Retained intraocular foreign bodies and endophthalmitis

Incidence and outcome of endophthalmitis over a 13-year period

OBJECTIVE

The aim of the study was to assess the functional results and long-term visual prognosis in patients treated for acute endophthalmitis (AE).

DESIGN

Observational clinical series.

PARTICIPANTS

Evaluation of 120 eyes of 114 patients with AE treated between 1991 and 2004.

METHODS

This retrospective institutional analysis included 120 eyes of 114 patients (52 male, 68 female) with AE and a minimum follow-up of half a year. The patients had a mean follow-up of 23 months (6-135 months) and a mean age of 61 years (8 months-94 years) at the time of treatment. Treatment included intraocular injection of antibiotics alone (n = 18, 15%) and vitrectomy combined with intraocular and topical postoperative antibiotics (n = 85, 70%).

RESULTS

The most frequent complications were vitreous or retinal hemorrhages (n = 17, 14%), retinal detachment (n = 17, 14%), choroidal detachment (n = 3, 3%), secondary glaucoma (n = 7, 6%), and recurrent endophthalmitis (n = 3, 3%). Four eyes had to be enucleated because of recurrent and uncontrollable infection. Positive microbiological results were achieved in 67 of 88 specimens (76%). The most common isolate was Staphylococcus. At the end of follow-up, visual acuity was reduced (more than 2 lines) in 18 eyes (15%), stable in 22 (18%), and improved (more than 2 lines) in 72 eyes (60%).

CONCLUSIONS

AE is a rare but severe disease with a potentially deleterious outcome in affected eyes. In our series, 78% of all eyes had stable or improved postoperative visual acuity following various treatment regimens, depending on the severity of each case.

Prognostic factors in vitrectomy for posterior segment intraocular foreign bodies

BACKGROUND

The aim of this study was to identify the prognostic factors influencing outcome after the removal of retained posterior segment intraocular foreign bodies (IOFBs) by pars plana vitrectomy.

METHODS

We reviewed the records of 42 consecutive patients (39 men and 3 women with a mean age of 26.8 years) operated between January 2001 and January 2005 by the same surgeon for IOFB after penetrating ocular injury. All eyes underwent IOFB removal via pars plana vitrectomy. Postoperative retinal detachment was considered as the anatomic failure. Associations between anatomic outcome and various preoperative, operative, and postoperative variables were statistically analyzed. Chi-square test and Mann-Whitney U test were used to evaluate the association between two categorical variables.

RESULTS

The mean time interval between the trauma and the IOFB removal was 5.3 days (range, 1 day to 30 days). The mean length of follow-up was 18.3 months (range, 6 months to 3 years). Preoperative retinal detachment was present in 19% of patients. After the IOFB removal, the retina was detached in 28.5% of patients (12 of 42 patients). After a second vitrectomy performed in 10 patients, final anatomic success rate was 89.8% (37 of 42 patients). Time between trauma and IOFB extraction, presence of intraocular hemorrhage, preoperative retinal detachment and primary surgical repair combined with the IOFB removal were significantly associated with the postoperative retinal detachment. Age, sex, entrance wound location, presence of endophthalmitis, location of IOFB, nature of IOFB, preoperative visual acuity, use of an encircling band, type of endotamponade, use of lensectomy were not significantly associated with the presence of postoperative retinal detachment.

CONCLUSIONS

Delay in IOFB extraction, presence of intraocular hemorrhage, preoperative retinal detachment, primary surgical repair combined with IOFB removal are the predictive factors for anatomic failure.

Combined lensectomy, vitrectomy, and primary intraocular lens implantation in patients with traumatic eye injury

PURPOSE

To analyse the postoperative anatomic and functional outcomes in addition to complications after combined lensectomy, vitrectomy, and primary intraocular lens (IOL) implantation in patients with traumatic eye injury.

METHODS

Retrospective review of patients with traumatic cataract and posterior segment injury who underwent combined lensectomy, vitrectomy, and primary IOL implantation with a minimum follow up (FU) of 3 months.

RESULTS

Thirteen consecutive patients (all male) with a mean age of 42.8 years (range 17-82 years) underwent combined lensectomy, vitrectomy, and primary IOL implantation from February 2000 to January 2006. Postoperative FU ranged from 3 to 54 months (mean 17.6 months). Best corrected visual acuity (BCVA) at presentation ranged from 20/30 to hand movement and was worse than 20/200 in eight patients (61%). Four patients (31%) had blunt trauma with no globe rupture. Of the nine patients (69%) with a penetrating eye injury (PEI), eight had an intraocular foreign body (IOFB) with one retinal detachment at presentation. Four patients had primary closure at the time of the vitrectomy. All eight IOFBs were removed. Seven patients had additional scleral buckling and four intravitreal gas injection. BCVA at last FU ranged from 20/20 to 20/300 and was 20/40 or better in eight eyes (62%). All patients had an attached retina at last FU. One eye had further surgery for epiretinal membrane proliferation and ptosis.

CONCLUSIONS

These results suggest that combined vitrectomy, lensectomy and primary intraocular implantation can offer good visual rehabilitation in patients with traumatic cataract and posterior segment injury.

Endophthalmitis: a review of current evaluation and management

PURPOSE

To review the current state of evaluation and management of various forms of endophthalmitis.

METHODS

A review of the literature is included, encompassing endophthalmitis occurring after ocular surgeries, intravitreal injection, trauma, and systemic infection. Based on this review, current principles and techniques for evaluating and treating these forms of endophthalmitis are discussed.

RESULTS

Postoperative endophthalmitis after cataract surgery is the most common presentation. Conclusions from the Endophthalmitis Vitrectomy Study (EVS) remain a foundation for management of postcataract surgery endophthalmitis, notably prompt intravitreal antibiotic administration after vitreous sampling, with consideration for pars plana vitrectomy in severe cases. The potential impact of advances since the EVS, such as oral fourth generation fluoroquinolones and new vitrectomy techniques, are also discussed. The management of postcataract endophthalmitis is compared and contrasted to endophthalmitis occurring after other ocular surgeries, intravitreal injection, trauma, and systemic infection. Although some principles remain common, treatment rationales differ with other forms of endophthalmitis based on differing clinical circumstances, such as the virulence of organisms that are frequently encountered.

CONCLUSIONS

Endophthalmitis is a serious, potentially vision threatening condition which can present in various settings. Prompt recognition and treatment are key in maximizing outcomes.

