**8. The mechanism of encephalopathy and risk factors in loiasis:**

A high density of microfilaria (> 30000 mf/ml) seems to be the most plausible risk factor for *Loa loa* encephalopathy (Figure. 7). In addition, the genetic heterogeneity of this parasite could explain th e higher prevalence of encephalopathy in certain regions. However, parasites isolated in parts of Cameroon with a high prevalence of encephalopathy were not found to differ genetically from those found in other regions of Africa (Gabon and Nigeria) (Higazi et al., 2004). It has also been suggested that hybridization between simian and human strains of *Loa loa* may be a cofactor for encephalopathy, but this remains to be demonstrated. Although it is possible to cross human and simian strains of *Loa loa*, animal strains do not develop in humans, as demonstrated by implantation of simian adult parasites or injection of infective larvae (L3) in human volunteers (Duke, 2004; Nutman et al., 1991). Moreover, vectors of simian strains of *C. langi* and *C. centurionis* do not bite humans and tend to be active after dark. In contrast, human strains of *Loa loa* can be transmitted to non human primates, and hybrids of human and simian strains can be produced experimentally. However, the two sets of strains normally develop in different host-parasite systems (Fain, 1988), and such hybrids are unlikely to occur in natural conditions. A genetic predisposition to developing microfilaremia (Garcia et al., 1999) could also favor the onset of encephalitis in some cases. Coinfection by *Loa loa* and other parasites such as *Plasmodium* (Hartgers et al., 2006; Kamgno et al., 2008) bacteria (Bonnet et al., 1943; Cattan et al., 1960) and viruses (Cauchie et al., 1965), might cause lesions through which microfilaria could enter the nervous system and brain. However, it has been reported that treatment of such coinfections has little impact on the outcome of encephalitis (Kamgno et al., 2008). Finally, alcohol consumption has also been forwarded as a possible risk factor (scientific working group,2003).

The pathophysiologic mechanisms underlying encephalitis in patients with loaisis may involve massive microfilarial death, leading to vascular embolism and inflammation. Interactions with drugs and other substances (alcohol, drugs, dietary components, etc.) may also play a role, through competition for biological carrier molecules. Glycoprotein P, a component of the blood-brain barrier, plays a role in drug entry to the brain. Substance P deficiency could lead to a rise in drug concentrations in the brain, resulting in severe

diuresis 1 ml/kg/hour. Perfuse 2000 ml of 5% glucose (including 6 g of NaCl, 3-4 g of KCl, 2 g of calcium) or mixt sera (with 4 g of KCl and 2 g of calcium) for 24 hours for an adult weighing 60 kg, or 500 ml every 6 hours (28 drops/minute). If dehydration persists, perfuse 1000 ml of these solutions over 8 hours, depending on clinical status. In case of fever, add 1 ml of solution per kilogram for each degree above 37°C. In case of coma, the treatment aim is to avoid bedsores, bronchial accumulation and intercurrent disorders, by mobilizing the patient every three hours and massages to prevent complications of decubitus; urinary probing, mouth care with sodium bicarbonate solution, and eye care with 9% NaCl. Pose of orpharyngeal canicle. If the Glasgow score is less than 10/15, transfer to a specialized intensive care unit. If recovery is slow, use gastric gavage with milk, soja…. Complementary examinations are necessary to rule out any other causes of coma (meningitis, hypo- or hyperglycemia, cerebral malaria, etc.), including thick blood smear (to search for *Loa loa* and malaria); glycemia, glycosuria, proteinuria and lumbar puncture (the liquid should be clear, but *Loa loa* microfilaria should be present between days 3 and 7 in *Loa loa* encephalitis). Removal of eye worm has been reported to cure spontaneous encephalitis (Kenney and

