Inammatory Markers in Patients With Treatment-Resistant Schizophrenia in Ethiopia: A Comparative Study

Background: Accumulating evidence indicates that schizophrenia is accompanied by an activation of the immune system; however, there is limited data from low and middle-income countries. Inammatory markers could even be more important in these settings where infectious conditions may play a more prominent role in the causation and maintenance of schizophrenia. The aim of this study was to assess the level of inammatory markers--High sensitive C-reactive protein (hsCRP) and Interleukin-6 (IL-6)--in patients with schizophrenia. Materials and methods: The study population consisted of a total of 132 study participants; 82 with schizophrenia and 50 controls. hsCRP and IL-6 were measured using Cobas Intgera 400 Plus and Cobas e 411 analysers respectively. Results: The levels of hsCRP and IL-6 were signicantly increased among patients with schizophrenia compared to controls. Controlling for potential confounders (age, sex and body mass index), having a diagnosis of schizophrenia remained signicantly associated with increased hsCRP (beta =0.29; 95% CI=0.10, 0.49) and IL-6 (beta=3.60; 95% CI: 1.35, 5.86). Conclusion: The nding is consistent with reports from high income countries and conrms that inammatory processes may have a role in the pathophysiology of schizophrenia regardless of setting. Despite failure of some interventions with anti-inammatory properties, interventions to reduce inammation are still worth pursuing.


Background
Schizophrenia is a relatively rare but serious mental disorder affecting about 1% of the adult population.
Due to medical co-morbidities and other patient and service factors, people diagnosed with schizophrenia have a high overall mortality rate [1,2], which may occur 20 to 30 years earlier than the general population [3]. The exact cause of schizophrenia is not established; however, epidemiological evidence indicates that several risk factors, including genetic susceptibility [4], season of birth [5], increasing parental age [6], and prenatal exposure to infection [7] may contribute to the development of schizophrenia.
Immunological dysfunction and in ammation, and exposure to infectious agents that lead to immune response, such as Toxoplasma gondii [8], in uenza [9] and interaction of environmental factors and stress are also considered important [10]. This theory has been supported by nding of elevated C-reactive Protein (CRP) and interleukin 6 ( IL-6) in mental disorders [11].
CRP is nonspeci c serum protein, traditionally considered as an acute phase immune response marker. It is mainly produced by liver cells and is directly modulated by both interleukin (IL) 1β and IL-6, both in ammatory markers increased during psychotic state [12][13][14].
A number of studies in both rst episode and persistent or recurrent schizophrenia show increased serum levels of acute phase proteins, such as CRP, and proin ammatory markers like tumor necrosis factor (TNF-alpha), IL-6, and IL-1β, although with some inconsistency [15][16][17][18]. A recent meta-analysis reported higher CRP levels in patients with schizophrenia than control groups [13] and two individual studies (case-control and longitudinal birth cohort study) indicated that increased CRP was associated with increased risk of schizophrenia [19,20]. Yet, other studies, albeit fewer, have found no difference between the level of serum CRP or IL-6 in patients with schizophrenia and control subjects [21,22].
Most of these studies have been conducted in high income countries. Although infectious causes may have more relevance in the causation of schizophrenia in low and middle-income countries, there is extreme dearth of data from these countries. Therefore, this study aimed to investigate the serum level of in ammatory markers (CRP and IL-6) among patients with schizophrenia.

Demographic and Clinical Characteristics
The socio-demographic characteristics of participants is presented in Table 1. Compared to the control group, participants with schizophrenia were predominantly male and slightly older. Over two-thirds of the patients with schizophrenia were single during study period, and lived with parental family. Patients with Schizophrenia and control participants were similar in terms of current BMI. The mean PANSS score was 89.2 and over 61% had markedly ill as de ned by their PANSS score.

Discussion
The main nding of the study is that signi cantly higher level of both hsCRP and IL-6 were observed in patients with schizophrenia than the control group. To the best our knowledge, this is the rst study to investigate the serum level of both hsCRP and IL-6 among patients with schizophrenia in Ethiopia. It is also one of the very few studies from Africa. More broadly, diagnosis of in ammatory diseases and in ammatory markers in Africa is rare. Nevertheless, there is evidence of increase in the incidence and prevalence of some in ammatory diseases in the developing world, which may increase the signi cance of in ammation in neuropsychiatric syndromes.
The result of the present study concur with studies from Western countries that consistently indicate that patients with schizophrenia have high serum levels of hsCRP and IL-6 [19,[23][24][25][26][27]. Elevated in ammatory markers in patients with schizophrenia have been reported in case control studies [19,28] and treatment studies [29]. This is also found in people with both acute [24], chronic [25] and treatment-resistant [30] illnesses. Because of the consistency of this nding, neuro-in ammation has been linked with the causation of schizophrenia and other mental disorders. However, such studies are rare in low and middle income countries where the majority of the population of the world lives. We believe that this study contributes to this particular knowledge gap and the broader issue of lack of such studies even in the general population [31].
A signi cant negative correlation were observed between hsCRP and total PANSS score in our study. The result in the literature in this regard is mixed: some studies have reported negative correlation as observed in our study [32] while others have reported either positive association [28,33] or no association between hsCRP and total PANSS score [34][35][36][37]. Despite these inconsistencies, hsCRP appears to be an important in ammatory marker in our particular setting although additional con rmatory studies would be needed.
The pathophysiology of schizophrenia has been linked with chronic in ammation, which stimulate in ammatory markers like CRP and IL-6 [38]. Both CRP and IL-6 have important roles in the in ammatory processes and CRP has been widely considered as a state marker along with other cytokines like TNFalpha. CRP is an acute phase protein and produced by hepatocyte cell when stimulated by in ammatory markers including IL-6. Under normal conditions, CRP doesn't cross the blood-brain barrier. Increasing serum level of CRP may increase the permeability of blood-brain barrier by affecting the function of tight junction which facilities the entry of pro-in ammatory cytokines and CRP itself into the central nervous system. This would support the potential role of CRP in the pathophysiology of schizophrenia, Moreover, studies based on cell culture indicate that CRP can induce a pro-in ammatory state in microglia, thus suggesting that CRP may be linked to neuro-in ammation in the central nervous system [39][40][41].

