Toxoplasma gondii and Schizophrenia

E. Fuller Torrey, Robert H. Yolken


Emerging Infectious Diseases. 2003;9(11) 

In This Article

Serologic Studies of Patients with Schizophrenia

In the course of doing our studies, we discovered that much research had been published in languages other than English and was not listed on searchable databases. Through direct contact with authors and by obtaining references listed on their papers, we identified 13 relevant studies published between 1953 and 1979,[15,16,17,18,19,20,21,22,23,24,25,26,27] as listed on the Table . Some publication bias is likely, since negative studies are less likely to have been submitted or published.

The 13 studies used a variety of immunologic methods for measuring antibodies, including the Sabin Feldman dye test, skin tests, and complement fixation (CF). One study used a test in which an alkaloid from T. gondii caused a tropical fish, Lebistes reticulatus, to change color.[19] Some of the studies compared the relative efficacy of two different tests. Most of the studies defined Toxoplasma-positive results as the presence of a skin reaction or antibodies above a certain titer but often without specifying the precise details of the method; thus, comparing the older studies with each other was not possible. Most of these studies also did not specify what diagnostic criteria were used for schizophrenia, but since at least 12 of them used inpatients, the patients likely had a severe psychiatric disorder. Similarly, most of the studies did not specify the origin of their control group other than saying such things as "681 healthy persons working or studying in the city of Gdansk".[15]

Despite these limitations, 12 of the 13 studies found that the patient group had a higher percentage of antibodies to Toxoplasma than the control group. In eight of the studies, the increase was statistically significant by chi square at the level of p < 0.05. In the two largest studies, Kozar[15] in Poland reported antibodies in 495 (52%) of 961 psychiatric inpatients compared with 170 (25%) of 681 controls, and Roch and Varela[25] in Mexico found antibodies in 836 (86%) of 973 patients with schizophrenia compared with finding antibodies in 30% of the general population.

We identified no studies that were done between 1979 and 1999. Since that time, six studies have been carried out, including our own.[28,29,30,31,32] All used enzyme immunoassay methods for measuring antibodies to Toxoplasma. All of the studies also used modern diagnostic criteria for schizophrenia; three studies included patients with chronic disease, and three included patients who were in the first episode of the disease. All of the studies identified their control groups, and some attempts were made to match them to the patient groups.

The results of these studies are summarized in the Table . In all of the studies, the patients had more antibodies to Toxoplasma than the control groups, and in the three studies, carried out in China and Germany, of patients who were having their first-episode of schizophrenia, the differences were statistically significant. One of the first-episode studies, carried out in Cologne by Leweke et al.,[32] divided the first-episode patients into those who had never received antipsychotic treatment and those who had received some treatment. The antibody levels for the treated group were intermediate between the levels of the never-treated group and those of the control group, suggesting that antipsychotic medication may have decreased the antibody levels. This conclusion is supported by a study that indicated that some antipsychotic medications inhibit the growth of T. gondii in cell culture.[33]

The Leweke et al. study also collected cerebrospinal fluid (CSF) from the first-episode patients. The level of Toxoplasma antibody in the CSF of untreated patients was significantly higher than the normal controls (p < 0.0001).[32] Treated first-episode patients had CSF antibody levels intermediate between those of the untreated patients and the controls, just as was found for the sera.

In addition to these studies on adults with schizophrenia, a study was also conducted by analyzing serum samples from pregnant women, obtained shortly before delivery, who gave birth to children in whom schizophrenia or other psychoses developed. Preliminary analysis indicates an increased rate of immunoglobulin (Ig) M (but not IgG) class antibodies to Toxoplasma gondii in mothers with infants in whom schizophrenia developed later, suggesting that the mothers were experiencing an active infection or that they had persistent IgM antibodies, as described in other studies. Increased levels of IgM antibodies were not found to other perinatal pathogens such as rubella virus or cytomegalovirus.[34]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: