Diagnostic value of single-photon–emission computed tomography in severe central nervous system involvement of systemic lupus erythematosus: A case-control study

Authors

  • Xuan Zhang,

    1. Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China
    Search for more papers by this author
  • Zhaohui Zhu,

    1. Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China
    Search for more papers by this author
  • Fengchun Zhang,

    Corresponding author
    1. Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China
    • Department of Rheumatology and Immunology, Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing 100730, China
    Search for more papers by this author
  • Huijun Shu,

    1. Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China
    Search for more papers by this author
  • Fang Li,

    1. Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China
    Search for more papers by this author
  • Yi Dong

    1. Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing, China
    Search for more papers by this author

Abstract

Objective

To evaluate the diagnostic value of single-photon–emission computed tomography (SPECT) in severe central nervous system (CNS) involvement of systemic lupus erythematosus (SLE).

Methods

Forty-three patients with SLE, including 22 with CNS-SLE and 21 with non–CNS-SLE, underwent SPECT and magnetic resonance imaging (MRI) examinations. SPECT was repeated 1–2 months after treatment in patients with abnormal findings.

Results

SPECT and MRI abnormalities were detected in 20 (90.9%) and 10 (45.5%) of the 22 patients with CNS-SLE, respectively (P < 0.01). For 4 patients with cerebral infarction or hemorrhage, SPECT was equally as sensitive as MRI (100%). For the patients with CNS-SLE with diffuse presentations, SPECT was more sensitive than MRI in revealing abnormalities (16 [88.9%] of 18 patients versus 6 [33.3%] of 18 patients; P < 0.01). In 19 (95.0%) patients, the abnormal SPECT finding manifested as moderate to severe perfusion defect (15 in frontal lobe, 11 in parietal lobe, 11 in basal ganglia, 3 in temporal lobe, and 17 in multiple regions). Although mild perfusion defect was also detected in 4 (19.0%) of the patients with non–CNS-SLE, it only involved a single region and spared the frontal and parietal lobes. Repeated SPECT after treatment showed that perfusion defect had improved significantly or even disappeared in 11 (84.6%) of 13 patients with diffuse CNS-SLE with abnormal findings before treatment.

Conclusion

Moderate to severe perfusion defect in SPECT involving multiple regions, especially in the frontal and parietal lobes and basal ganglia, in patients with lupus suggests CNS involvement. SPECT is more sensitive than MRI in revealing damage in diffuse CNS-SLE, and is useful in followup, especially for monitoring disease severity and guiding treatment.

INTRODUCTION

Central nervous system (CNS) involvement is a common complication in systemic lupus erythematosus (SLE). The incidence rate ranges from 24% to 51% depending largely on study design, particularly patient selection and the decision of which neurologic features to include (1–3). Although the incidence rate is high, the diversity of clinical presentations, the multiple potential etiologies, and the absence of sensitive and specific tests often make diagnosis difficult. The mortality rate can be very high, ranging from 7% to 40% according to the literature (1–3). Prompt diagnosis and treatment could significantly alleviate the disease and improve prognosis.

The most commonly applied complementary tests, such as computed tomography and magnetic resonance imaging (MRI), are static image techniques and are not sensitive in reflecting the pathophysiologic changes in CNS-SLE. Cerebral angiography is often declined by patients due to its risk and low sensitivity (4). It is imperative to use more sensitive tests for better diagnosis, especially for patients with nontypical neuropsychiatric manifestations or those at an early stage of the disease.

Since 1992, several groups have reported that single-photon–emission computed tomography (SPECT) is a sensitive test in detecting active CNS-SLE, and that the sensitivity could be as high as 88% (4–12). These groups proposed that SPECT analysis may provide additional information on potential CNS involvement in patients with lupus. The purpose of the present case-control study was to compare the SPECT and MRI findings in patients with CNS-SLE versus those with non–CNS-SLE, and to investigate the diagnostic value of SPECT in CNS-SLE.