Prognosis of penetrating eye injuries with posterior segment intraocular foreign body

PURPOSE

To identify the prognostic factors concerning the anatomy and visual acuity of eyes subject to trauma related posterior intraocular foreign body.

PATIENTS AND METHODS

The records of 28 eyes of 27 patients who underwent pars plana vitrectomy and intraocular foreign body removal during a 5 year period were retrospectively reviewed. Ocular trauma score was calculated for each eye.

RESULTS

The most common initial findings were corneal wound (68%), lens injury (50%), retinal lesion (50%), vitreous hemorrhage (25%), and endophthalmitis (14%). Multiple foreign body causing perforating injury with retained posterior segment foreign body occurred in 7% of the cases. The foreign body was found on the surface of the retina in 39% of the cases. Postoperative complications were retinal detachment (46%), proliferative vitreoretinopathy (25%), and phthysis (4%). No eye was enucleated and 1 eye (4%) lost light perception. The final best corrected visual acuity became better or equal to 0.5 Snellen E in 34% of the eyes. The mean follow-up was 19 months (1.5-60 months).

CONCLUSIONS

Prognosis was significantly worse in cases with lower trauma score, initial visual acuity less than 0.1 Snellen E, large foreign body, upset of bacterial endophthalmitis, and with proliferative vitreo-retinopathy. Visual outcomes in our cases were better than estimated follow-up visual acuity based on ocular trauma score parameters.

Delayed Intraocular Foreign Body Removal without Endophthalmitis during Operations Iraqi Freedom and Enduring Freedom

OBJECTIVE

To report the long-term follow-up results of intraocular foreign body (IOFB) removal at Walter Reed Army Medical Center during Operation Iraqi Freedom and Operation Enduring Freedom from February 2003 through November 2005 and to determine the prognostic factors for visual outcome in this patient population.

DESIGN

Retrospective, noncomparative, interventional case series.

PARTICIPANTS

Seventy-nine eyes of 70 United States military soldiers deployed in support of operations Iraqi Freedom and Enduring Freedom sustained IOFB injuries and subsequently were treated at the Walter Reed Army Medical Center with a minimum of 6 months of follow-up.

INTERVENTION

The principal procedure performed was 20-gauge 3-port vitrectomy with IOFB removal through limbal or pars plana incision.

MAIN OUTCOME MEASURES

Final visual acuity, rate of proliferative vitreoretinopathy, rate of endophthalmitis.

RESULTS

Average patient age was 27 years, with an average of 331 days of postoperative follow-up. Average IOFB size was 3.7 mm (range, 0.1-20 mm). Median time to IOFB removal was 21 days (mean, 38 days; range, 2-661 days). Mean preoperative visual acuity was 20/400 (1.36 logarithm of mean angle of resolution [logMAR] units) and mean final visual acuity was 20/120 (0.75 logMAR). Of the patients, 53.4% achieved visual acuity of 20/40 or better, whereas 77.5% achieved visual acuity of better than 20/200. There were no cases of endophthalmitis (0/79 eyes; 95% confidence interval, 0%-3.1%), siderosis bulbi, or sympathetic ophthalmia. Among the eyes, 10.3% evolved to no light perception or had been enucleated by the 6-month follow-up visit. Poor visual outcome correlated with extensive intraocular injury (P<0.032). Seventeen of 79 eyes (21%) experienced proliferative vitreoretinopathy. Proliferative vitreoretinopathy correlated with poor initial vision (hand movements or worse; P = 0.035) and extensive intraocular injury (P<0.001). Timing of vitrectomy did not correlate with visual outcome. The most common systemic antibiotic administered was levofloxacin, whereas the most common topical antibiotic administered was moxifloxacin.

CONCLUSIONS

Poor visual outcome and postoperative complication rates are related to extensive intraocular injury. Delayed IOFB removal with a combination of systemic and topical antibiotic coverage can result in good visual outcome without an apparent increased risk of endophthalmitis or other deleterious side effects.

Incidence and visual outcome of endophthalmitis associated with intraocular foreign bodies

PURPOSE

To determine the risk factors and visual outcome of endophthalmitis associated with traumatic intraocular foreign body (IOFB) removal and its allied management.

METHODS

A retrospective review was conducted of patients with penetrating eye trauma and retained IOFB with associated endophthalmitis managed at King Khaled Eye Specialist Hospital over a 22 year period (1983 to 2004).

RESULTS

There were 589 eyes of 565 patients (90.3% male; 9.7% female) which sustained ocular trauma and had retained IOFB that required management. Forty-four eyes (7.5%) developed clinical evidence of endophthalmitis at some point after trauma. From these 44 eyes, initial presenting visual acuity (VA) of 20/200 or better was recorded in 8 eyes (18.1%) and the remaining 36 eyes (81.9%) had VA ranging from 20/400 to light perception. Eleven eyes (25%) underwent IOFB removal and repair within 24 hours after trauma while 33 eyes (75%) had similar procedures done 24 hours or more after trauma. Thirty-one eyes (70%) underwent primary pars plana vitrectomy (PPV) at the time of removal of posteriorly located IOFBs. Definite positive cultures were obtained from 17 eyes (38.6%). Over a mean follow-up of 24.8 months, 21 eyes (47.7%) had improved VA, 6 eyes (13.6%) maintained presenting VA while 17 eyes (38.7%) had deterioration of their VA, including 10 eyes (22.7%) that were left with no light perception (NLP) vision. After the treatment of endophthalmitis, 20 eyes (45.4%) had VA of 20/200 or better at their last follow-up. Four eyes (12.9%) from the vitrectomy group (31 eyes) and 5 eyes (45.4%) from non-vitrectomy (11 eyes) group had final VA of NLP. Predictive factors for the good visual outcome included good initial presenting VA, early surgical intervention to remove IOFB (within 24 hours), and PPV. Predictors of poor visual outcome included IOFB removal 48 hours or later, posterior location and no PPV for the posteriorly located IOFB.

CONCLUSIONS

Delayed removal of IOFB following trauma may result in a significant increase in the development of clinical endophthalmitis. Other risk factors for poor visual outcome may include poor initial presenting VA, posterior location of IOFB and no vitrectomy at the time of IOFB removal.

