• Ocular surface
• cornea
• infection
• pharmacology
• treatment surgery

The human amniotic membrane (AM) possesses anti-inflammatory, antifibrotic and antiangiogenic properties, and these attributes makes it ideal for ocular surface reconstruction procedures.1 2 In addition, the AM also has antimicrobial properties due partly to its anti-inflammatory effects, and also due to secretion of elafin and secretory leucocyte proteinase inhibitor, both of which have antimicrobial actions and act as components of the innate immune system.3 4 It also contains cystatin E, an analogue of cysteine proteinase inhibitors, which has complementary antiviral properties.5 In spite of this, AM transplantation (AMT) is reserved for cases of postinfectious ulcers after an appropriate period of anti-infective treatment when clinical signs are improving.6 This is because the anti-infective properties of AM are non-specific and not considered to be potent enough to be effective in acute infective keratitis; this is the reasoning behind the concept of fortifying AM with antimicrobial drugs to make it a viable therapeutic modality in the setting of active infections of the cornea.

Antibiotic-impregnated medical devices such as catheters, bone and cardiac implants have been in use for over a decade.7 8 Various studies have shown the potential of such an approach—for example, in vascular surgery and arthroplasty, where they appear effective in reducing the risk of bloodstream infections or in limiting deep wound infections.9 10 In ophthalmology, the absorptive capacity and biodegradability of collagen corneal shields led to the development of collagen corneal shields that could deliver antimicrobial, anti-inflammatory, immunosuppressive and anticoagulant drugs.11 In vivo experimental studies showed that they are a suitable alternative to intensive topical treatment.12

The idea that the AM can be used as a drug delivery system akin to collagen shields and other prosthetic devices was first proposed by Kim et al in 2001.6 This concept was also proposed by Heilinghaus et al, who later in 2003 used AM grafts (AMGs) in the management of acute ulcerative and necrotising herpes simplex and zoster keratitis.13 The first successful in vitro results of using AM as a carrier of the antibiotic drug netilmycin (NTM) were reported by Menucci et al in 2006.14 Menucci et al (see page 28) provide us with further evidence of AM acting as a carrier for the antiviral drugs acyclovir (ACV) and trifluridine (TFU).15

Although it is a novel concept that may find extensive clinical utility if the efficacy of the model can be proven in in vitro studies, there are certain issues that must be addressed. The first and most important question is whether AM is capable of absorbing enough of the drug that it is supposed to carry in order to be clinically effective and what should be the strength of the solution in which the AM has to be soaked and for how long. The second is whether the in vitro cytopathic or inhibitory effects observed are due to the drug contained in the AM itself or the free drug that is present in the solution in which the AM is prepared. The third question is whether the same efficacy can be observed if the AM is used without incorporation of any drug as it is well known that AM posses anti-infective properties of its own.

In their previous study, using the aminoglycoside NTM, Menucci et al had shown that the efficacy of the NTM-treated AM (NTM-AM) was dose dependent—that is, the higher the drug concentration the higher the antibacterial activity; however, good antibacterial activity was also obtained with a low NTM concentration, such as 3 mg/ml, which is the concentration in the commercially available eye-drop formulation.14 Similarly in their new study, the efficacy of the ACV- or TFU-treated AM (ACV-AM, TFU-AM) was found to be dose dependent. In both studies care was taken to ensure that unabsorbed drug was washed off before testing the efficacy, so as to rule out any confounding beneficial effect of the unabsorbed drug. The additional and most noteworthy finding of this study, however, is that there was no significant difference between the untreated cell lines and those treated only with AM in terms of inhibition of the cytopathic effect of herpes simplex virus-1 (HSV-1). This means that the beneficial effect seen was solely due to the antiviral content of the AM and not due to the anti-infective properties of the AM itself. In murine models the improvement of HSV ulcerative keratitis after AMG was suggested to be due to reduced activity of matrix metalloproteases-8 and -9 and increased expression of tissue inhibitors of metalloproteases TIMP-1 and TIMP-2.16 AMG in the context of HSV keratitis may be effective in reducing the inflammatory sequelae of the infection but may not be effective in inhibiting active viral replication, and this explains the purpose of fortifying it with antiviral drugs.

Combining the anti-inflammatory properties of the AM and the anti-infective properties of potent antibiotic and antiviral drugs would be ideal for combating microbial and viral infections of the cornea. However, it is unlikely that ACV-AM by itself will be able to provide therapeutic concentrations of ACV for a prolonged period of time, and supplementation with antiviral drops or ointment will be necessary. Another issue that needs attention is faster degradation of AM in inflamed eyes. In spite of this, a drug delivery model can be extremely useful in situations where topical drug application may be plagued with problems such as compliance. The next obvious step will be to attempt to replicate the results in an animal model of viral keratitis, and subsequently in a human trial.