PATIENTS AND METHODS

Twenty-two consecutive inpatients with CNS-SLE in Peking Union Medical College Hospital were enrolled in this study, and 21 patients with non–CNS-SLE who were hospitalized at the same time were randomly selected as controls. All patients fulfilled ≥4 of the 1997 American College of Rheumatology (ACR) revised criteria for SLE (13). Both a rheumatologist and a neurologist diagnosed patients as having CNS-SLE due to significant and unequivocal change in neurologic or psychiatric function, identified by history, physical examination, and laboratory or radiographic tests, and further proved by clinical course and response to treatment, as required by the ACR criteria for CNS-SLE (14, 15). For example, cognitive dysfunction was diagnosed if significant deficits in any or all of the following cognitive functions were observed in the patients: simple or complex attention, reasoning, executive skills (e.g., planning, organizing, sequencing), memory (e.g., learning, recall), visual-spatial processing, language (e.g., verbal fluency), and psychomotor speed. A neurologic event was excluded if ≥1 criteria were present: hypertension (diastolic blood pressure >120 mm Hg), hypoxia (PaO2 <50 mm Hg), uremia (blood urea nitrogen >35.5 mmoles/liter), serious electrolyte imbalance, or culture-proved CNS infection. In accordance with the modified ACR nomenclature and case definitions (15), these 22 patients with CNS-SLE were further subdivided into diffuse presentations (n = 18), which included cognitive dysfunction, severe depression, acute confusional state, aseptic meningitis, movement disorder, and epilepsy, and focal presentations (n = 4), which included cerebral infarction and cerebral hemorrhage. All patients with CNS-SLE and controls received corticosteroids and/or cyclophosphamide treatment. For patients with positive antiphospholipid antibody, low-molecular heparin was also added if not contraindicated. Consent to participate in the study was obtained from all patients or their family. The research was approved by the Hospital Ethics Committee. All patients underwent SPECT and MRI, and 16 patients with CNS-SLE with abnormal SPECT findings at enrollment repeated SPECT 1–2 months after treatment.

In addition to neurologic assessment, each patient was evaluated for other evidence of SLE disease activity using a modification of Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) scores (16). Active disease was defined as a modified SLEDAI score ≥8 (CNS disturbance was excluded from the modified SLEDAI scores).

Single-photon–emission computed tomography.

All patients received an intravenous injection of 99mTc-ethylenedicysteinate dimer (99mTc-ECD) at a dose of 1,110 MBq (30 mCi), and the scan was initiated 20 minutes later. The SPECT camera (Helix-HR; Elscint, Haifa, Israel) was equipped with double rotating heads. The images were scanned with a 64 × 64–pixel matrix, without zoom, with 3° angles comprising 180° for each head for a total of 120 images, with a scanning time of 20 seconds per angle and a total time of 20 minutes. The images were fed into a computer and reconstructed by applying a recovery filter with a cut frequency of 0.25 and a threshold of 4 in the axial, sagital, and coronal views with a thickness of 0.9 cm, presented later on the screen for interpretation. The images were analyzed in a blinded manner by 2 independent experienced physicians of nuclear medicine who did not have access to any demographic or clinical data. The examination was considered normal when there was a symmetrical and homogeneous concentration of 99mTc-ECD in both cerebral hemispheres, and it was considered abnormal when there was asymmetry in the distribution of the 99mTc-ECD in the whole hemisphere or in some lobe and/or when there were focal areas of low tracer concentration.

Magnetic resonance imaging scan.

MRI was performed with a 1.5T MRI system (Philips Medical Systems, Best, The Netherlands). Conventional T1-weighted spin-echo imaging, fluid-attenuated inversion recovery (FLAIR)-weighted imaging, and dual (proton density and T2-weighted) imaging were performed on all patients. Images were also read in a blinded manner by 2 experienced neuroradiologists who did not have access to any clinical data. T1-weighted imaging was performed using a spin-echo sequence with an echo time of 20 msec, a repetition time of 600 msec, and a scanning time of 3 minutes and 8 seconds. Dual fast spin-echo imaging was performed using an echo time of 23/120 msec, a repetition time of 2,500 msec, an echo train length factor of 10, and a scanning time of 2 minutes and 35 seconds. FLAIR fast spin-echo imaging was performed using an echo time of 120 msec, a repetition time of 8,000 msec, an echo train length of 16, and a scanning time of 3 minutes and 28 seconds. All sequences consisted of 22 axial slices with a 6-mm thickness, a 222-mm2 field of view, a 0.6-mm interslice gap, and a 256 × 256–pixel matrix, with an 80% acquisition percentage.

Statistical analysis.

