• Users Online: 229
  • Home
  • Print this page
  • Email this page

   Table of Contents      
Year : 1982  |  Volume : 30  |  Issue : 6  |  Page : 635-640

Endophthalmitis : Diagnosis and management

Department of Ophthalmology, University of Illinois Eye and Ear Infirmary, Chicago, USA

Correspondence Address:
Gholam A Peyman
University of Illinois Eye and Ear Infirmary, 1855 W. Taylor S., Chicago, IL 60612
Login to access the Email id

Source of Support: None, Conflict of Interest: None

Rights and PermissionsRights and Permissions

How to cite this article:
Raichand M, Peyman GA. Endophthalmitis : Diagnosis and management. Indian J Ophthalmol 1982;30:635-40

How to cite this URL:
Raichand M, Peyman GA. Endophthalmitis : Diagnosis and management. Indian J Ophthalmol [serial online] 1982 [cited 2021 May 11];30:635-40. Available from: https://www.ijo.in/text.asp?1982/30/6/635/29302

Table 4

Click here to view
Table 4

Click here to view
Table 3

Click here to view
Table 3

Click here to view
Table 2

Click here to view
Table 2

Click here to view
Table 1

Click here to view
Table 1

Click here to view
Although aseptic techniques and newer antibiotics have greatly reduced the incidence of endophthalmits, this complication still conti­nues to occur in clinical practice, especially after surgery and ocular trauma. Traditionally, suspected cases of endophthalmitis have been treated with topical and systemic antibiotics; however, the prognosis for visual recovery with this mode of therapy has been poor. Poor ocular penetration of antibiotics by these routes explain the Poor results. Retrobulbar and subconjunctival injections of antibiotics were considered by some to be preferred methods for the treatment of endophthalmitis. However, in practice, these routes of adminis­tration have not proved to be superior to sys­temic administration of antibiotics. Although higher concentrations may be obtained in the aqueous with retrobulbar and subconjunctival injections, the levels achieved in the vitreous are subtherapeutic.

Good results with conventional m anagement have usually been reported in cases not proved to be caused by bacterial cultures. The good visual prognosis in these cases may be due to the noninfectious nature of sterile endophthalmitis. Newer diagnostic techniques have helped to determine the cause of infections in suspected cases of endophthalmitis. Intravitreally injected antibiotics and vitrectomy surgery have greatly improved the prognosis for bacterial endopha­lmitis in humans.


A good patient history can help to deter­mine whether the source of infection is endo­genous or exogenous. Injury caused by traffic accident or agricultural incident would indicate an exogenous source of infection, perhaps one of mixed bacteria or fungi. Previous eye surgery such as for glaucoma, may also suggest an exogenous infection. Patients receiving immu­nosuppressives or abusers of intravenous drugs. also may have an endogenous source of infec­tion. Determination of the source helps in insti­tuting appropriate systemic antibiotic therapy.

Clinical signs of endophthalmitis, such as blurring of visual acuity, redness, pain, hypo­pyon, and diminished red reflex, may be pre­sent. However all classical findings need to be present in cases of fungal endophthalmitis or anerobic bacterial endophthalmitis. Hypopyon formation may not suggest infection. Sterile hypopyon may be seen after intraocular lens implantation or in severe cases of uveitis, as in Beheet's syndrome. Samples of aqueous humor and vitreous should be obtained fm cultures.

Ultrasound examination using the B•mode helps to assess the status of the vitreous and retina. The presence of an intraocular foreign body, vitreous abscess, heamorrhage, or retinal detachment will influence the immediate management of these eyes. Electroretinography can help in selected cases to assess retinal function. X-ray film of the orbit help rule out suspected cases of intraocular foreign bodies.


  Background Top

In the 1940's von Salimann and others[1],[2], reported using intravitreal injection of peni­cillin to treat bacterial endophthalmitis. Because results were poor and intravitreal injection was considered dangerous, this tech­nique was abandoned.

In 1971, we began revaluating the intra­vitreal route of antibiotic administration to determine the nontoxic dose, clearance time, and efliicacy of intravitreal antibiotics in treat­ing experimental endophthalmitis. Using elec­trophysiologic, histopathologic, and clinical studies on laboratory animals. we were able to evaluate the nontoxic dose of most of the commonly available antibiotics, [2],[3],[4] sub [Table - 2]

Clearance studies showed that nontoxic doses of antibiotics remained in the vitreous for periods up to 96 hours. Studies of experi­mentally induced endophthalmitis in rabbits, showed that intravitreal antibiotics successfully cured endophthalmits within 12 to 24 hours after infection. When compared to systemic and subconjunctival routes of administration, intravitreal antibiotic injection proved superior in salvaging these eyes. [5] sub When dexamethasone was added to the intravitreal antibiotic injec­tions, the inflammation reaction was reduced [3]

sub We also evaluated antifungal agents in the treatment of fungal endophthalmits.[6] sub [Table - 2]

We were able to prove that by using a non­ toxic dose injection of antibiotics is feasible, safe, and effective in the treatment of endoph­thalmitis.

Although intravitreal antibiotics can kill infecting organisms, products of inflammatory reaction remain in the eye for a long time.