Outcomes of surgery for posterior segment intraocular foreign bodies—a retrospective review of 17 years of clinical experience

BACKGROUND

To determine the outcome of surgery in patients who presented with posterior segment intraocular foreign bodies (IOFBs) and to identify factors associated with poor visual outcome. To assess whether improvement in vitreo-retinal surgical techniques over the past 17 years has resulted in an improvement in the outcome of these patients.

METHODS

A retrospective cohort study carried out at Moorfields Eye Hospital, London. Patients who underwent a primary surgical repair and removal of a posterior segment IOFB were identified using a hospital database. Of the 140 patients identified, 114 (81%) were included in the study. Data on age, activity at time of injury, presenting clinical features, surgical techniques, complications and final visual outcome were recorded and analyzed.

RESULTS

The mean age at time of injury was 34.6+/-12.4 years. All patients were male. The majority of IOFBs were metal (83%). Hammering and chiselling were the most common activities at the time of injury (66%). Fifty-three percent of patients presented with a visual acuity of 6/ 36 or better. Primary closure was achieved in a median of 1 day (interquartile range 1-3 days) following injury. A final visual outcome of 6/60 or worse was documented in 31%. Presentation with a relative afferent pupillary defect (P<0.001), hyphaema (P=0.024), vitreous haemorrhage (P=0.003), uveal prolapse (P=0.038) or retinal detachment (P=0.031) was associated with a poor visual outcome (6/60 or worse). Postoperative complications of retinal detachment (P<0.001) and proliferative vitreoretinopathy (P<0.001) were prognostic of poor final visual acuity. Improvements in surgical techniques, e.g. panoramic viewing systems has resulted in a significant improvement in the final visual outcome of these patients (P=0.043).

CONCLUSION

This large series of posterior segment IOFBs shows that recent advances in surgical techniques are associated with a significant improvement in prognosis.

Intralenticular intraocular foreign body after stone impact: CT and US findings

INTRODUCTION

Intraocular foreign body (IOFB) is a relatively common entity in emergency departments worldwide. Appropriate ocular assessment is mandatory if an intraocular foreign body is suspected because it is associated with an increased risk of endophtalmitis and a wide range of complications including hyphaema, cataract, vitreous hemorrhage, and retinal tears and detachment.

CASE REPORT

We present a case of intralenticular intraocular foreign body after stone impact.

DISCUSSION

Ultrasonography (US) and computed tomography (CT) show an accurate location of the foreign body inside the lens. This finding enabled the surgeon to perform a phacoemulsification lens extraction with removal of the foreign body, the optimal method of removing intralenticular IOFB. To our knowledge, this is the first US and CT imaging report.

Intraocular foreign body injuries during Operation Iraqi Freedom

OBJECTIVE

To evaluate the number of intraocular foreign body (IOFB) injuries that occurred in Operation Iraqi Freedom, and to determine the cause of injury, the type of foreign body, and the associated injuries to other body systems.

DESIGN

Retrospective, noncomparative, interventional case series.

PARTICIPANTS

Fifty-five United States military personnel with an IOFB injury during Operation Iraqi Freedom.

INTERVENTION

Pars plana vitrectomy, foreign body removal, and additional surgical procedures as the clinical situation dictated.

MAIN OUTCOME MEASURES

Cause of injury, size of corneal/scleral laceration, number of foreign bodies, type of foreign body, time to foreign body removal, visual acuity, number of enucleations, and injuries to other body systems.

RESULTS

The foreign body was caused by a propelled explosive in 20 patients (36%) and a nonpropelled explosive in 31 patients (56%), and the cause of the foreign body was not known in 4 patients (7%). The size of the laceration of the cornea and/or sclera averaged 5.4 mm (range, 0.2-18). There were an average of 1.7 foreign bodies in the injured eye (range, 1-6). The size of those foreign bodies measured ranged from <1 mm to 12 x 14 mm. The most common type of foreign body was metal (68%), followed by glass (14%), stone/cement (14%), bone (5%), and cilia (3%). The time from injury to foreign body removal averaged 20.6 days (range, 0-90). No cases of endophthalmitis were seen. The most common associated injury was to the upper extremity, face, lower extremity, and neck.

CONCLUSIONS

Unlike trauma in the civilian sector, IOFB injuries in a military setting tend to be caused by explosive devices, which often result in multiple foreign bodies and simultaneous injuries to other body systems. Because of the lack of availability of specialty care in the combat theater, there is often a delay in removal of the foreign body.

Management of endophthalmitis with retained intraocular foreign body

AIMS

To assess the impact of primary injection of intravitreal antibiotics and delayed pars plana vitrectomy with removal of intraocular foreign body (IOFB) in patients with clinical features of bacterial endophthalmitis and retained IOFB.

METHODS

Retrospective review of all patients with clinical features of infective endophthalmitis and a retained IOFB who had immediate injection of intravitreal antibiotics and delayed pars plana vitrectomy with removal of IOFB in two vitreo-retinal centres during 1995-2001. Nine patients were identified and minimum follow-up was 3 months.

RESULTS

Four of the nine patients had a final visual outcome of 6/18 or better. One patient developed total retinal detachment.

CONCLUSIONS

The current series suggests that immediate injection of intravitreal antibiotics with delayed removal of IOFB in eyes with clinical features of infective endophthalmitis and a retained IOFB is a possible alternative to immediate removal of IOFB. This management may be associated with preservation of the eye and restoration of useful visual acuity.

VITRECTOMY FOR POSTERIOR SEGMENT INTRAOCULAR FOREIGN BODIES

PURPOSE

To evaluate the visual results and determine the prognostic factors after the removal of retained posterior segment intraocular foreign bodies (IOFBs) by pars plana vitrectomy.

METHODS

A retrospective study of 40 consecutive cases of pars plana vitrectomy for the removal of IOFBs during the period of January 1991 to June 2000 was conducted. Association between visual outcome and various preoperative, operative, and postoperative variables was statistically analyzed.

RESULTS

Of the 40 cases available for the study, the IOFBs could be removed in 36 eyes. Half the patients had removal of the IOFB within 14 days of the injury. Nineteen patients (47.5%) achieved a visual acuity of 20/40 or better. Postoperative retinal detachment occurred in 13 patients (32.5%). Poor visual outcome was found to be significantly associated with poor initial visual acuity and postoperative retinal detachment, whereas good visual outcome was significantly associated with the absence of lens injury and absence of postoperative retinal detachment.