**8. The mechanism of encephalopathy and risk factors in loiasis:** 

A high density of microfilaria (> 30000 mf/ml) seems to be the most plausible risk factor for *Loa loa* encephalopathy (Figure. 7). In addition, the genetic heterogeneity of this parasite could explain th e higher prevalence of encephalopathy in certain regions. However, parasites isolated in parts of Cameroon with a high prevalence of encephalopathy were not found to differ genetically from those found in other regions of Africa (Gabon and Nigeria) (Higazi et al., 2004). It has also been suggested that hybridization between simian and human strains of *Loa loa* may be a cofactor for encephalopathy, but this remains to be demonstrated. Although it is possible to cross human and simian strains of *Loa loa*, animal strains do not develop in humans, as demonstrated by implantation of simian adult parasites or injection of infective larvae (L3) in human volunteers (Duke, 2004; Nutman et al., 1991). Moreover, vectors of simian strains of *C. langi* and *C. centurionis* do not bite humans and tend to be active after dark. In contrast, human strains of *Loa loa* can be transmitted to non human primates, and hybrids of human and simian strains can be produced experimentally. However, the two sets of strains normally develop in different host-parasite systems (Fain, 1988), and such hybrids are unlikely to occur in natural conditions. A genetic predisposition to developing microfilaremia (Garcia et al., 1999) could also favor the onset of encephalitis in some cases. Coinfection by *Loa loa* and other parasites such as *Plasmodium* (Hartgers et al., 2006; Kamgno et al., 2008) bacteria (Bonnet et al., 1943; Cattan et al., 1960) and viruses (Cauchie et al., 1965), might cause lesions through which microfilaria could enter the nervous system and brain. However, it has been reported that treatment of such coinfections has little impact on the outcome of encephalitis (Kamgno et al., 2008). Finally, alcohol consumption has also been forwarded as a possible risk factor

The pathophysiologic mechanisms underlying encephalitis in patients with loaisis may involve massive microfilarial death, leading to vascular embolism and inflammation. Interactions with drugs and other substances (alcohol, drugs, dietary components, etc.) may also play a role, through competition for biological carrier molecules. Glycoprotein P, a component of the blood-brain barrier, plays a role in drug entry to the brain. Substance P deficiency could lead to a rise in drug concentrations in the brain, resulting in severe

Hewitt, 1950).

(scientific working group,2003).

neurotoxicity. This deficiency could be caused by genetic polymorphism, deficient glycoprotein P production, or glycoprotein P inhibition. Indeed, severe neurological adverse effects of ivermectin are observed in CF-I mice, that are deficient in MDRIA glycoprotein P (Kwei et al., 1999). Several drug carrier molecules such as MDR, MRP, OATP and glycoprotein P have been detected on the apical and basolateral membranes of epithelial cells in cerebral capillary membranes (Cordon Cardo et al., 1989; Huai-Yun et al., 1998; Kusuhara et al., 1998; Gao et al., 1999). Glycoprotein P is the most widely studied of these molecules. The risk of encephalitis could also be influenced by genetic factors. Indeed, dogs homozygous for a 4-bp deletion of the *MDR1* gene (resulting in premature termination of glycoprotein P synthesis) are highly sensitive to ivermectin (Mealy et al., 2001). In addition, CFI mice exhibiting low glycoprotein P production are more sensitive to ivermectin neurotoxicity than their wild-type counterparts (Umbenhauer et al., 1997). In addition to the neurotoxicity of ivermectin accumulating in the brain, through a deficiency in glycoprotein P or other carriers, neurotoxicity may result from interactions between drugs competing for the same carrier binding site. Other glycoprotein P substrates may compete with ivermectin, leading to a reduction in ivermectin efflux from the brain. This has been demonstrated in mice treated with both ivermectin and cyclosporine (Marques-Santos et al., 1999). It is important to note that ivermectin and DEC both lead to progressive neurological complications in patients with high microfilaremia and also ocular lesions (retinal or subconjunctival hemorrhage) linked to microemboli created by the parasite. It therefore appears that the etiology of *Loa loa*associated encephalitis associated with these two drugs is linked to clumping of dead microfilaria in vessels, leading to emboli and local vascular inflammation.

Fig. 7. Two views of *Loa loa* microfilariae in blood of an hypermicrofilaremic individual

It seems that the adult worm and microfilaria are both risk factors for spontaneous encephalitis. Adult worms have been implicated in the neuropsychological complications of loaisis in two European patients (Kenny & Hewitt, 1950). In both cases, extraction of the adult worm led to a clinical improvement and to a decline in eosinophilia from 56-50% to 3%. Location of adult worms in the subarachidonic space at the base of the brain has also been implicated in neuropsychological complications (Bertrand-Fontaine et al., 1948). The abundance of microfilaria is an important risk factor, because of their mobility in small

Encephalitis Due to *Loa loa* 353

lymphocytes and histiocytes may form in cardiac tissue, with a few leukocytes close to microfilaria in extracranial organs. There is no evidence of phagocytosis. Alternatively, complications could be due to obstruction caused by microfilarial and to the toxicity of their

The immune response could also play a role in encephalitis due to *Loa loa*, for example through the formation of circulating immune complexes that deposit in tissues and vessels. This could result in complement activation and influx of polymorphonuclear cells and basophils, followed by the release of vasoactive amines, causing retraction of endothelial cells and increased vascular permeability. Polymorphonuclear cells that fail to phagocytose deposits of immune complexes may degranulate, causing local tissue damage (Fig. 8A). The large amount of antigens associated with abundant microfilariosis may persistently stimulate an inefficient antibody response resembling type III hypersensitivity. Chronic lesions induced by such phenomena could lead to vessel destruction and encephalitis when they occur in the brain, or nephropathies when the kidneys are affected. Cardiac and renal involvement have both been described in loaisis. Alternatively, T lymphocytes sensitized by *Loa loa* antigens (Fig. 8B) could release cytokines, thus attracting activated macrophages. With the persistence of *Loa loa* antigens, activated macrophages could trigger chronic granulomatous reactions resembling type IV hypersensitivity. Such responses are observed in chronic infections such as schistosomiasis (Brian et al., 1983). Granulomas have been observed in the brain of a person infected by *Loa loa*. These observations suggest that type III and IV hypersensitivity reactions could be involved in the development of *Loa loa* encephalopathy. Finally, it is conceivable that spontaneous encephalitis is the end result of a long process involving deposits of immune complexes in several deep organs, the most sensitive being the

brain and kidneys. Cofactors (drugs, coinfection, etc.) could accelerate this process.

Administration of filaricidal drugs in massively infected patients often results in encephalitis, in the absence of any other known cause. The mechanism underlying encephalitis in this setting is unclear. Microfilaria have been found in *parasite*-infected animal brains (Hashimoto, 1939, quoted byJanssens,1952), including those coinfected by trypanosomes (Peruzzi, 1928). Consequently, the presence of *Loa loa* microfilaria in the brain cannot alone explain the onset of encephalitis. The heterogeneous nature of the associated clinical manifestations poses problems for prevention and timely patient management. Studies based on an experimental model, such as non human primates infected by human isolates, could help to identify predictive markers of *Loa loa* encephalitis and specific clinical complications (Orihel & Ebrehard 1985; Duke, 1960). Indeed, clinical expression of this filariosis is similar in humans and non human primates (Pinder et al., 1994), and hypermicrofilaremia can be reproduced in non human primates. Despite the existence of potent microfilaricidal drugs (DEC and ivermectin), new macrofilaricides or compounds capable of inducing a gradual decline in microfilaria without triggering encephalitis are needed. Most cases of encephalitis have been reported in Cameroon during mass treatment with ivermectin, but similar cases may go unreported in other endemic regions, especially if they occur in rural settings without adequate medical facilities. Specific studies are needed to evaluate the prevalence and characteristics of *Loa loa* encephalopathy in endemic areas. Studies of polymorphisms of human drug carrier molecules and proinflammatory cytokine synthesis are also necessary. As all current treatments, including albendazole (Blum et al., 2001), can induce encephalitis in

metabolic products (Weil et al., 1926).

**9. Ongoing and future directions** 

vessels and capillaries and outside the circulatory apparatus. How exactly microfilaria cross the blood-brain barrier remains to be determined. As stated above, coinfection by other pathogens (such as *Plasmodium*) could weaken this barrier, leading to vascular lesions that allow microfilaria to enter the brain. Head trauma could have a similar effect. In welldocumented cases of encephalitis (Bogaert et al., 1955), it has been shown that microfilaria cross the vascular barrier and penetrate into deep tissues, where focal necrosis occurs around dead parasites. These foci are surrounded by inflammation and fibrosis, and giant multinucleated cells arise in the spleen, liver and brain. Aggregates of neurological

Fig. 8. Potential immunological mechanisms for induction of *Loa loa* encephalitis

vessels and capillaries and outside the circulatory apparatus. How exactly microfilaria cross the blood-brain barrier remains to be determined. As stated above, coinfection by other pathogens (such as *Plasmodium*) could weaken this barrier, leading to vascular lesions that allow microfilaria to enter the brain. Head trauma could have a similar effect. In welldocumented cases of encephalitis (Bogaert et al., 1955), it has been shown that microfilaria cross the vascular barrier and penetrate into deep tissues, where focal necrosis occurs around dead parasites. These foci are surrounded by inflammation and fibrosis, and giant multinucleated cells arise in the spleen, liver and brain. Aggregates of neurological

Fig. 8. Potential immunological mechanisms for induction of *Loa loa* encephalitis

lymphocytes and histiocytes may form in cardiac tissue, with a few leukocytes close to microfilaria in extracranial organs. There is no evidence of phagocytosis. Alternatively, complications could be due to obstruction caused by microfilarial and to the toxicity of their metabolic products (Weil et al., 1926).

