Secondary Logo

Journal Logo

Original article

Development of sympathetic ophthalmia following globe injury

ZHANG, Ying; ZHANG, Mao-nian; JIANG, Cai-hui; YAO, Yi

Author Information
doi: 10.3760/cma.j.issn.0366-6999.2009.24.008
  • Free


Sympathetic ophthalmia (SO) is a special bilateral inflammation with unkown etilolgy. Accidental or surgical trauma to uvea inciting an autoimmune inflammatory response was one of the major presumed reasons for SO.1-3 In clinical practice, its importance lies in that injury to one eye (exciting eye) possibly leads to impairment of visual acuity (VA) or even blindness in both eyes. The innocently affected fellow eye is called sympathizing eye.

Moreover, due to its rare occurrence and lack of enough eye injury cases, the incidence and risk factors of SO after open globe injury have been difficult to be established. And the previous related studies in most merely focused on penetrating or postoperative cases. Regarding effect on development of SO, comparisons among different types of globe injury and between injured eyes with and without surgery were seldom made.

The purposes of this study were, first, by reviewing all cases of globe injury, to establish the demographic, occurrence rate and clinical profile of SO following all types of globe injury; second, to attempt to identify risk factors (injury types, intraocular surgery and endophthalmitis etc) for the development of SO after open globe injury; and, third, to describe the treatments and outcomes.



The medical records of all patients with ocular trauma admitted to 15 tertiary referral hospitals in China between January 1, 2001 and December 31, 2005 were retrospectively reviewed. Only the cases with trauma of eyeball were collected and the ones with simple lesions of eye appendix were excluded. Inpatient admissions were coded using International Classification of Diseases 10 (ICD10). Sufficient information was presented in all the medical files. Each case was recorded in a standardized data sheet pre-formulated and the computerized database of eye injury for statistic analysis. The SO cases were identified through a search of discharge diagnoses of admitted patients. The diagnosis of SO was confirmed by the development of bilateral or contralateral uveitis consistent with clinical features of SO and supported by the definit history of ocular trauma. Histopathological evidence of SO should be verified in enucleated or eviserated eyes. And any case with systemic autoimmune disease was screened out to exclude other oculopathy with similar clinical features. The records of patients with similar discharge diagnoses (e.g., endophthalmitis, uveitis) were also reviewed in detail to ensure that no cases of SO due to trauma were missed.


In this study, classification and definition of ocular trauma were based on the Birmingham Eye Trauma Terminology (BETT). An open globe injury was defined as traumatic full-thickness wound of eyewall, which included subtypes of rupture, penetration, perforation and intraocular foreign body (IOFB). A standardized data sheet was completed for each case of ocular trauma. Information, such as patient age, gender, occupation, medical and ophthalmic history, and previous vision, if known, when, where and how injury occured, clinical presentation, timing and way of management as well as outcomes at discharge, were collected in detail.

Data collection

In those patients who diagnosed with SO, the following information was gathered particularly: the timing of primary repair and secondary intraocular surgeries, total number of intraocular surgeries, the interval between the inciting injury or the last operation and the onset of SO symptoms, route and nature of therapeutic medicine administration, use of intravitreal steroid, final vision after treatment and enucleation or evisceration at discharge.

Data analysis

All data were collected in an electronic database and cross-checked for errors. Statistical analysis was performed using SPSS 17.0 (SPSS Inc, Chicago, USA). Categorical variables were analyzed using the chi-square test. Continuous variables were examined for normality, and means were compared using t test. Differences were considered to be significant at P <0.05.


Patient background data

A total of 9776 eyes of 9103 inpatients (673 bilateral cases) with globe injuries were inclusive. Among those, 4968 eyes of 4843 patients (125 bilateral cases) were diagnosed with open globe injures and 172 eyes were enucleated or eviscerated primarily within 24 hours after trauma.

Twenty-two cases were diagnosed with SO in the 9103 patients of globe injuries. Eighteen cases of SO occurred following 4843 patients of open globe injuries and the incidence of SO after open globe injury was 0.37%. Two cases of SO developed in closed globe injury (contusion) after intraocular surgery (vitrectomy). There were 537 eyes underwent vitrectomy after closed globe injury, in which the incidence of SO was 0.37% after vitrectomy. And another two occurred individually after corneal perforation of heat burn and alkali burn. The main findings in exciting eyes and corresponding sympathizing eyes were shown in Table 1.

Table 1
Table 1:
Summary of demographic data and clinical features of 22 sympathetic ophthalmia patients with globe injury

All 18 patients with SO following open globe injury (cases 1-18; Table 1) were Chinese, 15 male and 3 female. And there was no significant difference between sexes (0.35% vs 0.44%, P=0.73). The 18 patients were aged 11-57 years at the time of inciting event and the median age was (36.72 ± 13.59) years, which was higher than the median age of other patients without SO after open globe injury ((27.95 ± 15.17) years; t=2.45, P=0.01). The 15 male patients, aged 19-57 years, were generally older than their female counterparts.

