and say it out loud in the report to the aircrew. If you are a medical officer, put one eye casualty in your next training lane, with a second, louder wound beside it so the eye has to compete for attention. Grade what is under the shield as hard as what is on top of it, and grade whether the antibiotic time made it onto the card. The shield goes on before anything else touches that eye.
— Chester Shermer, MD, FACEP | Emergency Medicine, 25+ Years Clinical Experience | State Surgeon
If you're a combat medic, corpsman, flight medic, or military medical officer (or any clinician treating patients daily) trying to understand how AI will actually impact your clinical practice — not just the hype — I put together a free practical guide. You can download the AI in EM Survival Guide here.
Browse All Courses at Global MedOps Command
Relevant Reading on Global MedOps Command:
How to Avoid Becoming an AI Casualty — Dr. Shermer's guide to navigating AI tools in clinical and operational settings without compromising judgment or patient outcomes.
Emergency Department Efficiency Playbook — Practical systems for throughput, triage optimization, and operational efficiency.
Read more from Dr. Shermer on Medium
Connect with Dr. Shermer: LinkedIn — Chester "Chet" Shermer, MD, FACEP.
Related simulation training: EMS-MedSim for civilian prehospital trauma care · Books by Dr. Shermer