Ocular Toxoplasmosis

Key concepts

  • Ocular disease can occur after both congenital and acquired disease.

  • Recurrent disease frequently is seen as a satellite lesion.

  • Immunodeficient patients are at risk for acquired disease and possibly reactivation of old disease.

Toxoplasmosis is a common disease in both mammals and birds. The disease is caused by the obligate intracellular protozoan Toxoplasma gondii . It is thought that this organism infects at least 500 million persons worldwide, and at least 50% of the adult population in the United States has the chronic symptomless form of the disease. A survey of ophthalmologists in the US reported that 55% of those who responded saw one or more active ocular toxoplasmosis cases in last 2 years, and that 93% of those who responded had seen inactive cases in the last 2 years. In the United Kingdom the estimated lifetime risk for ocular toxoplasmosis has been calculated to be 18 in 100 000. In the developing world, its prevalence is probably underestimated.

A prospective study in Sierra Leone identified toxoplasmosis as the most common cause of uveitis. In Nepal, over 50% of those coming to hospital, not only for uveitis, but for other disorders such as malignancies and obstetric problems, had antibodies to toxoplasma, with over 5% being IgM positive, indicative of a recent infection. The disease can cause a passing flu-like condition that has little consequence, but it can also cause lymphadenopathy, serious and sometimes fatal disease in immunocompromised hosts, spontaneous abortions, and congenital disease. For the ophthalmologist it is one of the most frequently encountered posterior uveitides, classically producing a necrotic retinitis. It is also one of the few uveitides for which we can potentially make a definitive diagnosis. In the past few years our understanding of the organism and its interrelationship with its host has brought into question several concepts that had been readily accepted in ophthalmology practice.


In 1908 T. gondii was first found in the brain of the North African rodent the gondi, by Nicolle and Manceaux and then by Splendore in a rabbit in Brazil. Janku first described postmortem findings in a child who had died of disseminated toxoplasmosis. He noted what were probably Toxoplasma organisms in the eye, but inoculation of animals with infected tissue did not induce disease. The transmission of the organism to animals via inoculation of infected human tissue was accomplished by Wolf and coworkers. Helenor Campbell Wilder identified the presence of the organism in the eye in 1952, confirming that it was the cause of uveitis.

T. gondii is a ‘cosmopolitan’ parasite, being found all over the world. Members of the cat family are the definitive hosts. Oocysts of Toxoplasma are 10–12 µm in length and oval in shape. They are found uniquely in the intestinal mucosa of cats. Once they are released, they can be spread to humans or to other animals through a variety of vectors. Although invariably thought to be ingested, the organism may also enter the host through other mucosal surfaces. Humans can also be infected secondarily by ingesting meat (pork and lamb particularly, as well as chicken in endemic areas, but probably not beef) contaminated with Toxoplasma cysts. The two forms of the organism that can be found in humans are cysts and tachyzoites ( Fig. 14-1 ). The cysts are up to 200 µm in diameter, contain hundreds to thousands of organisms, and have a propensity for cardiac tissue, muscle, and neural tissue, including the retina. The cyst structure is complex and can include elements from the host. Cysts can remain intact outside of a host in soil for at least 1 year. Not all the factors that cause ultimate rupture of the cyst and the release of tachyzoites are totally clear. The tachyzoite is oval or arc shaped and about 6–7 mm in length. It is an obligate intracellular organism that actively proliferates and is the cause of the acute disease. The organism’s entry into and residence within the host cell are clearly complex and dynamic events, and much is still not known. Joiner found that the organism forms a parasitophorous vacuole that surrounds the parasite and that lacks plasma membrane markers from the host. It will not fuse with other compartments in the cell, and is sheltered from all cellular traffic.

Figure 14-1.

A, Tachyzoites in culture stained with immunofluorescent dye. B, Toxoplasma cysts containing large numbers of tachyzoites.

(Courtesy of Leon Jacobs, MD.)

Many of the antigens of the organism have been identified ( Table 14-1 and Fig. 14-2 ). Perhaps the most studied is SAG 1 or p30. This major surface antigen has a molecular mass between 27 and 30 kDa. It is useful in the serologic diagnosis of infection and may play a role in the parasite’s ability to invade a cell. In animal models, immunization with this antigen or adoptive transfer of immune cells recognizing this antigen will confer a degree of protection against active infection. The p30 gene sequence has been deduced and its mRNA appears to be 1500 nucleotides in length.

Table 14-1

Toxoplasma gondii antigens

Bradyzoite Surface antigen (SAG) 2C,2D, and 4
Bradyzoite specific recombinant (BSR) 4
Matrix antigen (MAG) 1
Lactate dehydrogenase (LDH) 2
Enolase (ENO) 1
Bradyzoite antigens (BAG) 1
Phosphatidylinositol (Ptdins) b
Tachyzoite SAG 1 (p30)
SAG 2A and 2B (p22)
Ptdins t
SAG-related sequences (SRS) 1–3

Figure 14-2.

Cartoon showing anatomic areas from which purified Toxoplasma antigens have been isolated.

A second antigen that has been characterized is SAG 2 or p22. This cell surface antigen (molecular mass 22 kDa) can participate in antibody-dependent, complement-mediated lysis of the tachyzoite. It appears to be part of a complex phagosomal reticular network. A third antigen that has been studied is known as the F3G3 antigen. This 58-kDa antigen is cytoplasmic and not expressed on the cell surface. Passive transfer of antibody that reacts to this antigen has been successful in protecting animals from a lethal challenge by the Toxoplasma organism. The study of excreted/secreted antigens of toxoplasmosis continues because it has been demonstrated that 90% of the circulating antigens detected during active infection are those that are actively excreted. These antigens could be used as a basis for vaccine development, in that immunization against these antigens might abrogate rapid entry of the tachyzoite into the cell. Of interest have been attempts to classify the specific clonal lineages that may cause human toxoplasmosis. Howe and Sibley determined the population genetic structure of T. gondii by using multiple restriction fragment length polymorphism analysis. They studied six loci in 106 independent Toxoplasma isolates from humans and animals. Although not separate strains, three distinct lineages seemed to be found, with only four of the isolates showing an extensively mixed genotype. In this study human isolates were found in all three lineages, although the majority of those had a type III genotype. However, one study performed in Europe reported that the cases evaluated were of the type I genotype. This was also the type reported from Brazil. However, the story is most probably more complicated than that. Grigg and colleagues reported an abundance of atypical strains (i.e., lineages) that were associated with toxoplasmic disease. In the 12 samples they evaluated, three had typical type I lineage whereas five of 12 had recombinant genotypes typical of two lineages. Howe and colleagues genotyped 68 of 72 samples isolated from human disease using the p22 (SAG2) antigen. They found that the vast majority of these 68 isolates (81%) were classified as type II, whereas only 10% were type I and 9% were type III. Khan et al. evaluated other strains from various parts of Brazil and found that the genotypes were highly divergent compared to previous clonal lineages. They argued that limiting the genotyping to just the SAG 2 region will not fully recognize the diversity of this organism. This line of work is important in understanding the human disease and in devising ways to stop the strains that most commonly cause disease in humans.

Clinical manifestations


The acquired disease in the nonimmunocompromised adult leads to lymphadenopathy in 90% of patients, with fever, malaise, and sore throat sometimes occurring. More severe disease can occur, affecting muscle, skin, brain, heart, and kidney, as well as other organs. Death due to toxoplasmosis rarely occurs in the immunocompetent. However, in the immunocompromised patient toxoplasmosis can be a fulminant central nervous system (CNS) disease that rapidly leads to death.


The acquired disease in the immunocompetent person had been thought to cause ocular disease only rarely, but this appears not to be the case. In part, this concept stemmed from a report by Perkins, discussed in some detail below (see Congenital versus Acquired disease ). This is not to minimize the importance of pregnancy and the transmission of this parasite to the child. Pregnancy is the period during which acquisition of the disease causes the most concern. An older article suggests that in the United States up to six in 1000 women acquire the infection while pregnant, with about a 40% risk of transmitting the infection to the fetus. In a report describing the national neonatal screening program for congenital toxoplasmosis in Denmark, there were 2.1 congenital toxoplasmosis cases per 10 000 newborns. The congenital disease can lead to a wide range of symptoms, but most important to this discussion is that most cases of ocular toxoplasmosis are conjectured to be acquired congenitally, with late activation. In the Danish study 9.6% of those with congenital toxoplasmosis were born with retinal or macular lesions, with 15.6% manifesting these changes at 3 years of age. In a European study reported by Koppe and Kloosterman in 1982, 5% of the infants infected would either die or become severely affected by the disease. Further, about 70% of infants with congenital infections will show chorioretinal scars compatible with toxoplasmosis after a follow-up of 16 years, with 1–2% suffering severe visual impairment because of this infection. However, reports also support the notion that acquired infection may lead to ocular disease. In following up the results of an outbreak of systemic toxoplasmosis that occurred in Atlanta, Georgia, in October 1977, Wilson and Teutsch reported that one patient of the original group who became ill or had serologic evidence of acute infection showed evidence of ocular disease.

