Brain perfusion imaging with voxel-based analysis in secondary progressive multiple sclerosis patients with a moderate to severe stage of disease: a boon for the workforce

Multiple sclerosis (MS) is an inflammatory demyelinating disease of the central nervous system (CNS) that affects myelin, oligodendrocytes, and axons, and eventually results in neuronal loss and progressive neurologic disability [1].

In total, four disease courses have been distinguished: relapsing remitting (RR), primary progressive (PP), secondary progressive (SP) and also progressive-relapsing (PR) [2]. The RR type is constituted as the common form of the disease. About two-thirds of patients with RR-MS change in the SP phase where neurologic disability has a progressive pattern between attacks [2, 3].

In the 1930s, Putnam explained the evidence of vascular occlusion in MS histopathology for the first time, and believed that vascular occlusion and inflammation were involved as antecedent processes in demyelinating disease evolution [4]. Wakefield et al. later reported similar findings and showed fibrin deposition and vessel thrombosis without obvious cellular infiltration, which suggested that the presentation of thrombosis in small veins and capillaries supported the possible role of ischemic process in MS disease [5].

Luke et al., who used single photon emission computed tomography (SPECT), showed hypoperfusion of cerebral blood flow (CBF) in the gray matter (GM) of the frontal lobe of patients with progressive MS compared with cerebral perfusion in control patients, although the study reported normal perfusion in patients with RR-MS [6]. Our previous study has similar results and normal brain perfusion of MS patients in the early stage and without significant disability [7]. In another study, Brooks et al. showed that there was decreasing CBF and oxygen utilization in both the white matter and the cortical GM in the MS patients compared to the healthy control group [8].

Some claims exist about the efficacy of HBOT in MS patients. Fisher et al. [9–11], in a double-blind controlled study, showed significant but transient improvements in mobility, fatigability, and walking in MS patients receiving HBOT. Warren et al., showed that HBOT resulted in an augmentation of experimental allergic encephalomyelitis, an animal model of MS [12]. Although different studies had been carried out [13, 14], the assessment of such therapy with functional modalities is scare in the literature. What’s more, recent clinical trial researches along these lines and provided evidence that HBOT can get better the symptoms and life quality of in some neurological disorders [15–25].

We aimed to evaluate cerebral perfusion in MS patients with a moderate to severe stage of disease. Besides, we provided firm appraisal of the HBOT effect on brain activity as assessed by SPECT imaging supplied by quantitative methods and its possible correspondence on clinical status of MS patients.

Of the 25 patients, 2 (8 %) have a normal SPECT and 23 (92 %) have abnormal brain perfusion SPECT studies. Brain perfusion SPECT in the abnormal group included: 2 (8 %), 20 (80 %) and 1 (4 %), which showed mild, moderate and severe perfusion scans, respectively (Table 1).

In total, 16 patients were treated with HBOT (Table 2). The brain perfusion scans that were obtained before and after therapy showed a significant improvement in perfusion scan results (P <0.05) (Figs. 1, 2, 3 and 4). However, the HBOT did not demonstrate a significant improvement in clinical subjective and objective evaluation of the patients (P >0.05).

Patient no.	EDSS	Interval time (day)	Disease duration (year)	Pre-treatment scan	Post treatment scan
1	4	90	9	Moderate bifrontoparital	Mild bifrontoparital
2	5	30	8	Severe Rt hemisphere	Mild Rt hemisphere
3	3	30	4	Moderate diffuse cortical	Mild diffuse cortical
4	6	50	10	Mild Rt hemisphere	Normal
5	2	240	7	Moderate Rt hemisphere	Moderate Rt hemisphere
6	2	52	2	Normal	Normal
7	3	60	6	Moderate diffuse (L > R)	Moderate diffuse (L > R)
8	3	32	2	Moderate Rt temporal	Moderate Rt temporal
9	3	50	3	Moderate bi temporal	Mild bitemporal
10	6	70	13	Moderate Rt posterior temporal	Normal
11	6	65	11	Moderate bi temporoparital and caudate nucleus	Normal
12	8	26	7	Moderate Rt occipital	Moderate Rt occipital
13	3	60	7	Moderate Rt occipital	Moderate Rt occipital
14	8	60	1	Moderate Lt occipital and putamen	Mild Lt occipital and putamen
15	4	40	5	Moderate bifrontoparital and diffuse subcortical	Normal
16	2	60	5	Moderate Lt occipital	Normal