The significance of the differences in the number of patients with positive baseline characteristics, laboratory findings, and SPECT and MRI abnormalities between study groups was determined using Fisher's exact test. The Statistical Package for the Social Sciences (SPSS) version 11.0 (SPSS, Chicago, IL) was used to analyze the data. In all tests, all probability values were 2-sided, and P values less than 0.05 were considered as significant difference.

RESULTS

Baseline characteristics and laboratory values of the patients with CNS-SLE and controls are listed in Table 1. The sex ratio, age, and mean disease duration were compatible between the patients with CNS-SLE and the non–CNS-SLE controls. Although they did not reach statistical significance, the positive rates of anti–ribosomal P and anticardiolipin antibody (aCL) as well as active disease in the CNS-SLE group were relatively higher than in the control group.

Table 1. Baseline characteristics and laboratory data in patients with CNS-SLE and the non–CNS-SLE controls*
CharacteristicCNS-SLE (n = 22)Non–CNS-SLE (n = 21)P
  • *

    Values are the number (percentage) of patients unless otherwise indicated. CNS-SLE = central nervous system systemic lupus erythematosus; anti-dsDNA = anti–double-stranded DNA; aCL = anticardiolipin antibody.

Men/women2/202/19
Age, mean ± SD years32 ± 1735 ± 22
Disease duration, mean ± SD years3.8 ± 2.43.3 ± 3.1
Nephritis18 (81.8)14 (63.6)0.36
Leukopenia/thrombocytopenia10 (45.5)10 (47.6)1.0
Hypocomplementia16 (72.7)12 (57.1)0.35
Elevated anti-dsDNA15 (68.2)12 (57.1)0.54
Anti–ribosomal P6 (27.3)2 (9.5)0.24
aCL8 (36.4)5 (23.8)0.51
Active disease stage20 (90.9)14 (66.7)0.07

CNS manifestations improved significantly or even disappeared in 17 of 18 patients with diffuse CNS-SLE 1–2 months after treatment, while no significant improvement was observed in 4 patients with focal CNS-SLE. Cerebrospinal fluid (CSF) examinations were routinely carried out in all patients with CNS-SLE and repeated 1–2 months after therapy. Before and after treatment, the mean ± SD CSF protein level was 148.3 ± 80.5 mg/dl versus 65.4 ± 48.2 mg/dl (P < 0.001), the mean white blood cell count was 28.4 ± 19.3/mm3 versus 6.3 ± 3.4/mm3 (P < 0.001), and the mean pressure was 204.5 ± 165.4 mm H2O versus 129.9 ± 100.3 mm H2O (P > 0.05).

SPECT abnormalities were detected in 20 of 22 patients with CNS-SLE, with a sensitivity of 90.9%, whereas MRI abnormalities were detected in only 10 (45.5%) of these 22 patients (P < 0.01) (Table 2). Nineteen (95.0%) of 20 abnormal SPECT findings manifested as moderate to severe perfusion defect, most commonly affecting multiple regions (17 cases), especially in the frontal and parietal lobes and the basal ganglia (15 in the frontal lobe, 11 in the parietal lobe, 11 in the basal ganglia, and 3 in the temporal lobe). For the 4 patients with focal presentations such as cerebral infarction and cerebral hemorrhage, SPECT was equally as sensitive as MRI (100%). For the patients with CNS-SLE with diffuse presentations, SPECT was more sensitive than MRI in revealing abnormalities (16 [88.9%] of 18 patients versus 6 [33.3%] of 18 patients; P < 0.01). In the 21 patients with non–CNS-SLE, mild perfusion defect was also detected in only 4 patients (19.0%), which only involved a single region and spared the frontal and parietal lobes. The specificity of SPECT for detecting CNS involvement in SLE was 89.5%. None of the 21 patients with non–CNS-SLE had any abnormal findings on MRI scanning. The most common MRI abnormalities were infarction, hemorrhage, leukomalacia, and diffuse alteration of white matter.

Table 2. SPECT and MRI abnormalities in patients with CNS-SLE*
CNS-SLETotal no.SPECTInvolved regionsMRI
  • *

    Values are the number of patients unless otherwise indicated. SPECT = single-photon–emission computed tomography; MRI = magnetic resonance imaging; CNS-SLE = central nervous system systemic lupus erythematosus; F = frontal lobe; P = parietal lobe; T = temporal lobe; B = basal ganglia; M = multiple regions.