A large atrophic scar, frequently in the macula – a result of congenital toxoplasmosis – is usually seen ( Fig. 14-3 ). Although such lesions help in diagnosing an old problem, they do not present the ophthalmologist with a therapeutic problem. Rather it is the reactivation or the recent acquisition of toxoplasmosis that poses the problem. In these instances, ocular toxoplasmosis manifests as a focal retinitis. The active lesion can vary greatly in size, but is usually oval or circular and rarely bullous. Frequently, reactivation sites will be ‘satellite’ lesions next to old atrophic lesions, indicative of previous toxoplasmic infections. The retina during the acute stage of infection appears thickened rather than transparent and is cream-colored ( Fig. 14-4 ). Cells are found in the vitreous, particularly overlying the active lesion. Some eyes may have one or two small, old lesions, whereas others may have many, with some being very large and involving several clock hours of peripheral retina. In some large and particularly recalcitrant lesions the vitreal haze and cellular reaction can be so profound as to cause decreased vision. In the area surrounding the active retinitis one may see hemorrhage, as well as sheathing of the retinal blood vessels. Fluorescein angiography of the active lesion demonstrates early blockage with subsequent leakage of the lesion ( Fig. 14-5 ). Indocyanine green angiography will show hypofluorescence of both active and inactive lesions. ICG has shown areas of choroidal hypofluorescence that do not correspond to the defects seen on fluorescein angiography ( Fig. 14-6 ). Third-generation OCT has shown that the retinal layers are abnormally hyperflective at the active lesions, with thickening of the posterior hyaloids, which is often focally detached. When lesions are close to the optic nerve, there can be considerable field loss; one study reported that 94% of toxoplasmosis patients had visual field loss due to the disease. As mentioned, there is a real risk of disease recurrence, hence the finding of ‘satellite’ lesions. It has been suggested that Toxoplasma cysts would be found in greater numbers near the site of a previous infection, which would seem reasonable. Holland and coworkers reported that the risk of recurrence was greatest immediately after an episode, and Garweg et al. found that younger ocular toxoplasmosis patients have a greater chance of recurrence than do older patients. In the questionnaire responses they received, two-thirds of the patients had a repeat attack.

Figure 14-3.

Large macular atrophic scar due to congenital toxoplasmosis.

Figure 14-4.

Large active toxoplasmic infection in retina; seen as ‘satellite’ to old healed atrophic lesion.

Figure 14-5.

A, Fluorescein angiogram showing leakage around borders of macular lesion of toxoplasmosis. Note staining of retinal vein below lesion. B, Large toxoplasmosis lesion encroaching on optic disc. C and D, Fluorescein angiogram of lesion showing hyperfluorescence peripherally with later photograph showing leakage of dye from both lesion and disc.

(Courtesy of JDM Gass, MD.)

Figure 14-6.

The ICG angiogram in the early phase can show hypofluorescent areas that do not correlate with that seen on fluorescein angiography (arrowheads).

(Reproduced with permission from Khairallah, Moncef, Acute Choroidal Ischemia Associated with Toxoplasmic Retinochoroiditis, 27(7):2007. Copyright 2007 Wolters Kluwer.)

Because the Toxoplasma organism has a propensity for neural tissue, it is important to bear in mind that the lesion classically begins in the retina, and only with ongoing inflammation will it involve not only multiple layers of the retina but also the choroid. Cells in the anterior chamber may also be noted and may appear to be either a granulomatous or nongranulomatous uveitis. In the immunocompetent host, although evidence of old toxoplasmic activity may be present, the disease usually activates in only one eye at a time. With continuing inflammatory disease the lesion and overlying vitreous will undergo several changes. The vitreous may contract, and a posterior vitreal detachment is not uncommon. Further vitreal condensation leads to a ‘scaffolding’ of vitreal strands ( Fig. 14-7 ). Roizenblatt and coauthors refer to the development of vitreous cylinders in toxoplasmosis, a result of condensation of the collagen fibers. As the lesion becomes less acute, the area of retinal involvement takes on a less bright-yellow appearance, ultimately becoming atrophic, often with pigment heaping around its edges. Pigment clumping, however, does not surround all old lesions and should not be used as a diagnostic sign. The associated retinal disturbance will also begin to resolve. All patients will have visual field defects that correspond to the interruption of the retinal nerve fiber layer. Although unusual, an unilateral toxoplasmic anterior optic neuropathy, presenting with sudden painless loss of vision, has been reported.