CONCLUSIONS

Acceptable visual results could be achieved after the removal of posterior segment IOFBs by vitrectomy, despite a delay in surgery. Postoperative retinal detachment, poor initial visual acuity, and the absence of lens injury were the factors that played a significant role in determining the visual outcome.

Open globe injuries with positive intraocular cultures: factors influencing final visual acuity outcomes

PURPOSE

To investigate the clinical features influencing final visual acuity outcomes of eyes with positive intraocular cultures after open globe injuries.

DESIGN

Retrospective, consecutive, interventional case series.

PARTICIPANTS

Thirty-seven patients.

METHODS

The medical records were reviewed of all patients with positive intraocular cultures after open globe injuries treated at Bascom Palmer Eye Institute between January 1, 1995, and December 31, 2001.

MAIN OUTCOME MEASURES

Final visual acuity. Clinical features investigated included the following: (1). presence or absence of clinical endophthalmitis; (2). virulence of the cultured organism (coagulase-negative Staphylococci, Corynebacterium, and Propionibacterium acnes were classified as nonvirulent organisms, whereas all other organisms were classified as virulent organisms); (3). presence of intraocular foreign body (IOFB); (4). presence of retinal detachment; (5). interval between ocular injury and surgical repair; (6). severity of vision loss at presentation; (7). zone of injury; (8). wound length; and (9). presence of vitreous hemorrhage.

RESULTS

The study included 37 eyes of 37 patients with a mean age of 30 years (range, 18 months-85 years) and a median follow-up of 13 months (range, 1-71 months). Study eyes were stratified into two groups: group 1 eyes (n = 16) were those in which clinical endophthalmitis did not develop, whereas group 2 eyes (n = 21) were those in which clinically diagnosed endophthalmitis developed at some point during their clinical course. Presenting visual acuity was similar in the two groups (mean logarithm of the minimum angle of resolution [logMAR] acuity, 1.91 and 2.22 [Snellen equivalents, 2/162 and 2/331] respectively; P = 0.33). Final acuities in the two groups were different, but not to a statistically significant level (mean logMAR acuity, 1.14 and 2.05 [Snellen equivalents, 20/276 and 2/224], respectively; P = 0.069). In group 1, final visual acuity ranged from 20/20 to no light perception (median acuity, 20/186); 12 eyes (75.0%) achieved a final visual acuity of 20/400 or better. In group 2, final visual acuity ranged from 20/25 to no light perception (median acuity, 7/200); of 20 eyes with known final visual acuity, 10 (50.0%) retained 20/400 or better vision. In group 1, three eyes (19%) eyes had virulent organisms, and 13 eyes (81%) had nonvirulent organisms. In group 2, 12 eyes (57%) had virulent organisms, and nine eyes (43%) had nonvirulent organisms. A final acuity of 20/60 or better was achieved in 14 eyes (41%), and a final acuity of 20/400 or better was achieved in 22 eyes (59%). Better presenting visual acuity (P = 0.038), culture of a nonvirulent organism (P = 0.011), lack of a retinal detachment (P = 0.002), absence of clinical endophthalmitis (P = 0.069), and shorter wound length (P = 0.024) were associated with better visual acuity outcome. In four of six eyes (67%) with both an IOFB and clinical endophthalmitis (group 2), the final visual acuity was no light perception (IOFB was not itself significantly associated with final visual acuity; P = 0.11).

CONCLUSIONS

Among eyes with positive intraocular cultures after open globe injury, the visual prognosis is guarded. Clinical features associated with better visual acuity outcomes include better presenting visual acuity, culture of a nonvirulent organism, lack of a retinal detachment, absence of clinical endophthalmitis, and shorter wound length.

Endophthalmitis after deadly-weapon-related open-globe injuries: risk factors, value of prophylactic antibiotics, and visual outcomes

PURPOSE

To identify clinical and microbiologic factors influencing the prevalence and visual outcomes in endophthalmitis after deadly-weapon-related open-globe injuries.

METHODS

In a retrospective study of 228 eyes of 212 patients with deadly-weapon-related open-globe injuries, clinical and microbiologic factors influencing the prevalence and visual outcome in endophthalmitis were analyzed in detail.

RESULTS

Twenty-nine eyes enucleated for irreparable damages at presentation were excluded. Nineteen eyes (18 patients) of 199 (186 patients) were associated with culture-proven endophthalmitis. Presence of grade 4 injury, a variable of Ocular Trauma Classification System, and lens disruption at presentation were the significant risk factors for development of endophthalmitis (P =.001; odds ratio = 15.9 [2.0 to 122.1]), and ([P <.001; odds ratio = 17.7 [2.3 to 136.3]), respectively. Favorable outcome (visual acuity of 5/200 or better) was achieved in only four eyes (21%). Five eyes (26.3%) were phthisical, and five eyes (26.3%) were enucleated or eviscerated. All eyes were under the coverage of prophylactic intravenous antibiotics when culture positivity was obtained. Except two eyes with Acinetobactersp., all were infected by gram-positive micro-organisms (89.5%), the most common of which (42%) was Staphylococcus epidermidis. None of the clinical characteristics and treatment modalities affected final visual outcome. Favorable outcome was associated with infection by the less virulent micro-organism S. epidermidis (P =.018; odds ratio = 0.50 [0.25 to 1.00]).

CONCLUSIONS

Endophthalmitis after deadly-weapon-related open-globe injuries has a dismal visual prognosis. Severity of injury and lens disruption are significant risk factors for development of endophthalmitis. Infection with a less virulent microbe is the only factor associated with favorable outcome.

Visual outcome and prognostic factors after vitrectomy for posterior segment foreign bodies

PURPOSE

To identify the prognostic factors that predict final visual outcome in eyes with posterior segment intraocular foreign body (IOFB) injuries managed by primary pars plana vitrectomy.

METHODS

Ninety-six consecutive patients with posterior segment IOFB injuries were retrospectively reviewed. Factors analyzed included initial visual acuity (VA), time between injury and presentation, site of entrance wound, uveal prolapse, vitreous prolapse, traumatized iris, endophthalmitis, location and size of IOFB, use of scleral buckling and/or an encircling band, gas tamponade, lensectomy, number of surgical procedures, and development of retinal detachment. Data were analyzed using univariate and multivariate logistic regression analysis.

