Neuropsychiatric manifestations of systemic lupus erythematosus: Iranian experience

Aims: To evaluate the prevalence and characteristics of different neurological and psychiatric presentations in patients admitted to hospital with systemic lupus erythematosus (SLE). Materials and Methods: In this retrospective hospital-based study, we examined the medical records of patients with SLE who were referred to the hospitals affiliated to Shiraz University of Medical Sciences from 1995 to 2005. All patients of SLE who had clinical neurological or psychiatric features were included in this study. The patients´ demographic data, findings on general examination, neuropsychiatric manifestations, and the results of laboratory investigations and imaging studies were recorded. Clinical manifestations were classified according to the American College of Rheumatology (ACR) case definitions. Results: Of the 407 study patients, 11.3% had neuropsychiatric complications. The most frequent findings were seizure (63%), headache (60%), and decreased level of consciousness (50%). Cerebrovascular disease (28.3%), seizure disorder (26.5%), and acute confusional state (19.6%) were the most prevalent syndromes. Conclusion: The prevalence and nature of different neurological presentations of SLE in Iranian patients has some similarities to that seen in other populations, as well as some differences. Ethnic and environmental factors may contribute to these differences.


Introduction
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with a variable clinical presentation and a variable course.It involves almost every organ of the body.Diverse psychiatric and neurological disorders may complicate the course of the disease.[3] These manifestations can occur frequently, may be highly variable, and are oft en incapacitating and sometimes life threatening.
electrodiagnostic studies are all used to study neurological involvement in SLE, but no single tool has proven to be completely specifi c and sensitive. [12]cause very little is known about the frequency of the neurological and psychiatric manifestations of SLE in Iran, this study was done to evaluate the frequency and characteristics of these manifestations in Iranian patients with SLE.

Materials and Methods
This is a descriptive retrospective study done to fi nd out the frequency and characteristics of the neurological and psychiatric manifestations of SLE patients referred to the hospitals affi liated to the Shiraz University of Medical Sciences during a 11 year period from 1995 to 2005.The clinical records of 407 patients discharged from hospital aft er a diagnosis of SLE were reviewed.Complete confi dentiality was maintained.The hospital fi les of the patients were examined for mention of the presence of neuropsychiatric symptoms and signs.Inclusion criteria were fulfillment of the ACR criteria for SLE and presence of a syndrome or disease with clinical neurological or psychiatric features likely to be causally related to SLE.The patients who had any clinical or paraclinical fi ndings which were consistent with any clinical entity other than SLE were excluded from the study.These exclusion criteria included positive CSF cultures or any other evidence of neurological infections, diabetes mellitus, hyperlipidemia, positive family history of premature atherosclerosis or other major risk factors of young adult stroke, intravenous drug abuse, neurological side eff ects of drugs, hypertensive or uremic encephalopathy, electrolyte disturbances, complications of renal transplantation, and positive HIV test.
The patients' demographic data, duration of SLE, general manifestations of the disease, and the neurological and psychiatric manifestations of SLE were studied.The results of paraclinical data such as antinuclear antibodies, anti-doublestrand DNA, CSF analyses, electroencephalography, nerve conductions studies and electromyography, and neuroimaging studies were also recorded from the medical fi les.
Cognitive dysfunction was considered according to clinical fi les of the patients.andcomprehensive psychometric batt eries were not conducted for most of the patients.
The patients were categorized by a neurologist (A.B.H) into defi nite neurological syndromes according to American College of Rheumatology (ACR) criteria. [13] used SPSS 10.0 soft ware for analysis of the data.The Chi square test, the t-test, and Fisher's exact test were used as appropriate.Statistical signifi cance was assumed at P ≤ 0.05.