Figure 14-7.

A, ‘Scaffolding’ seen in vitreous of patient with severe recurrent ocular toxoplasmosis. B, Keratic precipitates on cornea of same eye during inflammatory episode.

Decreased Vision

The vision of the patient with toxoplasmosis is decreased for several reasons. As already mentioned, the vitreal inflammation by itself may be so great as to significantly reduce vision. Also, the lesion may be situated in the posterior pole (although not in the fovea per se), with edema and probable inflammatory byproducts of the reactions affecting central vision; however, the fovea may be relatively far from the center of the retinitis. This may be similar to the phenomenon seen in pars planitis. Another cause of decreased vision is infection involving the macula. Once the fovea is involved in the actual infection itself, the potential for a significant return of good vision is poor, and all efforts must be made to prevent this from occurring. Schlaegel and Weber noted that in 60 attacks of ocular toxoplasmosis seven patients (12%) had active retinitis within 5° of the umbo, and another seven (12%) demonstrated some evidence of mild macular edema. The observer should not forget that ocular toxoplasmosis patients often have elevated intraocular pressures. These need to be monitored, as clearly a loss of vision could be due to glaucoma.

Ocular toxoplasmosis has been noted to present in a variety of ways besides the classic manifestation already described. Friedmann and Knox and Doft and Gass observed a subset of ocular toxoplasmosis that was characterized by gray-white fine punctate lesions of the deep retina and retinal pigment epithelium and, initially, with little or no overlying vitreal activity ( Fig. 14-8 ). The resolution of these lesions may leave a typical toxoplasmosis scar. Another important presentation is papillitis. These patients may have severe papillitis even with a central retinal artery occlusion, white centered retinal hemorrhages, vitreal inflammation, and sector nerve, fiber-bundle defects, with no apparent retinal foci, as described by Folk and Lobes. Other presentations that we are aware of are bullous-like inflammatory lesions in the midperiphery of the retina ( Fig. 14-9 ), a wide ringlike lesion near the extreme periphery of the retina that resembles a severe uniocular pars planitis, and a scleritis due to the severe retinal inflammation caused by the Toxoplasma infection. The acute inflammation may be so severe that choroidal ischemia manifesting as retinal whitening ( Fig. 14-10 ) Numerous anecdotal reports concerning various presentations can be found in the literature. Although some may be without foundation, these reports emphasize the fact that one should suspect this disorder in a variety of clinical situations. Toxoplasmosis lesions can on occasion be confused with other infectious disorders (see Case 14.4 ). In a patient who received an allogeneic bone marrow transplant for leukemia, ocular toxoplasmosis presented as a bilateral CMV retinitis with patient also found to have CNS lesions as well (see Fig. 14-5 ). Friedmann and Knox noted that the 63 patients whose cases they reviewed averaged 2.7 episodes each.

Figure 14-8.

Punctate inner retinal layer toxoplasmosis. A, Lesion begins as subtle alteration in normal macular pattern. B, In a follow-up photograph lesion takes on appearance of more classic type of toxoplasmic infection.

(Courtesy of J.D.M. Gass, MD.)

Figure 14-9.

A, Large bullous lesion due to toxoplasmosis, with titer of 1 : 125. B, Lesion involuted only after anti- Toxoplasma medication.

Figure 14-10.

Toxoplasmosis lesion appearing as cytomegalovirus-like lesion which cleared with anti-toxoplasmosis medication.

(Reproduced with permission from Bilateral Toxoplasma Retinochoroiditis Simulating Cytomegalovirus Retinitis in an Allogeneic Bone Marrow Transplant Patient, Hyewon Chung, MD1, June-Gone Kim, MD1, Sang-Ho Choi, MD2, Sun Young Lee, MD1, Young Hee Yoon, MD1 Korean Journal of Ophthalmology 2008;22:197–200.)

The involvement of the retinal vasculature needs to be emphasized. In one review of 64 patients 59 had vascular changes in the quadrant of the active lesion. Five eyes had changes in all quadrants. Three of these patients also had retinal vascular occlusions. Frosted branch angiitis has been reported in ocular toxoplasmosis eyes even years after the acute event.

In patients with ocular toxoplasmosis who are undergoing eye surgery there is a risk of reactivation of the disease. Bosch-Driessen and colleagues reported that five of 15 eyes in patients with toxoplasmosis undergoing cataract extraction had a reactivation. They suggested that prophylactic anti- Toxoplasma therapy given perioperatively may be warranted.

Loss of Vision

The most important sequela of ocular toxoplasmosis is loss of vision due to direct involvement of the fovea by the infection. However, choroidal neovascularization can also occur as a late complication of the disease. , Skorska and coworkers found that subretinal neovascular lesions were present in seven of 36 patients studied. The new vessels were located either directly on the border of the scar or at a distance, with feeder vessels arising from the scar ( Fig. 14-11 ). Retinochoroidal anastomoses have also been observed to occur in toxoplasmosis, with one study reporting an incidence of 2.7% ( Fig. 14-12 ). Other vascular complications have been reported. Rose reported on a patient with ocular toxoplasmosis with a retinal vein occlusion, papillitis, and florid disc neovascularization. Pakalin and Arnaud observed an occurrence that manifested with an arteriolar branch occlusion at a site passing through an area of necrosis. They emphasized the need to include toxoplasmosis in the differential diagnosis of even arterial inflammatory phenomena, something not usually done. Bosch-Driessen and associates reported that nine of 150 patients (6%) with ocular toxoplasmosis had retinal detachments, and another seven (5%) had retinal breaks. Retinal detachment was seen more commonly in myopes, and the visual prognosis is guarded: five of nine eyes with retinal detachment were left with 20/200 visual acuity or worse. As with any disorder that disrupts Bruch’s membrane, choroidal neovascularization, usually at the rim of an old lesion, needs to be looked for. One report describes multiple CNV lesions in one eye with punctate outer retinal Toxoplasma lesions.

Figure 14-11.

Subretinal neovascularization 8 months after acute attack of ocular toxoplasmosis. A, Nasal portion of lesion that stains corresponds to area of old scarred lesion. B, In the late-stage angiogram nasal hyperfluorescence has not increased in size; however, temporal subretinal neovascular lesion can be seen.

(Courtesy of Prof. G. Coscas. Reproduced with permission from Journal Francais d’Ophtalmologie 7: 211–218, 1984 © Masson.)

Figure 14-12.

Fundus photograph of eye with congenital toxoplasmosis, with retinochoroidal shunt.

(Courtesy of J. Melamed, MD, Porto Alegre, Brazil.)

A curious association between Fuchs’ heterochromia and ocular toxoplasmosis was initially made by Toledo de Abreu and coworkers. They studied 13 patients with Fuchs’ syndrome who had focal necrotizing chorioretinal toxoplasmic lesions; none of these patients had ciliary injection or posterior synechiae but most had keratitic precipitates, anterior chamber reactions, and cataracts. Six of the 13 had iris transillumination. La Hey and colleagues evaluated this possible association in a series of 88 patients with Fuchs’ heterochromia, comparing them with control subjects and other patients with uveitis. Although nine of these patients (10.2%) with Fuchs’ heterochromia had scars compatible with toxoplasmosis, the authors were unable to establish a special relationship between the two entities. Others have confirmed these observations . Schwab, in an interesting review, also looked for a relationship between toxoplasmosis and Fuchs’ heterochromic iridocyclitis. In his review he found that 13 of 25 patients with Fuchs’ heterochromic iridocyclitis had scars typical of toxoplasmosis and serologic evidence of the organism. This result was compared with results for 590 patients seen in the retina clinic at West Virginia University, with only 24 of these patients (4%) having retinal lesions typical of toxoplasmosis. Schwab concluded that a causal relationship between the two entities appears to exist, at least for a subgroup of patients with Fuchs’ heterochromia.