RESULTS

After a mean follow-up of 8.6 months, 63 eyes (65.6%) achieved VA of 20/200 or better, and 9 eyes (9.4%) had total retinal detachment complicated by inoperable proliferative vitreoretinopathy. On univariate analysis, predictors of poor vision (hand movements or less) were poor initial VA, corneoscleral entrance wound, uveal prolapse, vitreous prolapse, traumatized iris, and development of retinal detachment. In contrast, predictors of good visual outcome (20/200 or better) were absence of uveal prolapse, no endophthalmitis, and no retinal detachment. Multivariate analysis identified corneoscleral entrance wound, uveal prolapse, and development of retinal detachment as the only factors significantly associated with poor visual outcome. Absence of uveal prolapse was the only factor significantly associated with good visual outcome.

CONCLUSIONS

Final visual outcome is greatly determined by the severity of the primary injury. On multivariate analysis, significant predictive factors of final VA were corneoscleral entrance wound, presence or absence of uveal prolapse, and development of retinal detachment.

Prognostic factors in ocular injuries caused by intraocular or retrobulbar foreign bodies

OBJECTIVE

To evaluate prognostic factors associated with final visual outcome, development of posttraumatic infectious endophthalmitis, and occurrence of proliferative vitreoretinopathy in patients with penetrating ocular injuries caused by intraocular or retrobulbar foreign bodies (FBs).

DESIGN

Clinic-based cross-sectional study.

PARTICIPANTS

One hundred thirty patients presenting with penetrating ocular injuries caused by lacerations from FBs were operated on by one of two surgeons between 1989 and 1997. Follow-up time was an average of 20.84 +/- 20.76 months (median: 17.0 months). All FBs were located posterior to the lens.

INTERVENTIONS

Pars plana vitrectomy; foreign body removal; additional surgical procedures according to the clinical situation.

MAIN OUTCOME MEASURES

Postoperative visual acuity; posttraumatic infectious endophthalmitis; proliferative vitreoretinopathy.

RESULTS

Occurrence of posttraumatic infectious endophthalmitis developing in seven patients (7/130 = 5.4%) was significantly (P = 0.026) associated with removal of the FB later than 24 hours after the accident and with the type of the FB (P < 0.01). Size (P = 0.37) of the FB, preoperative visual acuity (P = 0.62), presence of traumatic cataract (P = 0.75) or a retinal lesion by the FB (P = 0.16), age (P = 0.39), and gender (P = 0.46) did not show a statistically significant influence on the occurrence of endophthalmitis. Statistically significant risk factors for the development of proliferative vitreoretinopathy occurring in 27 patients (27 of 99 [27.6%] patients with a minimal follow-up of 3 months) were size of the FB (P < 0.001), preoperative visual acuity (P = 0.02), presence of a retinal lesion (P = 0.002), and traumatic cataract (P = 0.03). The time between FB removal and the accident was statistically marginally associated with the development of proliferative vitreoretinopathy (P = 0.07). Postoperative visual acuity depended significantly on size of the FB (P = 0.002), preoperative visual acuity (P < 0.001), presence of a retinal lesion (P = 0.049), and location of the retinal lesion (P < 0.001). Three eyes had to be enucleated because of endophthalmitis or phthisis bulbi.

CONCLUSIONS

Prognosis in open-globe injuries with intraocular or retrobulbar foreign bodies depends on the size and type of the foreign body, presence and location of retinal lacerations, additional involvement of other intraocular structures, preoperative visual acuity, and timing of surgery. These factors may be important in preoperative counseling of the patient and for planning surgery.

Intraocular foreign bodies: management, prognostic factors, and visual outcomes

OBJECTIVE

To determine prognostic factors and visual outcomes in patients with intraocular foreign bodies (IOFBs).

DESIGN

Retrospective, noncomparative, interventional case series.

PARTICIPANTS

Fifty-nine consecutive patients undergoing removal of an IOFB with a minimum of 6 months of follow-up.

INTERVENTION

Surgical removal of the IOFB and repair of associated ocular trauma.

MAIN OUTCOME MEASURES

Final best corrected visual acuity was the main outcome measured. Ocular findings at presentation were compared with final visual acuity to determine prognostic factors for visual outcome.

RESULTS

Final best corrected visual acuity of 20/40 or more was obtained in 42 patients (71%) and ambulatory vision (>5/200) was achieved in 50 patients (85%). Presenting visual acuity was predictive of final visual outcome (P < 0.01). Prognostic factors for a better visual outcome (P < 0.05) included better presenting visual acuity and hammering metal on metal as the mechanism of injury. Prognostic factors for a poor outcome (P < 0.05) included poor presenting visual acuity, the presence of an afferent pupillary defect, and vitreous hemorrhage.

CONCLUSIONS

Final visual outcomes were excellent in 71% of patients. Presenting visual acuity was the strongest predictor of final visual outcome in this series. Additional predictive factors included the mechanism of injury, the presence of an afferent pupillary defect, and vitreous hemorrhage.

Controversies in the management of open-globe injuries involving the posterior segment

There are numerous unresolved issues and controversies regarding the management of open-globe injuries involving the posterior segment. These areas include, but are not limited to, the following issues. Although vitrectomy has been shown to improve visual outcomes and allow retention of the eye in many cases, the extent of visual improvement is often limited because of the nature of the injury. Timing of vitrectomy surgery has been and will continue to be debated by proponents of early versus delayed intervention. The multiple features of acute ocular injury make it very difficult to interpret retrospective data regarding the most appropriate timing for surgical intervention. The use of prophylactic cryotherapy, in the setting of a scleral laceration with possible retinal damage, is not as controversial at present, as there is now sufficient data indicating that cryotherapy may actually exacerbate intraocular proliferation and worsen the situation. The role and benefit of a prophylactic scleral buckle is very widely contested, and it is not known if it truly decreases the risk of subsequent retinal detachment. Another area of debate centers on the use of antibiotics. When there is a known clinical infection, intravitreal antibiotics are the mainstays of therapy. However, in the absence of clinical infection, the use of prophylactic antibiotics and their routes of administration are quite controversial. Although there are significant data regarding the use of antibiotics in the postoperative setting, this information cannot be extrapolated into the setting of open-globe injuries, as organisms and virulence factors differ. Similarly, the use of vitrectomy versus vitreous tap in the setting of traumatic endophthalmitis is not fully resolved, although vitrectomy is used in most cases to repair concurrent damage from the injury itself. Finally, the placement of intraocular lenses in the acute trauma setting is controversial, as the risk of complications is quite high. Prospective, controlled clinical studies have not been done. This article reviews pertinent data regarding these management issues and controversies, and provides recommendations for treatment based on the available published data and the authors' personal experience.