As for open globe injuries, there was no significant effect of injury type or uvea proplaps on SO incidence (Table 2). Four traumatized eyes (cases 5, 6, 8 and 9; Table 1) developed endophthalmitis and subsequently SO which accounted for 0.70% of 571 endophthalmitis eyes following open globe injury. However, endophthalmitis here was not associated with the incidence of SO (Table 2).

Table 2
Table 2:
The effect of mechanism of injury, uvea incarceration, multiple intraocular surgeries, vitrectomy and endophthalmitis of open globe injury eye, on sympathetic ophthalmitis incidence

The intervals between ocular trauma and the onset of SO were variable. As for the 18 cases of open globe injury, it ranged from 26 days to 10 years, while 55.56% occurred within 2 months, 72.22% within half of a year and 83.33% within 1 year after injury. The case 15 was admitted to hospital for rupture ten years ago and returned again for SO later. And the case 16 suffered twice open globe injuries: first 10 years before and then rupture of the atrophy globe 2 months before following which SO occurred. Furthermore, the interval between the last intraocular surgery and the onset of SO was as short as 1 day and long as 10 years, 16.67% occurred within 2 weeks, 38.89% within 1 month, 61.11% within 2 months and 83.33% within 1 year after the last ocular surgery. For the case 21 (Table 1), SO occurred 1 week after evisceration of the perforating burn eye.


The 18 eyeballs of open globe injury inciting SO later had all undergone ocular surgery after trauma. Of these, 16 cases (88.9%) underwent primary repair after trauma, 8 cases (44.4%) received intraocular operation of trans-pars plana vitrectomy among which 2 cases twice for vitrectomy and 7 cases (38.9%) had ocular surgery twice at least. One case underwent extracapsular cataract extraction prior to SO when first injured. None of once or multiple (≥2) intraocular surgeries was associated with SO incidence (Table 2).

There was no eyeball prophylactically enucleated within 2 weeks after injury. Topical and systemic steroids therapies were administered on all cases and periocular steroids on 19 cases. In addition, 4 cases received immunosuppressive treatment simultaneously, otherwise all of which were enucleated or eviscerated at last and verified by histopathological diagnosis. In 7 (38.89%) of the 18 cases of open globe injury, the inciting eyes were removed eventually after surgery and medicine treatments because of relapsing SO in the sympathizing eyes. The interval between the onset of SO and enucleation of inciting eyes ranged from 1 day to 6 months. And the interval between injury and enucleation ranged from 2 months to 6 years, among which 5 (71.43%) within half a year.

As shown in Table 3, the overall follow-up time in our series ranged between 2 months and 2 years. VA of exciting eyes were worse (range from no light perception (NLP) to 0.1) before onset of symptoms and recovered limited (range, from NLP to 0.4) after treatments. The extent of decline of VA in sympathizing eyes was moderate comparatively and outcome was satisfactory by medicine treatments.

Table 3
Table 3:
Summary of visual acuity (VA) conditions of 22 sympathetic ophthalmia patients with globe injury



The incidence of SO after open globe injury in our study is higher than 0.2%4 and lower than 1.65% reported by a study based on data of inpatient.5 The incidence of SO after penetrating ocular injuries is keeping with 0.14%-0.5%.6-10 Some recent reports indicate that more aggressive surgical management, such as vitrectomy, of severely traumatized eyes may be a greater risk for SO especially in developed countries.11-14 This can be reflected in our study, the two cases of SO after closed globe injury were both undergone vitrectomy. Even though, whether vitrectomy comparing with other ocular surgeries is a more dangerous contributor to SO or not needs to be confirmed. However, people traditionally think that any type of intraocular surgery is a risk factor for SO development.15 It was shown the incidence of SO after intraocular surgery varying from 0.007% to 0.6%.7-9 In 1982, Gass reported 0.01% incidence of sympathetic ophthalmia following pars plana vitrectomy, and a 0.06% incidence when there was a history of another penetrating wound.16 We could not get data of SO following vitrectomy, but our outcome of SO incidence after vitrectomy with penetrating background (0.3% in Table 3) is obviously higher than 0.06%. Different severity of injuries of sample perhaps leads to this discrepancy.

In this series, we found that in terms of SO incidence after open globe injury, there was no significant difference between sexes and the elder were predominant. The viewpoints were divergent. About sexes, Kilmartin et al11 and Albert et al17 have the same opinion in a prospective surveillance for SO, while some other authors presume SO occurs more frequently in men.18,19 Some found equal incidence in any age, but others reported more in young people or in the elder.11,17 Probably the age group of accidental trauma or ocular surgery is the different point.