Results
Among the 407 patients, 59 had neurological or psychiatric symptoms.Thirteen patients were excluded due to the presence of one of the following: uremic encephalopathy resulting from renal involvement by SLE (seven patients), hypocalcemia, bacterial meningitis, thrombocytopenia-induced intracranial hemorrhage, hypertension-induced intracranial hemorrhage, cardiac embolism originating from a prosthetic valve, and idiopathic epilepsy (each in one patient).The remaining 46 patients (11.3% of the patient population) were considered to have neuropsychiatric lupus.Among these 46 patients, 42 (91.3%)were female and four (8.7%) were male.Thirty-one patients (67.4%) were from urban areas and 15 patients (32.6%) from rural areas.The age at fi rst manifestation of SLE ranged between 16 and 47 years, with a mean of 24 years.The mean age at fi rst manifestation of neuropsychiatric lupus was 27 years, with a mean interval of 4 years between diagnosis of SLE and manifestation of neuropsychiatric symptoms/signs.In 11 patients (23%), neurological or psychiatric manifestations were the initial features of SLE.
Among the neuropsychiatric manifestations in 46 patients, general seizures were the most frequent fi nding, followed by headaches, decreased level of consciousness, and weakness (including mono/hemi/para/quadriparesis) [Table 1]. Figure 1 shows the prevalence of diff erent clinical syndromes among our patients according to ACR criteria.As can be seen, cerebrovascular diseases, seizure disorders, and acute confusional state were the 'big three' among our patients.
In this study, 92% of patients had high erythrocyte sedimentation rate (ESR).Antinuclear antibody (ANA) and anti-double-strand DNA were positive in 94% and 78% of patients, respectively.C3 and/or C4 complement were decreased in 68% of patients.
Thirty-nine CT scans and 33 MRIs were conducted in all.Among the CT scans of brains, 23 (59%) revealed no abnormality.In seven (18%), brain atrophy was present, fi ve scans (13%) showed diff use brain edema, and four scans (10%) showed large infarcts.Among the 33 available brain MRIs, 12 (36%) studies were normal, small lesions were seen in 17 (52%) brain MRIs, and large lesions were observed in four (12%) MRIs.All of the abnormal MRIs revealed hyperintense lesions in T2-weighted images but only in 9 (43%) MRIs were corresponding hypointense lesions in T1-weighted images seen.These characteristics were compatible with ischemic/ demyelinating lesions.There were 22 CSF analyses performed in the study subjects.Fift y-fi ve percent of the CSF studies were completely normal, 27% revealed pleocytosis ( segment predominant in half and lymphocytic predominant in half), 18% showed increased protein content, and 2% showed decreased glucose levels.
Nerve conduction studies and electromyography revealed myogenic patt ern in two patients; there was myositis and a neurogenic patt ern with conduction block in one patient with the Guillain-Barré syndrome.

Haghighi and Haza: Neuro-psychiatric lupus in Iran
The mortality rate due to direct neuropsychiatric complications of SLE in our 46 patients was 22%.The major contributor to mortality was large cerebrovascular accidents, which were seen in 69% of the patients who died.

Discussion
This is a retrospective hospital-based study, with data collected from clinical fi les of patients.Therefore, patients with noncatastrophic neuropsychiatric manifestations of SLE were not included in our study, which had some infl uences on our results.
[3] This could be because of the retrospective, hospitalbased nature of our study, which failed to detect the cases of SLE with neurological manifestations that did not require hospitalization.The high mortality rate among our patients is also a consequence of the nature of selection.
[16] Seizures can occur at any time during the course of SLE and all kinds of seizures may be seen in these patients (including generalized tonic-clonic, simple focal, and complex partial seizures).They may occur in isolation or along with the other neurological manifestations. [14,17,18]xty percent of our patients had headache as part of their neuropsychiatric event.[24] For example, the accumulated frequency of cerebrovascular disease (infarcts, transient ischemic att acks, and hemorrhages) was 23.1% according to Futurell and Milikan. [22]The possible pathomechanisms are premature atherosclerosis, vasculitis, and coagulopathy. [23]ychotic att acks are relatively common in SLE and sometimes recur.Psychotic att acks in SLE must be diff erentiated from corticosteroid-induced psychosis. [25]vement disorders which occur in SLE include chorea (which was seen in two of our patients), [26] ballismus, [27] and parkinsonism. [28]rebral venous thrombosis presents with severe headache, papilledema, and/or hemorrhagic infarcts and was seen in one of our patients.Due to the availability of magnetic resonance venography the diagnosis of venous sinus thrombosis in patients with lupus is relatively easy these days. [29]though infl ammatory polymyositis is not included in the ACR nomenclature system, we considered it as a separate entity in this study so as to cover all the neurologic complications of SLE.
There are several series which have used the ACR nomenclature Figure 1: Frequency of the observed of SLE in the studied patients according to criteria of the American College of Rheumatology. [13]nfl ammatory polymyositis is not an ACR entry Haghighi and Haza: Neuro-psychiatric lupus in Iran for categorization of neuropsychiatric manifestations of SLE.
In Brazilian and Thai series, [15,16] seizure disorder was the most common syndrome (whereas it was the second most common one in our study).Cognitive dysfunction, one of the least common syndromes in our study, was one of the most common ACR syndromes in American, [30] Finnish, [31] and Italian [32] studies.This variation may be due to not only geographic and ethnic factors but also to the variations in the selection criteria in the diff erent studies.
The current study has some shortcomings, the major limitations being its retrospective and hospital-based nature.Also, we had very strict exclusion criteria; If both SLE and its complications contributed to the neuropsychiatric manifestations in any one patient, exclusion of that patient may have induced a selection bias.Also considering a patient in just one ACR category might have been arbitrary.
In summary, our study restates the great diversity of neuropsychiatric manifestations in SLE.Based on our experience in this study, we stress the need for conducting a prospective multicentric study on neuro-lupus, with inclusion of both inpatients and outpatients.