Effects in immunocompromised host

Special mention seems appropriate for ocular toxoplasmosis in the immunocompromised host, particularly the patient with AIDS not being treated with HAART (see Chapter 11 ). In these patients the development of ocular toxoplasmosis does not meet the criteria for an opportunistic infection but can indicate to the clinician that a change in the patient’s immune state has occurred. Indeed, we have seen active ocular toxoplasmosis in a patient with the AIDS-related complex in whom the full-blown picture of AIDS soon developed. In the patient with AIDS the degree of inflammatory disease associated with a toxoplasmic retinitis is usually far greater than that seen with cytomegalovirus (CMV) retinitis, perhaps indicating that reactivation (or acquisition) of the disease occurs earlier in the course of AIDS than in CMV retinitis, when the immune system is still capable of mustering a significant inflammatory response. In the United States, toxoplasmic retinitis is still a relatively uncommon disorder in patients with AIDS. Often in toxoplasmic retinitis numerous lesions will appear to be active at the same time, a most unusual finding in the immunocompetent patient.

The diagnosis of ocular toxoplasmosis in a patient with AIDS should initiate an evaluation of possible CNS disease as well. CNS toxoplasmosis has become a problem frequently seen in this patient population. The reason for the CNS and the eye incidence disparity is not known. The therapeutic management of these ocular lesions can be difficult. Despite therapy, disease can progress. Moorthy and colleagues reported two patients in whom a severe necrotizing retinitis was seen, with one developing a panophthalmitis and orbital cellulitis despite sulfadiazine, pyrimethamine, and folinic acid therapy. Whether the ocular manifestations of this disease represent reactivation or acquired disease is not known. One could conjecture that both routes are possible. Gagliuso and coworkers offer the observation that most ocular lesions in patients with AIDS are unassociated with a preexisting retinochoroidal scar and would therefore suggest acquired disease. This may be, but it is possible that cysts in the retina may remain dormant and become activated only in the immunosuppressed state. It is also possible that small lesions indicative of previous disease are simply engulfed in the large retinal necrotic lesion present in the patient with AIDS.

The patient who has undergone iatrogenic immunosuppression has a high risk of reactivation of ocular toxoplasmosis. As with patients with AIDS, if the immunosuppression cannot be reversed serious consequences may ensue, as reported by Yeo and colleagues. Singer and coworkers and Blanc-Jouvan and associates have reported cases of ocular toxoplasmic retinochoroiditis occurring after liver transplantation. In the patient of Singer and coworkers, because of the difficulty of diagnosing the disorder, the eye was ultimately enucleated. These authors emphasize the fulminant nature of the disease, suggestive far more of the type of ocular disease seen in patients with AIDS.

We conclude with the interesting observation that immune recovery uveitis may be driven by Toxoplasma antigens and not only by CMV, as it is generally thought. Sendi and colleagues reported the case of a 34-year-old HIV-positive man with a CD4 count of 11, who, after being placed on HAART therapy developed a uveitis after an increase in his CD4 count. Aqueous PCR for CMV was negative but positive for Toxoplasma , and his disease abated only after periocular steroid injections.

In summary, the diagnosis of ocular toxoplasmosis is primarily clinical. The typical lesions as described constitute the most important factor in our decision making. The one additional supportive test we believe to be very important is a positive toxoplasmosis titer at any dilution (see below). When the presentation is highly unusual, the diagnosis rests on a combination of multiple factors. In young children, infection with lymphochoriomeningitis virus, found in rodent feces, urine, and saliva, has been reported to mimic the ocular lesions associated with toxoplasmosis. In these patients toxoplasmosis could be confused with the ocular histoplasmosis syndrome, although vitreal cells are not present in the latter entity and the peripapillary changes are rarely evident in toxoplasmosis. The deep retinal presentation of toxoplasmosis may be confused with the white-dot syndromes, such as an unusual case of acute posterior multifocal placoid pigment epitheliopathy (see Chapter 29 ). In the immunocompromised host a single toxoplasmosis lesion early on may be confused with CMV retinitis. One possible way to help discriminate between the two entities is the use of fluorescein angiography. In an active toxoplasmosis lesion the central area will block fluorescein early and stain late, because the central part is where the greatest inflammatory response is taking place. This is in contrast to CMV retinitis, in which the central area will be atrophic and more readily hyperfluorescent early in the angiogram (Phuc LeHoang, MD, personal communication, 1989). Indocyanine green angiography will show that the lesion extends beyond the visible area, with hypofluorescent foci at all phases. A very good review of ocular disease can be found in Dr Gary Holland’s Jackson Memorial lecture. ,