Affecting factors

The cases 1-18 (Table 1) of open globe injuries, except the case 10 merely with ECCE, all underwent surgical repair of wounds or even more than one procedure to treat other complications. Then, complex is that we should attribute SO to accidental trauma or to surgery. A report from Singapore discussed that the initial injury may have sensitized the immune system to uveal antigens, but it was the subsequent surgical manipulation that released sufficient uveal protein to trigger the development of SO.14 However, they subsequently questioned this assumption. We think both are critical, but initial trauma is the essential cause. This is supported by our finding that intraocular surgery was not a risk factor for SO development of eyes with open globe injury. Probably, in this instance, surgery is a following factor of promotion and primary trauma is the initial trigger. However, this does not mean that surgery is not dangerous to non-injured eye because some surgeries, especially vitrectomy, may accidentally become a kind of injury to intraocular tissues particularly when occurring operating delinquency or maneuver not gentle enough. Here, for eyes with open globe injury, trauma occurred prior to surgery in chronologic order and was more powerful than surgery in extent of destroying.

As reported, penetration, IOFB and perforation are risk factors for SO development.17,20 We found no statistically difference among each type of open globe injury. The destructive power of injury may be the truth. And the evaluation on dangerous levels of injury for inciting SO needs further investigation. Uvea proplaps is not a risk factor here, which doesn't mean there's no difference in damage extent or time of proplaps of uvea.

It has been reported that any type of intraocular surgery is risk for SO15 and the incidence of SO increases with multiple ocular surgeries. But we found there's no significance between once and multiple intraocular surgery including vitrectomy which induced SO. We supposed that development of SO depend on susceptivity of eye tissues and efficacy of irritating intraocular immuno-response after surgery or injury. Multiple ocular surgeries merely increase the chance of triggering SO. Based on this, maybe we can explain why the other cases undergone multiple ocular surgeries, but did not develop SO.

In our study, there was coexisting SO in four cases and posttraumatic endophthalmitis after open globe injuries (Table 3). Our analysis also showed that endophthalmitis was not the necessary factor for SO. Many authors agree with this opinion.12,17,21,22 But the incidence of coexistent endophthalmitis is variable. It can be as high as 3%-11%.10,21,22 While Lubin et al,18 in a large clinicopathologic study of SO, found that it was less than 1% which is close to our data.

The interval between ocular injury and the onset of SO reported can vary from 5 days to 60 years,17,23,24 but 65% occur between 2 weeks and 2 months, 80% within 3 moths and 90% within 1 year after injury.25 Our data are within the range.


The controversial part of treatment is prophylactic enucleation of the injured eyes within 2 weeks of the traumatism before its sensitization.9 We do not recommend this impetuous action as it does not absolutely prevent SO.26 The case 21 in our study seemed to support this, because SO occurred one week after evisceration of a perforating burn eye. This is why we did not exclude the eyes directly removed after trauma. Perhaps someone would argue that evisceration is not safer than enucleation.27 But Ruedemann et al28 found no case of SO in 200 cases of evisceration during the long period follow-up of 18 years. On the other hand, SO may not develop whether undergoing enucleation or not. At the same time, enucleation of exciting eye after SO development is still subject to controversy about whether it could improve visual prognosis as well as lead to fewer and milder relapses in the sympathizing eye.9,18,22,28,30 As shown in Table 2, the visual prognosis of sympathizing eyes is not totally related with enucleation of traumatic eyes. By prompt treatment the conditions of SO could be controlled and the recovery of VA in sympathizing eyes was satisfactory. The relatively worse final VA of traumatized eyes seems due to severity of original injury. So we do not recomment enucleation as either prophlatic or remedial measure.