Vitrectomy for the treatment of endophthalmitis can help in two ways. First, it can be used to remove organisms, toxins, cellular infiltration, and exudates from the vitreous cavity to allow the media to clear. Second, it can be the primary treatment of endoph­tha lmitis by administering antibiotics directly into the vitreous cavity. We performed a series of experiments to determine the non­toxib dose of antibiotics that can be added to vitrectory infusion fluids.[7],[8] The nontoxic dose of antibiotics that can be injected directly into the vitreous is quite different than that which can be added to vitrectomy infusion fluid. [Table - 3] gives the nontoxic doses of antibiotics for vitrectomy infusion fluid that were safe in experimental studies with animals. Antibiotic selection and dosage for addition to vitrectomy infusion fluid is indicated in [Table - 4].

Management of endophthalmitis

When endophthalmitis is suspected the patient should be hospitalized. A thorough eye examination, including B-scan ultrasono­grapv, electroretinography, and orbit x-ray films, when indicated, is performed. Cultures and smears are prepared for scrapings of the conjunctival and wound margins. Cultures are plated on sheep blood agar, chocolate agar, thioglycolate broth, and Sabrouraud's dextrose. The smears are treated with Giemsa and Gram staining. In the operating room, under sterile conditions, anterior chamber fluid and vitreous samples are obtained.

Following van Lint block and retrobular anesthesia, the eye is fixed by the assistant with forceps. A 25-gauge needle mounted on a tuberculin syringe is introduced into the anterior chamber through the limbus. One tenth milliliter of aqueous fluid is aspirated, and the needle is removed. One tenth milliliter of air is injected into the anterior chamber through the same needle track to re-form the globe. A 2-mm snip incision is made in the conjunctiva 3-mm behind the limbus. The sclera, 4-mm behind the limbus, is cleared at this site. A small Ziegler knife is introduced into the vitreous through a stab incision 4-mm behind the limbus. A 22-gauge needle mounted on a tuberculin syringe is introduced into the vitreous after removal of the knife. Care must be taken to direct the needle toward the center of the vitreous and away from the crystalline lens. One tenth milliliter of vitreous is aspirated, and the needle is removed. The aqueous and vitreous fluids are cultured and smeared as described. One tenth milliliter of antibiotic solution [Table - 1] is injected into the vitreous through the Ziegler knife track. It is very important to inject the antibiotic into the anterior vitreous, behind the anterior vitreous face. The injection should be performed slowly, with the bevel of the needle pointing upward. injection of antibiotic near the retina can result in a local toxic reaction by the retina. One tenth milliliter of antibiotic solution containing 400 ug of gentamicin and 360 ug of dexamethasone is the usual choice for intravitreal injection. When mixed infections are suspected, a mixture of antibiotics is suggested. In cases of fungal endophtha­ Imitis, a single injection of 5 ug of amp­hotericin B is indicated [Table - 1].

Appropriate systemic and topical antibiotic therapy is started at the same time. In cases of exogenous infection, systemic antibiotic therapy may appear to be suprafluous, however this is the present standard of care. We do not advocate the use of subconjunctival antibiotic injections.

If the vitreous are cultures positive within 24 to 48 hours and the vitreous remains hazy, we perform vitrectomy with appropriate anti­biotics in the infusion fluid. The choice of antibiotic [Table - 3] is determined by the vitro sensitivity testing of the organism.

Exceptions are when the vitreous reaction is so severe that the vitreous appears white or when vitreous abscess or retinal detachment is suspected by ultrasonography. Also, when fungal endophthalmitis is suspected, cultures results can delay treatment for 5 or 6 days. Under these circumstances, pars plana vitrectomy is the primary method of management. Before inserting the vitroph­age in the eye, a vitreous fluid sample is obtained through the sclerotomy site. [Table 5] outlines the management of suspected endophthalmitis.[9]

  References Top

Von Sallmann L, Meyer, K, Di Grandhi J 1944 Arch Ophthalmol 32:197  Back to cited text no. 1
Leopold IH: 1945, Arch Ophthalmol 33:211  Back to cited text no. 2
Graham RO, and Peyman GA: 1974, Arch Ophthalmol 92:146  Back to cited text no. 3
Daily MJ, Peyman GA, and Fishman G: 1973, Am J. Ophthalmol 76:343  Back to cited text no. 4
Bennett TO, and Peyman GA: 1974, Can, J Ophthalmol 9:475  Back to cited text no. 5
Axelrod AJ, Peyman GA: 1973, Ani. J. Ophthalmol 76:584  Back to cited text no. 6
Stainer GA, Peyman GA, Meisel H,: 1977. Ann Ophthalmol 9:615  Back to cited text no. 7
Peynian G, Vastine D, Crouch E, Herbst R : 1974, Otolaryngol 78:862  Back to cited text no. 8
Bennett TO, Peyman GA: 1974, Albrecht won Graefes Arch Klin Exp Ophthalmol 191:93  Back to cited text no. 9


  [Table - 1], [Table - 2], [Table - 3], [Table - 4]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
Article Tables

 Article Access Statistics
    PDF Downloaded0    
    Comments [Add]    

Recommend this journal