Histopathology and immune factors

In ocular toxoplasmosis, cysts and tachyzoites can be found in the retina ( Fig. 14-13 ). The Toxoplasma organism most frequently is seen in the superficial portions of the retina. The lesion induced is necrotic, destroying the architecture of the retina. In many cases the underlying structures are destroyed as well, so that the disease at this point can certainly be classified as a chorioretinitis. Clinically, this destruction will permit the examiner to see underlying sclera quite clearly. Dutton and coworkers , were able to study these alterations more carefully using a murine model of congenital toxoplasmic retinochoroiditis. They noted that the inflammation ranged from a low-grade mononuclear infiltrate to total destruction of the outer retina, the retinal pigment epithelium, and choroid. Of great interest was the fact that photoreceptor outer segments were phagocytosed by macrophages, whereas the Toxoplasma cysts did not appear to be the center of the inflammatory attack. Roberts and colleagues characterized histologically the eyes from 10 fetuses and two infants with congenital toxoplasmosis. Retinitis was present in 10 of 18 eyes, necrosis in four of 18, retinal pigment epithelial changes in 12 of 18, choroidal inflammation in 15 of 18, and optic neuritis in five of eight fetal eyes. Parasites were found on immunohistologic examination in 10 of 18 eyes. It appeared to the authors that the inflammatory response mounted by the host accounted for part of the damage seen.

Figure 14-13.

Photomicrograph showing toxoplasmic cysts in retina. Cysts are larger circular structures, located mainly in or close to the nerve fiber layer.

(Courtesy of D. Cogan, MD.)

Because of the selective photoreceptor destruction, one can speculate that autoimmune mechanisms may be important in the tissue destruction seen (see discussion of autoimmunity in Chapter 1 ). Indeed, when we had the opportunity to evaluate this issue, we found that in 16 of 40 patients (40%) with ocular toxoplasmosis an in vitro proliferative response to the retinal S-antigen was seen. This finding could indicate that an autoimmune component to the inflammatory disease is initiated, with destruction of the retina by a parasite and subsequent sensitization to the uveitogenic antigen. We also found that proliferative responses to the p22 antigen approached those seen to crude toxoplasmosis antigen, whereas the response to the p30 membrane antigen was considerably less striking. Others have shown that Toxoplasma antigen stimulates CD25+ helper cells.

Immune response

The host’s immune response is exceptionally important in the ultimate expression of toxoplasmosis. In animal models, resistance, as measured by survival after challenge with the organism, has been reported to be regulated by at least five genes, one of which is within the region of the H-2 antigen (the equivalent of the major histocompatibility complex [MHC] in humans). Brown and McLeod have found that class I MHC genes, as well as the CD8+ fraction of T cells, determine the cyst number in Toxoplasma infection. Jamieson et al. reported polymorphism associations at the COL2A1 encoding type II collagen associated only with those who had ocular disease. Others have looked at the effect of cytokines on the multiplication of the Toxoplasma organism. Interferon (IFN)-γ, as well as tumor necrosis factor (TNF)-α and transforming growth factor-β all appear to play a role in inhibiting multiplication. , Gazzinelli and colleagues reported that a reactivation of T. gondii , at least in the experimental mouse model, is due to a downregulation of IFN-g and TNF-a, which leads to reduced macrophage (with a decrease in inducible nitric oxide synthase and macrophage activation gene 1) and glial activation, a release of parasite growth, and tissue damage. Shen and associates found the presence of IFN in the eyes of Toxoplasma -infected mice, but that inflammatorily induced apoptosis was caused by several factors, not only Fas/FasL interactions. Beaman and colleagues reported that interleukin (IL)-6 enhanced intracellular reproduction of T. gondii and actually reversed the effect of IFN-g-mediated killing, which contradicts the finding of Lyons and colleagues who stated that IL-6 knockout mice in a chronic toxoplasmosis model had more severe disease and an increased parasite burden. This finding is particularly important in light of the fact that the retinal pigment epithelium produces large amounts of IL-6. An analysis of ocular fluids from uveitic eyes (including those with from patients with toxoplasmosis) demonstrated the presence of various cytokines, including IL-6, IFN, and IL-10. Recently, Zamora and coworkers showed that the Toxoplasma organism invades human retinal endothelial cells more efficiently than they do human dermal endothelial cells. In another study, Feron and colleagues characterized 10 T-cell specimens from the vitreous of patients with toxoplasmosis. Although the cell lines were initially generated by mitogenic stimulation, they were all CD4+ and appeared reactive to Toxoplasma antigens and not to any retinal antigens. The majority had a Th2 profile. It has also been suggested that CD8+ T cells directed against the Toxoplasma organism appear during the acute phase of the disease, whereas CD4+ parasite-specific T cells appear with chronicity of the disease. Denkers and colleagues reported that the Toxoplasma organism possesses a superantigen that expands murine Vβ5-expressing cells, most of which were CD8+. It may be that this superantigen-driven expansion of predominantly IFN-g-secreting CD8+ cells is partly why the early immune response is seen. Curiel and coworkers reported the cloning of human CD3+, CD4+ T cells that lyzed autologous target cells which had been pulsed with Toxoplasma antigen or infected with live tachyzoites. These results suggest that specific immunotherapy through the development of vaccine may be possible. Patients with acute toxoplasmosis have increased serum levels of CXCL8, part of the chemokines that control leukocyte infiltration and can even modulate angiogenesis.

Although antibodies are usually readily made, it is the cellular component of the immune system that must be intact for a resolution of the disease process. However, antibody production may play an important role in establishing a state of premunition (immunity from infection) in Toxoplasma infection. In an attempt to evaluate why newborn infants seem to have difficulty in fighting the Toxoplasma infection, Wilson and Haas evaluated the cellular defenses against T. gondii in newborns. They noted that newborn and adult macrophages killed the organism equally well, but supernatants from cord blood-derived concanavalin A-stimulated mononuclear cells activated macrophages less effectively than supernatants produced from adult blood cells. This difference appeared to lie in the CD4+ fraction of T cells. The cord blood appeared to produce fewer lymphokines capable of activating macrophages, including IFN-g. The authors did not believe that enhanced generation of reactive oxygen intermediates was important in explaining the differences between the adult and newborn responses, although recent notions would suggest that nitrous oxide and its effects in macrophages may indeed play a very important role. Roberts and associates, in a murine model, demonstrated that inhibition of nitric oxide by administration of l ω- nitrol -arginine methyl ester made the disease worse. Of interest is the fact the organism replicates in the macrophage, and this reduced killing in the newborn may then lead to greater susceptibility.

Inflammatory response

What then is the cause of the focal, retinal inflammatory response? So far the data still support the notion that it is the release of actively proliferating tachyzoites, often from a long-dormant cyst. We know now of many immune components that can stimulate this tachyzoite–bradyzoite interconversion ( Fig. 14-14 ) However, immune studies with patients with ocular toxoplasmosis suggest that other factors are involved. Wyler and coworkers noted that lymphocytes from patients with ocular toxoplasmosis demonstrated in vitro responses not only to Toxoplasma antigens but also to a crude retinal preparation. As already mentioned, our patients with toxoplasmosis have demonstrated in vitro cellular responses to the retinal S-antigen, a purified antigen from the photoreceptor region, and the site of the most intense inflammatory response in Dutton and colleagues’ murine model for toxoplasmosis. Abrahams and Gregerson detected circulating antibodies to various retinal antigens in patients with toxoplasmosis. Whittle and colleagues, using indirect immunofluorescent techniques over normal human cadaver retina, determined the presence of human antiretinal antibodies in patients with ocular toxoplasmosis. Of the sera from 36 toxoplasmosis patients, 94% demonstrated photoreceptor layer reactivity. However, only 27 of these patients had anti-S-antigen antibodies as determined by an enzyme-linked immunosorbent assay (ELISA). The authors concluded that antiretinal activity can be accounted for by antibodies directed not only against S-antigen but also against other antigens. These observations raise the provocative argument that the inflammatory disease we see is at least partly, or at times, autoimmune driven. However, Vallochi et al. reported that the peripheral blood from ocular toxoplasmosis patients also recognized retinal antigens. But their data suggest that such autoimmune responses were associated with less severe disease. We know that the Toxoplasma cyst may include tissue from the host, and perhaps S-antigen is sequestered there until the cyst breaks open. Alternatively, the initial destruction caused by the proliferating organism releases immunogenic antigens to the general circulation, thereby causing sensitization. This then results in a secondary autoimmune response, which can prolong the initial response or become the center of a recurrent inflammatory episode some time in the future.

Oct 21, 2019 | Posted by in OPHTHALMOLOGY | Comments Off on Ocular Toxoplasmosis
Premium Wordpress Themes by UFO Themes