Incidence of Endophthalmitis after Intravitreal Anti-vascular Endothelial Growth Factor: Experience in Saudi Arabia.

PURPOSE: To report the incidence of endophthalmitis, the clinical and microbiological aspects, after intravitreal (IVT) injection of anti-vascular endothelial growth factor.
METHODS: A chart review was performed of patients diagnosed with endophthalmitis after receiving IVT injections of bevacizumab (Avastin) and ranibizumab (Lucentis) presenting to King Khaled Eye Specialist Hospital (KKESH) from May 2006 to December 2012. Endophthalmitis was diagnosed clinically as an intraocular infection with vitreous involvement that required treatment with IVT antibiotics or had undergone pars plana vitrectomy (PPV) to remove the suspected microorganism. Main outcome measures were the incidence of endophthalmitis and the clinical and microbiological features.
RESULTS: Seven cases of endophthalmitis were identified, there was 1 (0.004%) case of endophthalmitis of 22674 IVT injections performed at KKESH. All cases were after IVT bevacizumab. Three (42.85%) cases were culture-positive and caused by Staphylococcus epidermidis. The initial management was vitreous tap and IVT injection of antibiotics followed by PPV in 6 (85.7%) cases. One (14.3%) case underwent evisceration. Visual acuity improved at last visit in only 2 (28.6%) cases. The rate of endophthalmitis was 0.0004% for bevacizumab.
CONCLUSIONS: The rate of endophthalmitis after IVT bevacizumab and ranibizumab was very low. We recommend following a standardized injection protocol, adherence to sterile techniques, and proper patient follow-up are determinant factors for low incidence rates. In addition, endophthalmitis after IVT bevacizumab and ranibizumab have poor visual outcomes despite prompt treatment.

INTRODUCTION

Over the last decade, the use of intravitreal (IVT) injections of bevacizumab (Avastin®; Genentech, South San Francisco, CA) and ranibizumab (Lucentis®; Genentech, South San Francisco, CA) have increased exponentially in retina practices worldwide. Conventional treatment modalities have been replaced due to the success of bevacizumab and ranibizumab as effective pharmacologic therapy to treat macular edema and choroidal neovascularization with minimal complications. However, the increased frequency of use can also increase the potential risk of intraocular infection.1

Although rare endophthalmitis, after IVT injection of bevacizumab and ranibizumab, is a serious complication, which can be irreversible in spite of prompt and appropriate management.23 However, better injection technique, patient education, and use of broad-spectrum antimicrobial solutions for sterilizing the surgical field have controlled the incidence of endophthalmitis.

This study evaluates the incidence and outcomes, and the clinical and microbiological features of endophthalmitis after IVT injection of anti-vascular endothelial growth factors (VEGF).

METHODS

This study was approved by the Institutional Review Board and Ethics Committee at the King Khaled Eye Specialist Hospital (KKESH). Oral informed consent was obtained from all patients. This study adhered to the tenets of the Declaration of Helsinki. None of the participants received a stipend for this study.

A retrospective chart review was performed for all patients with endophthalmitis after undergoing IVT injections of the anti-VEGFs, bevacizumab (Avastin), and ranibizumab (Lucentis) including cases referred to KKESH from May 2006 to December 2012. Endophthalmitis cases were diagnosed clinically as an intraocular infection with vitreous involvement and receiving treatment with IVT antibiotics or had undergone pars plana vitrectomy (PPV) to remove the suspected microorganism. The total number of IVT injections of bevacizumab and ranibizumab at KKESH was calculated using the KKESH pharmacy coding system.