Conclusion
To the best of our knowledge, this is the rst study that compares the in ammatory markers of patients with treatment resistant schizophrenia with a control group. The result suggest that there is a higher level of hsCRP and IL-6 in patients with schizophrenia compared to their control groups.While the crosssectional design and relatively small sample size are potential limitations of this study, taken together with ndings from developed countries, underscore the fact that in ammation plays an important role in the pathogenesis of schizophrenia globally.

Participants
Eighty-two patients with Schizophrenia according to the Diagnostic and Statistical Manual of Mental Disorders Fourth edition (DSM-IV) [42] were recruited from Amanuel Specialized Mental Hospital, the main national institution for the care of those with severe mental illness in the country, between January 2015 and March 2016. Participants were recruited as part of a clinical trial, the MINOS (MINOcycline for Schizophrenia) Trial [43] (Clinicaltrials.gov identi er: NCT01809158) and were at least 18 years of age with a con rmed diagnosis of schizophrenia using standardized evaluation (Operational Criteria for Research-OPCRIT) administered by a psychiatrist and a recent onset of illness (duration under ve years).
They had at least moderately severe illness with a Positive and Negative Syndrome Scale score of 75 or more and have been on antipsychotic treatment for at least four weeks with little response. They had no gross physical or neurological co-morbidity and substance abuse. Because of the nature of the trial, women of child bearing age were excluded. For this study on in ammatory markers, the rst 82 participants who provided blood sample were included. Fifty healthy control subjects were recruited in the same geographic area as summarized in Table 1. The full medical history of study participants were evaluated to exclude any existing systemic diseases that may affect the parameters of in ammatory markers, and all had no detected medical illness.
Assessments A demographic questionnaire that included age, sex, and body mass index (BMI) was completed by trained Clinical Nurse. Obesity estimations were made from the BMI: Underweight (< 18.50), Normal (18.5-24.99), Overweight and Obese (25 and above). The severity of the symptoms of schizophrenia was assessed with the PANSS, a widely used semi-structured instrument in schizophrenia research. The Total PANSS score was used in this study. The PANSS has been used successfully in a clinical trial in Ethiopia [44] Serum IL-6 and hsCRP measurement Experienced phlebotomists collected 4-5 ml of blood from both study participants and controls after 8-10 hours of overnight fasting. Centrifugation was done at 5000 rpm on clotted blood for 10 minutes, and aliquots were stored at -80°c until analysis performed at the Clinical Chemistry laboratory of the Ethiopia Public Health Institute. Both hsCRP levels and IL-6 were measured by turbidimetric and Electrochemiluminescence immunoassay method using Cobas Integra 400 Plus and Cobas e411 (Roche Diagnostics GmbH, Mannheim, Germany) respectively. The lowest detectable limit for hsCRP was 0.1 mg/L and for IL6 was 1.0 pg/ml. IL-6 values of > 7 pg/ml were considered high while the corresponding value for hsCRP was > 1 mg/L.

Statistical analysis
Statistical analyses were done using SPSS Version 22.00 (SPSS Inc. Chicago, IL, USA). Simple descriptive and comparative analyses were carried out initially. For more advanced analysis, linear regression was used after evaluating the normality of the distribution of both hsCRP and IL-6. hsCRP was not normally distributed and thus was log-transformed. Gender, age, and BMI, factors previously reported to be associated with hsCRP and IL-6, were considered confounders and adjusted for in the linear regression model. All value of p < 0.05 were considered signi cant. Written informed consent was obtained from each participant after detailed explanation of the objectives of the study, risk, and bene ts. Guardians or next of kins' informed consent was obtained for those individuals that do not have the capacity to consent. All methods were carried out in accordance with the Declaration of Helsinki.

Consent for publications
Not applicable Availability of data and materials The datasets used and analyzed during the current study are available from the rst author Feyissa Challa on reasonable request.

Competing Interests
The authors declare no con ict of interest.

Funding
This study was funded by The Stanley Medical Research Institute (SMRI).
Authors' contributions FC and AF were the principal investigators of the study. FC, AF, and YW took the leading role from conception, design, and supervising the data collection process up to the nal analysis and preparation of the manuscript. DS and DK participated in reviewing the method part and provided critical comments. MA contributed to the writing of the manuscript. MM and MA coordinated the project and were responsible for the data acquisitions. MS and TG contributed to the methodology and laboratory analysis. All authors read and approved the nal manuscript.

Figure 1
Mean serum level of hsCRP and IL-6 in schizophrenia and control group. P values derived after adjustment for gender, age, and BMI