The immune response could also play a role in encephalitis due to *Loa loa*, for example through the formation of circulating immune complexes that deposit in tissues and vessels. This could result in complement activation and influx of polymorphonuclear cells and basophils, followed by the release of vasoactive amines, causing retraction of endothelial cells and increased vascular permeability. Polymorphonuclear cells that fail to phagocytose deposits of immune complexes may degranulate, causing local tissue damage (Fig. 8A). The large amount of antigens associated with abundant microfilariosis may persistently stimulate an inefficient antibody response resembling type III hypersensitivity. Chronic lesions induced by such phenomena could lead to vessel destruction and encephalitis when they occur in the brain, or nephropathies when the kidneys are affected. Cardiac and renal involvement have both been described in loaisis. Alternatively, T lymphocytes sensitized by *Loa loa* antigens (Fig. 8B) could release cytokines, thus attracting activated macrophages. With the persistence of *Loa loa* antigens, activated macrophages could trigger chronic granulomatous reactions resembling type IV hypersensitivity. Such responses are observed in chronic infections such as schistosomiasis (Brian et al., 1983). Granulomas have been observed in the brain of a person infected by *Loa loa*. These observations suggest that type III and IV hypersensitivity reactions could be involved in the development of *Loa loa* encephalopathy. Finally, it is conceivable that spontaneous encephalitis is the end result of a long process involving deposits of immune complexes in several deep organs, the most sensitive being the brain and kidneys. Cofactors (drugs, coinfection, etc.) could accelerate this process.

#### **9. Ongoing and future directions**

Administration of filaricidal drugs in massively infected patients often results in encephalitis, in the absence of any other known cause. The mechanism underlying encephalitis in this setting is unclear. Microfilaria have been found in *parasite*-infected animal brains (Hashimoto, 1939, quoted byJanssens,1952), including those coinfected by trypanosomes (Peruzzi, 1928). Consequently, the presence of *Loa loa* microfilaria in the brain cannot alone explain the onset of encephalitis. The heterogeneous nature of the associated clinical manifestations poses problems for prevention and timely patient management. Studies based on an experimental model, such as non human primates infected by human isolates, could help to identify predictive markers of *Loa loa* encephalitis and specific clinical complications (Orihel & Ebrehard 1985; Duke, 1960). Indeed, clinical expression of this filariosis is similar in humans and non human primates (Pinder et al., 1994), and hypermicrofilaremia can be reproduced in non human primates. Despite the existence of potent microfilaricidal drugs (DEC and ivermectin), new macrofilaricides or compounds capable of inducing a gradual decline in microfilaria without triggering encephalitis are needed. Most cases of encephalitis have been reported in Cameroon during mass treatment with ivermectin, but similar cases may go unreported in other endemic regions, especially if they occur in rural settings without adequate medical facilities. Specific studies are needed to evaluate the prevalence and characteristics of *Loa loa* encephalopathy in endemic areas. Studies of polymorphisms of human drug carrier molecules and proinflammatory cytokine synthesis are also necessary. As all current treatments, including albendazole (Blum et al., 2001), can induce encephalitis in

Encephalitis Due to *Loa loa* 355

The risk of *Loa loa* encephalitis must be taken in to account when managing patients in and from endemic areas. This severe form can occur spontaneously or be triggered by antifilarial treatment in highly microfilaremic patients. The underlying mechanism appears to include embolism following massive death of microfilaria, genetic polymorphism of biological drug carriers, and immunological processes. More work is needed to develop a diagnostic test, as well as new drugs and possibly a vaccine. Further characterization of *Loa loa* encephalitis in endemic regions and in animal models is needed to understand the mechanisms underlying

The work on filariae *Loa loa* is supported by CIRMF. CIRMF is sponsored by total Gabon, Gabonese state and Ministère de la cooperation Française. We thank Ms Line Mengome, Hubert Moukana and Mbou Moutsimbi Roger Antoine for their assistance during the

Abel L, Joly v, Yeni P, Carbon C. (1986). Apheresis in the management of loiasis with high