EDSS expanded disability status scale

Obtained hypoperfusion regions in a patient with multiple sclerosis compared with healthy control group through SPM was mapped on T1 MRI transaxial images (Fig. 5).

The study showed a significant association of severity of perfusion impairment with disease duration and also with EDSS (P <0.05).

The time interval between the two before- and after-treatment SPECT examinations was 30–180 days (mean time interval, 64.43 ± 38.66 days).

The SPM results table plus its corresponding glass brain of a patient with two subsequent HBOT sessions were presented. There are three follow up results indicating therapy was effective and also based on the obtained figures the regions with significantly reduced perfusion became smaller and smaller and finally in the last scan there was no significant hypoperfused region where the results are in alignment with the visual physician assessment (Tables 3 and 4) (Fig. 6).

p (FWE-corr)	T	Z	Coordinates x, y, z {mm} max. peak	Brain area
0.009429	9.117918	4.899239	26, 46, −6	Right superior frontal gyrus, orbital part
0.021354	8.183681	4.672336	−0, −54, 34	Left precuneus
0.02832	7.874513	4.590642	−38, −10, −18	Medial temporal lobe
0.028749	7.858227	4.586238	−46, −8, −20	Left temporal lobe
0.03078	7.784257	4.566107	−40, −18, −24	Left inferior temporal gyrus

p (FWE-corr)	T	Z	Coordinates x, y, z {mm} max. peak	Brain area
0.02072	8.219753	4.681639	26, 46, −6	Right superior frontal gyrus, orbital part
0.026836	7.935467	4.607033	−36, −10, −20	Medial temporal lobe

A group semi-quantitative analysis using SPM was performed which clearly showed a significant hypoperfusion in the right medial orbitofrontal cortex (Table 5, Fig. 7).

p (FWEcorr)	T	Z	Coordinates x, y, z {mm} max. peak	Brain area
0.003843	7.936576	4.986516	18, 36, −8	Right medial orbitofrontal cortex

In this study, we investigated cerebral perfusion by SPECT in SP-MS, and evaluated the effects of HBOT in the CBF of these patients. Our study revealed abnormal cerebral perfusion in 92 % of SP-MS patients and for the first time showed that cerebral perfusion significantly improved after HBOT. In addition, it depicted a remarkable association between brain perfusion defects and clinical status. Such findings were also seen on voxel-based analysis of whole-brain data using SPM analysis.

Statistical parametric mapping is a neuroimaging analysis tool which evaluates the statistical significance of cerebral blood flow (CBF) alterations on a voxel by voxel basis, which is more reliable than conventional ROI analysis due to the inherent subjectivity of these methods [28]. SPM was used to identify regions with decreased CBF in each patient compared to the control group [15].

Perfusion studies of MS have different results that may be due to various perfusion measurement techniques and stages of disease [6, 7]. In our previous study, we investigated cerebral perfusion in early RR-MS patients with mild or no disability, which determined normal cerebral perfusion in all patients [7]. Other studies showed different brain perfusion in MS patients particularly in those with a progressive type of disease [6].

By using an arterial spin labeling sequence in early RR-MS patients, Debernard et al. found reduced perfusion in deep GM and different cortical regions, including lingual gyrus, intracalcarine, insular and operculum cortex, temporal, parietal, frontal and occipital cortical regions [29]. In another study, Rashid et al. by arterial spin labeling, found hypoperfusion in several cortical regions of patients with primary and SP-MS but not in patients with RR-MS [30].