Clostridium beijerinckii endophthalmitis secondary to penetrating ocular injury

Endophthalmitis occurs in five to 10% of injuries involving intraocular foreign bodies. A 52 year old abattoir worker sustained such penetrating ocular trauma and developed fulminant endophthalmitis. Clostridium beijerinckii was isolated from the vitreous humor. Intravitreal vancomycin and amikacin and intravenous penicillin and clindamycin were given. Despite therapeutic vancomycin and amikacin levels in the vitreous, vision was lost and enucleation was ultimately required.

Post-traumatic endophthalmitis: causative organisms and visual outcome

PURPOSE

Post-traumatic endophthalmitis makes up a distinct subset of intraocular infections. The purpose of the present study was to identify the causative organisms and record the visual outcome after infectious endophthalmitis in eyes with penetrating trauma.

METHODS

We reviewed 18 consecutive cases of culture-positive endophthalmitis that developed after penetrating ocular trauma. All cases were treated with pars plana vitrectomy and intravenous and intraocular antibiotics.

RESULTS

The 15 males and 3 females ranged in age from 4 to 43 years (mean 25.1 +/- 11 years). Nine (50%) had intraocular foreign bodies. A single species was isolated in 16 cases, and multiple organisms in two. Staphylococcus epidermidis and gram-negative organisms were the most frequent and were cultured either alone or in association with other organisms in respectively five (27.7%) and four cases (22.2%). Clostridium perfringens was isolated in three cases (16.6%). Bacillus was not found as a cause of endophthalmitis. Final visual acuity was better than 20/400 in eight cases (44%). In five cases (27.7%), the eye was saved but visual acuity was counting fingers. Two eyes (11%) had no light perception. The remaining three eyes (16.6%) were enucleated or eviscerated. Clostridium perfringens was isolated from two eyes and Aspergillus niger from one. Postoperative retinal detachment developed in four eyes, which were successfully operated.

CONCLUSIONS

Organisms isolated in this series were similar to those in previous reports of post-traumatic endophthalmitis from other parts of the world, except that the frequency of Clostridium perfringens isolation was high and no Bacillus species were cultured. In view of its devastating outcome, post-traumatic endophthalmitis must be treated promptly with vitrectomy and intravitreal antibiotics.

Endophthalmitis

Endophthalmitis is an inflammatory reaction of intraocular fluids or tissues. Infectious endophthalmitis is one of the most serious complications of ophthalmic surgery. Occasionally, infectious endophthalmitis is the presenting feature of an underlying systemic infection. Successful management of infectious endophthalmitis depends on timely diagnosis and institution of appropriate therapy. Recognition of the different clinical settings in which endophthalmitis occurs and awareness of the highly variable presentation it may have facilitate timely diagnosis. Biopsy of intraocular fluid/tissue is the only method that permits reliable diagnosis and treatment. The different presenting clinical settings, a rational approach to diagnosis (i.e., when, what, and how to biopsy), and the treatment of infectious endophthalmitis are reviewed.

Management of intra-ocular foreign bodies impacting or embedded in the retina

PURPOSE

To identify both the clinical features of eyes suffering penetrating ocular injury with intra-ocular foreign bodies (FB) impacting or embedded in the retina and the results of surgical outcome and complication rates.

METHODS

Thirty-eight consecutive cases of penetrating ocular injuries with intra-ocular FB impacting or embedded in the retina were retrospectively analysed.

RESULTS

The FB were metallic in 30 eyes (78.9%; 19 ferromagnetic) and were non-metallic in eight eyes (21.1%). Magnetic extraction was performed in 17 cases (44.7%) and instrumental extraction was performed in 21 cases (55.3%). In two cases, pars plana vitrectomy and FB removal was combined with penetrating keratoplasty, temporary keratoprosthesis and anterior segment reconstruction because of severe explosive injury to the eye. At the final postoperative examination, 32 eyes (84.2%) had completely attached retinas. Visual acuity (VA) improved postoperatively in 18 eyes (47.4%). A final VA of 0.5 or better was achieved in nine of 23 eyes (39.1%) with FB in the peripheral retina and in one of 15 eyes (6.7%) with posterior pole and/or optic nerve head involvement. There was no difference in the final VA regarding the time elapsed between trauma and the removal of the FB.

CONCLUSION

The late anatomical and functional outcome in eyes with penetrating ocular injury and FB impacting or embedded in the retina is mostly uncertain, despite immediate and complex surgery. Final anatomical and functional outcome in the present study was influenced by location and extent of the initial damage to the posterior pole. The time that had elapsed between trauma and removal of the FB did not influence final visual function.

Epidemic Bacillus endophthalmitis after cataract surgery I: acute presentation and outcome

OBJECTIVE

The purpose of the study is to report the clinical outcome of acute Bacillus endophthalmitis after cataract surgery.

DESIGN

The study design is a cohort study.

PARTICIPANTS

Fourteen eyes of 14 patients with epidemic acute postoperative inflammation after exposure to bacteria-contaminated viscoelastic material were studied.

INTERVENTION

Three patients with milder clinical presentations were treated without vitrectomy or antibiotics. Eleven patients with more severe infection were treated with vitrectomy as well as intravitreous and topical fortified antibiotics.

MAIN OUTCOME MEASURES

Final visual acuities were obtained in all 14 study patients. Results of microbiologic studies of aqueous and vitreous specimens from 11 vitrectomized eyes also were analyzed.

RESULTS

One patient with late presentation had severe inflammation and had phthisis bulbi develop with no light perception. The remaining 13 patients had successful resolution of inflammation after treatment by 1 month of follow-up. Twelve of these 13 patients, including 1 nonvitrectomized patient, had final visual acuities of 20/100 or better at 6 months' follow-up. Six patients, including two patients with nonvitrectomized eyes, had outcomes of 20/40 or better visual acuity. Bacillus species were grown from all 11 (100%) vitreous and 7 (88%) of 8 aqueous specimens obtained from vitrectomized patients, as well as from the contaminated viscoelastic material remaining in the operating room.