Injection technique

Standard injection protocols were followed for all IVT injections in this study. A trained nurse who followed the standard preinjection protocol for the KKESH retina clinic prepared the injection. Preparation involved instilling a few drops of 5% povidone-iodine and 4% lidocaine drops from single-use dispensers into the conjunctiva. Then, the eyelid skin and lashes were cleansed with 10% povidone-iodine swabs. Topical antibiotics were not used prior to or after the injection. A sterile wire lid speculum was inserted and lashes directed away from the eye. Povidone-iodine and lidocaine drops were instilled in three alternating cycles to the inferior fornix and conjunctiva. Depending on the surgeon's preference, either a lidocaine-soaked cotton-tipped applicator was applied with gentle pressure or xylocaine gel applied at the injection site at least 5 min before the injection. The injection was performed with the physician wearing sterile gloves. For the last 3 years, the physician has been asked to wear a mask as well. The medication was administered through a pars plana approach, followed by instilling 5% povidone-iodine drops. After the injection, retinal artery perfusion was assessed by indirect ophthalmoscopy and visual acuity was checked to the count fingers level. The patients were educated on the visual symptoms and potential risk of endophthalmitis, and to present to the emergency room if necessary. A follow-up visit was scheduled. From May 2006 to May 2008, topical antibiotics were prescribed after the injections routinely; however, this practice has been abandoned for the last 4 years.

Preparation of bevacizumab

Bevacizumab was prepared under laminar airflow hoods by pharmacists following standardized KKESH policies and procedures (infection control policy, pharmacy). Briefly, the pharmacist used a disposable hair cover, yellow isolation gown, and mask, all jewelry including watches were removed, nails must be acceptable length (<1/4 length), and nail polish is prohibited. The hands, nails, and forearm up to elbow should be meticulously washed for at least 30 s with antibacterial soap in addition to using a nailbrush.

The bevacizumab is withdrawn by syringe from the vial ready for 25 syringes (1.25 mg/0.005 ml plus 0.07 ml overfill) into a 1 ml tuberculin syringe cap. The syringes and needles should be disposable. Finally, two syringes are randomly chosen for a sample culture to ensure no bacterial contamination. Each syringe is labeled with the following instructions: Do not filter; do not shake; do not freeze; protect from light; refrigerate; and expired after 14 days.

RESULTS

Seven cases with a diagnosis of endophthalmitis were identified, and all occurred after IVT bevacizumab injections [Table 1]. There was 1 (0.004%) case of endophthalmitis out of 22,674 (21,387 bevacizumab injections and 1287 ranibizumab) IVT injections performed at KKESH. This case presented within 3 days of IVT bevacizumab. The case was culture-positive for Staphylococcus epidermidis. The rate of endophthalmitis after bevacizumab was 0.004% and 0% after ranibizumab injections.

Table 1

Clinical features, indications, management, and outcome of endophthalmitis cases after intravitreal injection of anti-vascular endothelial growth factor*

Table 1 demonstrates the clinical features, indications, management, and outcomes of the seven patients with endophthalmitis after IVT injection of anti-VEGFs. The mean age of the patients with endophthalmitis was 55.4 years old. In all the cases, endophthalmitis was secondary to IVT injection of bevacizumab, and the indication for therapy was complications associated with diabetic retinopathy. Three (42.85%) cases were culture-positive and caused by S. epidermidis. The initial management was vitreous tap and IVT injection of vancomycin and ceftazidime followed by PPV in 6 (85.7%) cases. One (14.3%) case underwent evisceration. Visual acuity improved in 2 (28.6%) cases at the last visit.

DISCUSSION

The reported incidence of endophthalmitis after anti-VEGF IVT injection ranges from 0.02% to 0.08%.45678 A small study reported an incidence of 0.26%.9 However, in the current study, we found a very low incidence of endophthalmitis (0.004%) compared to some reports. This difference might be related to our standard protocol for preinjection preparation. Our results are consistent with a recent retrospective study by Shimada et al.10 that reported a 0% incidence of endophthalmitis. Shimada et al.10 evaluated a total of 15,144 injections comprising 548 injections of pegaptanib sodium, 846 injections of bevacizumab, and 13,750 injections of ranibizumab. They10 recommended a protocol that included wearing surgical masks by operating room personnel (surgeons and nurses), draping patient's face, and irrigating the conjunctiva with 0.25% povidone-iodine prior to performing IVT injection.