  • P < 0.01.

Diffuse, no. (%)1816 (88.9)14F, 10P, 1T, 9B, 15M6 (33.3)
 Cognitive dysfunction766F, 3P, 4B2
 Severe depression222F, 1P, 1B0
 Acute confusional state433F, 2P, 1B1
 Aseptic meningitis222F, 1P, 1T, 1B1
 Movement disorder111F, 1P0
 Epilepsy222P, 2B2
Focal, no. (%)44 (100)2F, 1P, 2T, 2B, 2M4 (100)
 Cerebral infarction331F, 1P, 1T, 2B3
 Cerebral hemorrhage111T1

SPECT examinations were repeated 1–2 months after treatment in 16 patients with CNS-SLE who had abnormal findings before treatment. Of these, 13 patients had diffuse CNS-SLE. Repeated SPECT showed significant improvement or even disappearance of perfusion defect in 11 (84.6%) of these 13 patients (Figure 1), and only one slight improvement in 1 of 3 patients with focal CNS-SLE.

Figure 1.

Single-photon–emission computed tomography images of a patient with systemic lupus erythematosus with diffuse central nervous system involvement. Serious perfusion defects could be seen in both frontal lobes (arrows) before treatment. A, axial view; B, coronal view; C, sagital view, which improved after treatment; D, axial view; E, coronal view; F, sagital view.

DISCUSSION

CNS involvement is a common complication of lupus. The onset of CNS-SLE is often acute and fulminating. Mortality rate has been high in the past due to lack of awareness and early diagnostic criteria of this complication. The pathogenesis remains unclear. Small vessel vasculopathy mediated by immune complexes, antiphospholipid antibody, anti–ribosomal P antibody, and other autoantibodies and vasculitis were all suggested in the pathogenesis of CNS-SLE (2, 3, 17, 18). Our present study demonstrated that the positive rates of anti–ribosomal P antibody and aCL were higher in patients with CNS-SLE than in those with non–CNS-SLE. This finding is consistent with our previous report and is similar to reports from other groups (1, 19–21).

Traditional tests, such as computed tomography and MRI, are less sensitive and are only positive in severe or advanced stages of CNS-SLE. Therefore, it is imperative to use other tests that are more sensitive for prompt diagnosis of nontypical or early-stage CNS-SLE. SPECT is a technique that obtains information on cerebral vascular perfusion using 99mTc as a tracing reagent. As a functional imaging technique, SPECT could be used to evaluate regional cerebral blood flow through calculation of tracer concentration. Similar to previous reports (8–12), our study showed that SPECT was more sensitive (90.9%) in detecting CNS abnormalities than MRI (45.5%), especially in patients with diffuse involvements. Compared with patients with non–CNS-SLE, the most common finding in patients with CNS-SLE was moderate to severe perfusion defect involving multiple regions, especially the frontal and parietal lobes and basal ganglia. Also, in most patients with CNS-SLE, especially those with diffuse presentations, repeated SPECT examinations after prompt and effective treatment showed significant improvement or even disappearance of perfusion defect. This result indicated that most patients with SLE with diffuse CNS involvement during the early active stage have reversible impairment and might have a good response to prompt treatment. It also indicated that SPECT, as a technique revealing functional alteration of CNS, could detect abnormalities long before the development of irreversible impairment. SPECT proved to be a very sensitive method, and therefore was useful in monitoring disease severity and effectively guiding clinical treatment.

In this study, SPECT also showed mild abnormalities in 4 of the 21 patients with non–CNS-SLE (19.0%). It is noteworthy that none of these patients had disease that was very severe or involved multiple regions, as shown in most of the patients with CNS-SLE, and none of them had any abnormal findings on MRI scanning. Although none of these patients had any clinical manifestation of CNS involvement, we could not completely exclude the possibility of insidious subclinical CNS-SLE. Also, we tended to treat these patients more aggressively if they also had nephritis.

SPECT was proved to be a sensitive method in detecting CNS-SLE, especially in patients with diffuse CNS involvement. CNS-SLE was suggested if SPECT showed moderate to severe perfusion defect involving multiple regions, especially in the frontal and parietal lobes and basal ganglia. As a functional imaging technique, SPECT was more sensitive than MRI in revealing damage and was also useful in followup, especially for monitoring disease severity and guiding treatment.

Ancillary