CONCLUSIONS

Postoperative Bacillus endophthalmitis need not result in poor outcome. Results depend on factors including bacterial load, specific bacterial species, timing of treatment, and immune status of the patient.

Endophthalmitis following penetrating eye injuries

Postinjury endophthalmitis is the eye infection with the worst prognosis. A retrospective 9-year study was made of penetrating eye injuries, with an analysis of the incidence of infection and its relation to the type of wound and the presence of intraocular foreign bodies. There were 403 cases of penetrating eye injury; of these, 233 affected the cornea and 170 involved the posterior pole. Intraocular foreign bodies were present in 40 cases. Endophthalmitis developed in 4.2% of cases (17/403), and was more common in patients with posterior pole involvement (7%) than in purely corneal trauma (2.1%) (p = 0.03, Chi-square). Infection was in turn more frequent in the presence of intraocular foreign bodies (15%) (p = 0.17, Chi-square). Staphylococcus epidermidis was the most common cause (23.4%), while in three cases (17.6%) mixed infection was detected. The visual results were evisceration or non-perception of light in 82.3% of cases.

Useful visual outcomes after treatment of Bacillus cereus endophthalmitis

PURPOSE

Bacillus cereus endophthalmitis occurring after penetrating ocular trauma has been almost always associated with a poor visual outcome. The purpose of our study was to review and report patients who had useful visual acuity outcomes.

METHODS

The study group consisted of five patients from a single medical center with penetrating ocular trauma and endophthalmitis caused by B. cereus. The study population was derived from a review of the microbiology records, clinical records, and operative reports of patients with culture-proven, post-traumatic endophthalmitis over a 15-year period. Patients were only included if the final visual acuity outcomes were 20/200 or better.

RESULTS

All five patients had penetrating ocular injuries, and four patients had a retained intraocular foreign body. Endophthalmitis was diagnosed preoperatively in three patients and intraoperatively in two patients. All patients underwent pars plana vitrectomy and injection of intravitreal and periocular antibiotics. Postoperatively, a rhegmatogenous retinal detachment developed in three patients between 4 weeks and 12 months after the injury (average, 19 weeks); all retinal detachments were reattached with additional vitreoretinal surgery. Final postoperative visual acuities were 20/200 (two patients), 20/30 (one patient), and 20/25 (two patients). The postoperative follow-up time interval ranged from 12 months to 30 months (average, 19.2 months).

CONCLUSION

The current series adds further support to the observation that certain eyes with post-traumatic B. cereus endophthalmitis may be associated with preservation of anatomic integrity and restoration of useful visual acuity.

Endophthalmitis after penetrating trauma. Risk factors and visual acuity outcomes

PURPOSE

To identify clinical characteristics that were associated with an increased incidence of endophthalmitis in eyes with penetrating ocular trauma.

METHODS

In part 1, a retrospective analysis was performed on 258 consecutive patients with penetrating ocular trauma presenting to the Bascom Palmer Eye Institute between October 1987 and January 1991. In part 2 of the study, 28 consecutive patients with culture-proven endophthalmitis were identified from the Clinical Microbiology Registry from April 1987 through September 1987 and February 1991 through August 1993. Clinical variables were evaluated in each part for association with an increased risk of endophthalmitis.

RESULTS

In part 1 of the study, endophthalmitis developed in 13 (5%) of the 258 patients. Endophthalmitis did not occur in eyes that had blunt injury. In those eyes with a lacerating injury, there was an increased relative risk of infection in eyes with disruption of the crystalline lens. This risk factor was found statistically significant by univariate and multivariate analysis. In part 2 of the study, lens disruption was present in 24 (86%) of 28 patients with culture-proven endophthalmitis. Of the 41 patients with infection from part I and part II, 22 (54%) achieved visual acuity of 20/ 400 or greater. Endophthalmitis caused by coagulase-negative staphylococci had the best visual outcome, with 7 (64%) of 11 patients obtaining visual acuity of 20/ 400 or greater.

CONCLUSION

Lens disruption in eyes with penetrating trauma is a significant risk factor for the development of endophthalmitis. The prognosis for useful vision in eyes with posttraumatic endophthalmitis is best when infection is caused by less virulent organisms.

Paediatric post-traumatic endophthalmitis

AIMS

A retrospective analysis of children with post-traumatic endophthalmitis was performed to determine if microbiological differences exist between this disease in the paediatric population compared with this disease in adults.

METHOD

Twelve cases of post-traumatic endophthalmitis in children were analysed to determine characteristics of this disease in youth. Patient ages varied from 18 months to 13 years; the mean age was 8 years. Gram positive organisms were isolated in eight eyes, Gram negative organisms from four eyes, fungus from one eye, and negative cultures in three eyes. The most common isolates were streptococcal species (56.6%) and staphylococcal species (22.2%). Vitrectomy was performed on eight (66.7%) eyes.

RESULTS

Visual acuity of 20/200 or better was obtained in eight eyes (66.7%). Three eyes had vision less than 5/200. One eye developed phthisis bulbi. Nine (75%) patients were younger than 10 years of age, and six (66.7%) of these nine obtained a final visual acuity of 20/200 or better.

CONCLUSION

Useful vision can be obtained in children with post-traumatic endophthalmitis with early, aggressive treatment. The microbiology of paediatric post-traumatic endophthalmitis differs from adult disease with streptococcal species as the most common infecting organisms.

Microorganisms cultured from the anterior chamber of ruptured globes at the time of repair

PURPOSE

We studied events leading to the development of posttraumatic endophthalmitis by examining the significance of 15 factors on microbial contamination of injured eyes.

METHODS

A prospective study was done of 30 ruptured globes in patients admitted to an urban medical center. Cultures were taken from the conjunctiva before and after preoperative disinfection and from the anterior chamber at the beginning and end of wound repair. Twenty-five of 30 patients received a three-day regimen of intravenous antibiotics that were begun before surgery.

RESULTS

Anterior chamber samples grew microorganisms in ten (33%) of 30 eyes, with positive cultures recovered from specimens taken at the beginning of wound repair in eight eyes and at the end of wound repair in six eyes. Contamination with indigenous flora may have occurred at the time of injury in one eye and during repair in another eye. Microbes recovered included Staphylococcus, Corynebacterium, and Aspergillus species. No patient developed endophthalmitis. Of the 15 factors studied, only intravenous antibiotics significantly decreased the incidence of positive anterior chamber cultures in eyes treated before wound repair compared with eyes not receiving such therapy (P = .002).

