Article Text

Vancomycin and ceftazidime incompatibility upon intravitreal injection
  1. Department of Ophthalmology, Soroka University Medical Center, Ben Gurion University of the Negev, Beer Sheba, Israel
  1. Tova Lifshitz, Department of Ophthalmology, Soroka University Medical Center, Ben Gurion University of the Negev, POB 151, Beer Sheba 84101, Israel

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Editor,—The recommended treatment of endophthalmitis includes intravitreal injection of broad spectrum antibiotics. Vancomycin is the drug of choice for Gram positive bacteria, while Gram negative coverage can be obtained by either amikacin or ceftazidime. Owing to potential retinal toxicity with amikacin, the option of its substitution by ceftazidime seemed very attractive.1-7

Fiscella,8 and the pharmacological handbooks have already described physical incompatibility of vancomycin and ceftazidime.9 10 We encountered this phenomenon in two cases of post-traumatic endophthalmitis, which were treated with intravitreal vancomycin and ceftazidime. Immediately upon injection the antibiotics were seen to form yellow-white precipitates along the needle tract.


Case 1

A 17 year old male was treated for post-traumatic endophthalmitis. Following anterior chamber and vitreous tap and cultures, intravitreal ceftazidime 2.2 mg/0.1 ml and vancomycin 1 mg/0.1 ml were slowly injected, as well as subconjunctival injection ceftazidime 100 mg/0.5 ml and vancomycin 25 mg/ 0.25 ml. This was done using different needles and syringes for each drug. Immediately upon injection dense yellow-white precipitates were observed along the needle tract in the vitreous cavity.

The subconjunctival vancomycin (25 mg/0.25 ml) and ceftazidime (100 mg/0.5 ml) were injected separately at different sites of the lower conjunctiva. Some of both diffused over the conjunctival surface and immediately whitish precipitates formed, which were washed away from the ocular surface.

Postoperatively indirect ophthalmoscopy revealed yellow-white precipitates along the needle tract through which the ceftazidime and the vancomycin were injected.

The vitreous precipitates were observed to dissolve gradually with a complete clearing of the vitreous cavity over a period of 2 months, with a final visual acuity of 1.0 (20/20).

Case 2

A 44 year old man was treated for late post-traumatic endophthalmitis. Anterior chamber and vitreous tap for stains and cultures were performed. Intravitreal ceftazidime 2.2 mg/0.1 ml and vancomycin 1 mg/0.1 ml were slowly injected via different syringes and needles, through the same port. Immediately upon intravitreal injection of the antibiotics, a dense yellow-white material was observed along the needle tract in the vitreous cavity.

Subconjunctival vancomycin and ceftazidime were injected, and again there was instant precipitation over the ocular surface.

On the first postoperative day a white precipitate could be seen in the vitreous cavity along the injection tract of the antibiotics. During follow up the vitreous opacities gradually disappeared over a period of 2 months, with complete resolution. Best corrected visual acuity stabilised at 0.67 (20/30).

In our laboratory we tried to simulate the precipitates formed by vancomycin and ceftazidime. Into the vitreous cavity of a fresh pig's eyes we injected vancomycin and ceftazidime, through one port, with two different syringes, in doses exactly as we had used on the patients. Instant precipitation of the substances occurred (Figs 1 and 2). On a glass slide we put a drop of vancomycin and a drop of ceftazidime and mixed the two: one can see that the mixture is opaque compared with the clear liquids on either side.

Figure 1

On a slide: clear drops of vancomycin and ceftazidime, and the central opaque drop of the mixed antibiotics in which precipitation occurred.

Figure 2

Opacification after injection of vancomycin and ceftazidime into the vitreous cavity of a fresh pig's eye.


We encountered two similar cases of endophthalmitis following perforating trauma, which were treated with intravitreal injection of vancomycin and ceftazidime.

Vancomycin was chosen for its broad spectrum Gram positive coverage, and ceftazidime for the broad spectrum Gram negative coverage it provides.

Recent reports describe retinal toxicity due to aminoglycosides, including amikacin.1 This has led some authors to recommend the use of ceftazidime as an alternative to amikacin.4 6 However, the Endophthalmitis Vitrectomy Study Group prefers the continued use of amikacin.1 The reasons are that even though amikacin is potentially more retinotoxic than ceftazidime, it has a concentration dependent bactericidal effect, is not dependent on the inoculum size for its effect, and is synergistic with vancomycin.5

The theoretical advantages of ceftazidime lie in its lower potential for retinal toxicity, and better efficacy in acidic and hypoxic environments (such as endophthalmitis) compared with amikacin.

From the pharmacological literature it is clear that vancomycin and ceftazidime are incompatible.6 8 Vancomycin and ceftazidime precipitate when mixed. This is thought to be due to the alkaline pH of vancomycin (pH 5–7.5) compared with ceftazidime (pH 2.5–4.5) and the presence of bicarbonate in most ceftazidime preparations, even though precipitation also occurred in preparations that did not contain bicarbonate.9

Most authors who recommend the combination of intravitreal vancomycin and ceftazidime were unaware of its physical incompatibility. Those who were aware thought the clinical consequences of the incompatibility minor.6 Fiscella stated the possibility of the formation of microprecipitates, and the unknown pharmacological consequences of this combination.8 From our experience, the precipitates that form after the injection of vancomycin and ceftazidime clear completely without adverse sequelae within 2 months. It is unclear whether the antibiotic properties of these drugs after precipitation in terms of bioavailability, efficacy, and toxicity, have not been compromised.

Further studies to evaluate those properties should be conducted. Meanwhile we have stopped using the combination of vancomycin and ceftazidime for intravitreal injection, until conclusions can be drawn.