1. Damico FM, Kiss S, Young LH. Sympathetic ophthalmia. Semin Ophthalmol 2005; 20: 191-197.
2. Chu DS, Foster CS. Sympathetic ophthalmia. Int Ophthalmol Clin 2002; 42: 179-185.
3. Chan CC, Mochizuki M. Sympathetic ophthalmia: an autoimmune ocular inflammatory disease. Springer Semin Immunopathol 1999; 21: 125-134.
4. Robert AM. Sympathetic ophthalmia. In: Ferenc Kuhn, Dante JP, ed. Ocular trauma-principles and practice. New York, Stuttgart: Thieme Medical Publishers Inc; 2002: 301.
5. Luo Y, Wang Z, Lin X, Hu S. Sympathetic ophthalmia caused by ocular penetration with endophthalmitis. Eye Sci (Chin) 2003; 19: 75-78.
6. du Toit N, Motala MI, Richards J, Murray AD, Maitra S. The risk of sympathetic ophthalmia following evisceration for penetrating eye injuries at Groote Schuur Hospital. Br J Ophthalmol 2008; 92: 61-63.
7. Gürdal C, Erdener U, Irkeç M, Orhan M. Incidence of sympathetic ophthalmia after penetrating eye injury and choice of treatment. Ocul Immunol Inflamm 2002; 10: 223-227.
8. Liddy L, Stuart J. Sympathetic ophthalmia in Canada. Can J Ophthalmol 1972; 7: 157-159.
9. Makley TA Jr, A Azar. Sympathetic ophthalmia: a long-term follow-up. Arch Ophthalmol 1978; 96: 257-262.
10. Marak GE Jr. Recent advances in sympathetic ophthalmia. Surv Ophthalmol 1979; 24: 141-156.
11. Kilmartin DJ, Dick AD, Forrester JV. Prospective surveillance of sympathetic ophthalmia in the UK and Republic of Ireland. Br J Ophthalmol 2000; 84: 259-263.
12. Chan CC, Roberge RG, Whitcup SM, Nussenblatt RB. 32 cases of sympathetic ophthalmia: a retrospective study at the National Eye Institute, Bethesda, Md, from 1982 to 1992. Arch Ophthalmol 1995; 113: 597-600.
13. Kilmartin DJ, Dick AD, Forrester JV. Sympathetic ophthalmia risk following vitrectomy: should we counsel patients? Br J Ophthalmol 2000; 84: 448-449.
14. Su DH, Chee SP. Sympathetic ophthalmia in Singapore: new trends in an old disease. Graefes Arch Clin Exp Ophthalmol 2006; 244: 243-247.
15. Lyons C, Tuft S, Lightman S. Sympathetic ophthalmia from inadvertent ocular perforation during conventional retinal detachment surgery. Br J Ophthalmol 1997; 81: 612.
16. Gass JD. Sympathetic ophthalmia following vitrectomy. Am J Ophthalmol 1982; 93: 552-558.
17. Albert DM, Diaz-Rohena R. A historical review of sympathetic ophthalmia and its epidemiology. Surv Ophthalmol 1989; 34: 1-14.
18. Lubin JR, Albert DM, Weinstein M. Sixty-five years of sympathetic ophthalmia: a clinicopathologic review of 105 cases (1913-1978). Ophthalmology 1980; 87: 109-121.
19. Ganesh SK, Sundaram PM, Biswas J, Babu K. Cataract surgery in sympathetic ophthalmia. J Cataract Refract Surg 2004; 30: 2371-2376.
20. Power WJ, Foster CS. Update on sympathetic ophthalmia. Int Ophthalmol Clin 1995; 35: 127-137.
21. Rathinam SR, Rao NA. Sympathetic ophthalmia following postoperative bacterial endophthalmitis: a clinicopathologic study. Am J Ophthalmol 2006; 141: 498-507.
22. Kuo PK, Lubin JR, Ni G, Wang KM, Albert DM. Sympathetic ophthalmia: a comparison of the histopathological features from a Chinese and American series. Int Ophthalmol Clin 1982; 22: 125-139.
23. Drews RC. Delayed onset of sympathetic ophthalmia. Ophthalmic Surg 1994; 25: 62-63.
24. Towler HM, Lightman S. Sympathetic ophthalmia. Int Ophthalmol Clin 1995; 35: 31-42.
25. Aude Danan-Husson, Joseph-Alain Sahel. Sympathetic Ophthalmia. In: Daniel M Albert, Joan W Miller, ed. Principles and Practice of Ophthalmology. Vol1 3rd ed. Philadelphia, US: Elservier Inc; 2008: 1222.
26. Bellan L. Sympathetic ophthalmia: a case report and review of the need of prophylactic enucleation. Can J Ophthalmol 1999; 34: 95-98.
27. Cytryn AS, Perman KI. Evisceration. In: Migliori ME, ed. Enucleation, evisceration and exenteration of the eye. Boston, MA: Butterworth-Heinemann; 1999: 105-112.
28. Ruedemann AD. Sympathetic ophthalmia after evisceration. Trans Am Ophthalmol Soc 1963; 61: 274-314.
29. Jennings T, Tessler HH. Twenty cases of sympathetic ophthalmia. Br J Ophthalmol 1989; 73: 140-145.
30. Reynard M, Riffenburgh RS, Maes EF. Effect of corticosteroid treatment and enucleation on the visual prognosis of sympathetic ophthalmia. Am J Ophthalmol 1983; 96: 290-294.

sympathetic ophthalmia; eye injury; intraocular sugery; risk factors

© 2009 Chinese Medical Association