CONCLUSIONS

Despite the frequency of anterior chamber microbial contamination during injury or repair of the wound, with our treatment protocol and the presence of physiologic mechanisms to reduce intraocular microbes, no eyes developed clinical endophthalmitis. With our limited sample size only intravenous antibiotic therapy was found significantly to reduce anterior chamber microorganisms at the time of surgical repair, supporting their prophylactic use against the development of posttraumatic endophthalmitis.

Primary intraocular lens implantation in the setting of penetrating ocular trauma

PURPOSE

To evaluate the clinical outcome of patients who underwent lensectomy and intraocular lens (IOL) implantation at the time of primary repair of a penetrating ocular injury.

METHODS

A review of 14 patients who sustained cataracts and lens rupture in the setting of a corneal laceration to determine anatomic and visual outcome, in addition to complications related to the primary IOL.

RESULTS

The IOL remained anatomically stable in all 14 patients with no complications encountered at implantation or after surgery. Final visual acuity in 9 of the 14 patients was 20/40 or better. Six patients underwent pars plana vitrectomy for removal of an intraocular foreign body.

CONCLUSION

Intraocular lens implantation at the time of lensectomy and primary repair of a corneal laceration allows good visual rehabilitation with restoration of binocular function and serves as an alternative to contact lens correction in select patients.

Concurrent endophthalmitis and retinal detachment

PURPOSE

Eyes with concurrent endophthalmitis and retinal detachment usually have a poor anatomic and visual outcome after treatment. The purpose of this study is to define the relation among the causative organism, the results of retinal detachment repair, and the final visual acuity.

METHODS

Data were retrieved by a retrospective, computer-assisted review of the coded inpatient diagnoses from April 1987 through March 1992.

RESULTS

This study included 16 patients (9 males, 7 females) ranging in age from 5 to 88 years (average, 58.7 years). Endophthalmitis was classified as exogenous in 13 (81%) patients and endogenous in 3 (19%). Two groups were identified: a virulent group that included eight (50%) patients (Staphylococcus aureus, streptococci, gram-negative, Bacillus), and a less-virulent group that included eight (50%) patients (Staphylococcus epidermidis, Propionibacterium acnes, fungal). The initial surgical procedure consisted of diagnostic vitreous fluid collection by pars plana vitrectomy (11 cases), anterior vitrectomy (1 case), or vitreous aspiration (4 cases). Additional initial adjunctive surgical procedures included pars plana lensectomy (2 cases), scleral buckling (6 cases), fluid-gas exchange (7 cases), and intraocular antibiotic injections (9 cases). In six (75%) of eight patients with endophthalmitis in the virulent group, the retina remained detached. However, in seven (88%) of eight patients with endophthalmitis in the less-virulent group, the retina remained reattached postoperatively, and the remaining patient had a stable, nonprogressive peripheral tractional retinal detachment. None of the eight patients with endophthalmitis in the virulent group retained a postoperative visual acuity of better than 3/200, and four (50%) lost all light perception. Five (62%) of the eight patients with endophthalmitis in the less-virulent group retained a postoperative visual acuity of 5/200 or better, and none lost all light perception.

CONCLUSIONS

Concurrent endophthalmitis and retinal detachment patients with virulent organisms have a poor prognosis. Visual and anatomic outcomes were better in the less-virulent group.

Infectious endophthalmitis after penetrating injuries with retained intraocular foreign bodies. National Eye Trauma System

PURPOSE

To determine the risk factors and prognostic indicators of infectious endophthalmitis in eyes with penetrating injury and retained intraocular foreign body.

METHODS

From the National Eye Trauma System (NETS) Registry, 492 eyes with intraocular foreign bodies were reviewed for signs of infectious endophthalmitis.

RESULTS

Thirty-four eyes (6.9%) with intraocular foreign bodies had evidence of infectious endophthalmitis, and 31 (91.2%) of those eyes had signs of infection at the time of removal of the intraocular foreign body. The majority of eyes with an intraocular foreign body with or without endophthalmitis were in patients between 10 and 39 years of age, but the risk of endophthalmitis developing increased with age, especially in patients 50 years of age or older with delayed primary repair (P = 0.005). Endophthalmitis was more likely to develop in eyes with home or occupational injuries (33/358, 9.2%) than in those with injuries from other settings (1/128, 0.8%; P = 0.001). Infectious endophthalmitis was much less likely to develop in eyes with primary repair within 24 hours of the injury (10/287 = 3.5%) than in eyes with primary repair more than 24 hours after the injury (22/164, 13.4%; P < 0.0001). Bacilli or staphylococci were isolated in 21 (95%) of 22 eyes with positive cultures. Visual prognosis was reasonably good with 15 (58%) of 26 eyes attaining a visual acuity of 20/200 or better.

CONCLUSIONS

Removal of a retained intraocular foreign body within 24 hours of injury markedly reduces the risk of infectious endophthalmitis developing. Older persons are at high risk for endophthalmitis developing after retaining an intraocular foreign body when there is delayed surgical repair.

Intraocular foreign body missed by radiography

Images

Criteria for intravitreal antibiotics during surgical removal of intraocular foreign bodies

Published opinion supports the early use of prophylactic antibiotics soon after presentation of a potentially contaminated intraocular foreign body (IOFB) in the posterior segment, preferably within 12 hours of trauma when the visual acuity is still good. Recommended treatment includes topical, subconjunctival, parenteral and especially intravitreal antibiotic therapy to reduce the chance of endophthalmitis. Intravitreal therapy should include gentamicin with vancomycin or clindamycin to cover Bacillus spp., since this organism is responsible for half the endophthalmitis cases and produces beta-lactamase giving resistance to penicillins and cephalosporins. Intravitreal therapy is important because systemic and topical antibiotics do not penetrate the globe in sufficient concentration to control a fulminant infection associated with damaged tissue. Early therapy is essential, as delaying treatment until endophthalmitis occurs is less likely to save useful vision. All vitreous aspirated, plus pus if present, and the IOFB should be cultured for bacteria and fungi to identify pathogens and gain antibiotic sensitivities. The prevention of IOFB-associated endophthalmitis requires early recognition of the IOFB and enthusiastic antibiotic therapy at the time of its removal.