Andi JJ, FF Bishara, OO Soyinka. (1981). Relation of severe eosinophilia and microfilariasis

Akue Jean Paul, Devaney Eileen. (2002). Transmission Intensity Affects Both Atigen –

Akue JP, Egwang TG, Devaney E. (1994). High level of parasite specific IgG4 in the absence

Akue Jean Paul, Marcel Hommel, Eileen Devaney. (1997). high levels of Parasite-Specific

Akue JP Hommel M, Devaney E. (1998). IgG subclass recognition of *Loa loa* antigens and

Akue JP, Hommel M, Devaney E. (1996). Markers of *Loa loa* infection in permanent residents

Ajuh PM, Akue JP, Boutin P, Everaere S Egwang TG. (1995). *Loa loa*: structural diversity of a

Baize S, Wahl G, Soboslay P T, Egwang T G, Geaorges AJ. (1997). T helper responsiveness in

*Clinical and Experimental Immunology*, Vol. 108,pp.272-278,ISSN:00099104

microfilariaemia and renal disease. *British Medical Journal* , Vol. 292, No. 4 January

to chronic African endomyocardial fibrosis. *British Heart Journal*, Vol.45, No. 6,

Specific and Nonspecific T-cell Proliferative Responses in *Loa loa* Infection . *Infection* 

of microfilaremia in Loa loa infection. *Tropical Medicine and Parasitology* ,Vol.45,No.

IgG1 correlate with the amicrofilaremic State in *Loa loa* Infection. *Journal of Infectious* 

their correlation with clinical status in individuals from Gabon. *Parasite* 

of a loiasis endemic area of Gabon. *Transactions of the Royal Society of Tropical* 

15kDa repetitive antigen. *Experimental Parasitology* ,Vol. 81, pp.145-153, ISSN:0014-

human *Loa loa* infection: defective specific proliferation and cytokine production by CD4+ T cells from microfilaraemic subjects compared with amicrofilaraemics.

the onset and outcome of encephalitis in patients with loaisis.

**10. Conclusion** 

**11. Acknowledgement** 

preparation of the manuscript.

1986) pp 24, ISSN

(June 1981), pp. 672-680, ISSN:

3, pp. 246-248, ISSN 9177-2392

*and Immunity*, Vol. 70,pp.1475-1480,ISSN: 0019-9567

*Diseases*, Vol. 175, pp:158-163 ISSN: 0022-1899

*Immunology* , Vol. 20,pp.387-393,ISSN:0141-9838

*Medicine and Hygiene,* Vol. 90,pp. 115- 118, ISSN:

**12. References** 

4894

highly microfilaremic patients, the antifilarial activity of African traditional herbal remedies may be of interest. Most of these plants are well accepted and tolerated, and preliminary cytotoxicity results are encouraging (Mengome et al., 2010). There is currently no evidence of the existence of a symbiont in *Loa loa* (McGarry et al., 2003; Buttner et al., 2003), that might warrant concurrent antibiotic therapy for patients with loasis. Because *Loa loa* infection often goes undiagnosed, cases of encephalitis in *Loa loa* endemic areas may be attributed to viruses, bacteria or other parasites. This underdiagnosis is due partly to the lack of a simple, specific and rapid diagnostic test available at points of treatment for use by non specialists. Some candidate antigens have been identified and produced (Azzibrouck et al., 2010). Although spontaneous encephalitis may be caused by the adult worm, life-threatening forms are generally due to massive death of microfilaria. However, in endemic areas, about one-third of infected persons are microfilaremic and only 5% are strongly microfilaremic, the remainder being amicrofilaremic (Van Hoegaerden et al., 1986; Dupont et al., 1988). The fact that these latter persons live permanently in areas of continuous transmission without becoming microfilaremic points to the existence of a natural control mechanism. Further studies of these subjects could help to find ways of clearing microfilaria without triggering encephalitis. Noteworthy immunological differences have been found between microfilaremic and amicrofilaremic subjects. The latter patients exhibit a stronger immune response against *Loa loa* antigens, both qualitatively and quantitatively (Pinder et al., 1988; Pinder et al., 1992; Egwang et al., 1988a; Egwang et al., 1988b;Egwang et al.,1989; Akue et al., 1997; Akue et al., 1998; Baize et al.,1997; Akue & Devaney, 2002). Finally, more work is needed to determine the role of immune complex deposition in the onset of encephalitis in patients with loaisis.

Fig. 9. *Loa loa* adult worm