Our study concurs with previous reports that prove that reduction of bacteria in the operative field is an important factor in preventing endophthalmitis. The use of povidone-iodine,1112 donning a surgical mask,1113 and silence when a surgical mask is not used1113 have been shown to be effective measures to achieve ocular surface antisepsis. However, use of a lid speculum, conjunctival displacement, injection site (inferior hemisphere or superior hemisphere) location, and type of anti-VEGF agent (bevacizumab or ranibizumab),4 have no significant association with the risk of endophthalmitis.

The value of topical antibiotic before or after IVT injections is uncertain, previous reports showed that the onset of the bactericidal effect of topical antibiotics was longer (approximately 60 min) than povidone-iodine (approximately 60 s). Hence, instillation of topical antibiotics immediately prior to an IVT injection is likely insufficient for adequate bactericidal effect.14 Therefore, installation of topical antibiotics either before the day of injection or immediately prior to injection is not generally recommended.15 A recent prospective study found that topical moxifloxacin 0.5% had no additional effect on reducing conjunctival bacterial counts beyond the effect of 5% povidone-iodine alone.16 Bhatt et al.,17 showed that the incidence of endophthalmitis after IVT injections is similar with and without postinjection prophylactic antibiotics. In addition, Cheung et al.,18 found a higher rate of endophthalmitis after the administration of topical antibiotics given immediately or for 5 days after the injection, compared with no antibiotics. In our hospital, the use of postinjection antibiotics was discontinued 4 years ago, and the one case of endophthalmitis occurred during period when we routinely used postinjection antibiotics. Our outcomes also support the findings that topical antibiotics after anti-VEGF injections are not required. Reports in literature indicate that in a nonsterile environment for IVT injection, bacteria are inoculated into the vitreous, and topical antibiotics after IVT injections do not adequately penetrate the vitreous.19

The initial management for our patients (except one who underwent an evisceration because of a perforated globe) was vitreous tap, and IVT injection of antibiotics followed by PPV due to increasing severity of the condition. The value of a vitreous tap versus PPV as initial management remains uncertain.1 A recent study,20 did not draw any conclusions with regard to the benefit of PPV as a superior management to vitreous tap, and the latter was recommended as a viable primary intervention for endophthalmitis after anti-VEGF injection. In our study, only 2 (28.6%) patients had an improvement in visual acuity (20/80). The poor visual outcome might be attributed to the severity of the infection and, possibly, that the infection may have accelerated the underlying retinal pathology which in our series was proliferative diabetic retinopathy and diabetic macular edema.

The limitations of our study include that it was a retrospective, single-center consecutive case series, and uncontrolled study. Therefore, it is not possible to delineate which factors help prevent endophthalmitis after anti-VEGF IVT injections. However, our study shows that a low incidence of endophthalmitis can be achieved by following a standardized protocol that includes instilling 5% povidone-iodine in the conjunctiva and fornix, cleaning the eyelid skin and lashes with 10% povidone-iodine, a sterile wire lid speculum, wearing sterile gloves, and surgical masks. Another limitation of our study is that we did not have an adequate number of patients to detect an appropriately selected safety outcome rate for ranibizumab (0%). However, we can be reasonably confident (95%) that the true rate of endophthalmitis after IVT ranibizumab is <3/1287, or 0.002%.21

In summary, in this series, a very low rate of endophthalmitis was observed after IVT injections of bevacizumab (0.0004%) and ranibizumab (0%) in an outpatient setting. We believe that following a standardized injection protocol, adherence to sterile techniques, and proper patient follow-up are determinant factors for low incidence rates. This series also showed that endophthalmitis after IVT injections of bevacizumab and ranibizumab have poor visual outcomes despite prompt treatment.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.