Ota et al. assessed the CBF of MS patients by arterial spin labeling and revealed reduced CBF in the bilateral thalami and right frontal region of the patients compared to the healthy volunteers [31]. Lyke et al. evaluated the CBF of MS patients by brain SPECT and showed that CBF was significantly reduced in the patients—particularly in those with the progressive type of MS [6]. Adhya et al. by using dynamic susceptibility contrast perfusion MRI, reported that the CBF was significantly decreased in all normal-appearing white matter (NAWM) regions in the MS patients, and that PP-MS patients showed a significantly lower CBF compared to RR-MS patients [32]. Law et al. determined that the NAWM region of patients with RR-MS had significantly lower CBF compared with the NAWM of control group [4].

Several direct and indirect mechanisms might have contributed to the change in cerebral perfusion of MS patients. The reduction of white matter perfusion may occur following a decrease in tissue metabolism, or it may be due to a primary vascular pathology that may impair cerebral flow [29]. Lower metabolic activity of the cerebral region secondary to neuronal loss (due to the demyelination process) might decrease CBF. Neuronal loss may originate from hypoperfusion, iron deposition, demyelination, or Wallerian degeneration [33]. Neuronal metabolic dysfunction secondary to inflammation with consequent mitochondrial dysfunction may decrease the demand for blood supply and eventually brain perfusion [29]. Axonal transection in the demyelination lesion of the white matter could result in anterograde and retrograde degeneration of axons that are associated with neuronal dysfunction and will consequently decrease GM perfusion [33]. An abnormal cerebral vasculature is another possible mechanism of the GM hypoperfusion [29]. The inflammatory process may cause microvascular damage by several mechanisms. Reaction of cytotoxic T cells with the specific antigen on endothelial surface may start a clotting cascade that eventually results in thrombus formation. Similarly, specific antibodies may react with their antigen on the endothelial cells and result in vascular damage via complement activation. Furthermore, inflammatory vascular edema may interfere with vascular blood flow following focal tissue swelling; whereas acute and chronic venous obliterations may develop as a result of exudation of inflammatory cells and intravascular fibrin deposition [34]. In addition, obliterative vasculitis might cause chronic ischemia resulting from modulation of the vascular tone and the CBF [33, 35]. An increase in the calcium concentration of extracts, neuronal and glial cells shown in MS lesions may be associated with an increase in the tone of blood vessels. Also, a high blood level of endothelin-1, a vasoconstrictor compound, is seen in MS [29, 35]. An abnormal brain cortical-subcortical circuit is considered to be the other mechanism of decreasing brain perfusion in MS patients, which theoretically may be caused by deep GM lesions like the thalamus even by a small lesion that is not obvious [31].

In addition, it revealed a noticeable association between brain perfusion abnormalities and clinical status. In our previous study, we investigated cerebral perfusion in early RR-MS patients with mild or no disability (EDSS score <1) [6]. In contrast, current study depicted significantly cerebral perfusion impairment in MS patients with moderate to severe disability may indicate cerebral perfusion defect increasing with worsening EDSS score. In line with our findings, Lycke et al. [5] evaluated cerebral perfusion in 19 patients with MS and visualized low cerebral perfusion in the frontal grey matter correlated with neurological disability and low perfusion in frontal grey and white matter correlated with impaired cognitive functions. Up to now, clinical and magnetic resonance imaging examinations considered as routine tools to predict disability progression and then appropriate management in MS, however no single parameter is expected to have sound prognostic significance due to complex pathogenesis of MS. What’s more, no conventional MRI procedures has shown strong correlation with long-term disability progression in MS patients, while more developed MRI techniques are currently being investigated with promising results in preclinical settings [35]. The relation between EDSS score and perfusion defect suggest that brain perfusion defect may be helpful parameter to monitor clinical disability and eventually disease progression in MS patient and brain perfusion SPECT can be complementary to other diagnostic modalities such as MRI and clinical examinations in disease surveillance and monitoring. We didn’t evaluated association between MRI lesions and perfusion defect because beyond the scope of our study and should be considered in follow up studies.

Based on the group analysis, right medial orbito frontal cortex was obtained as the significant perfused area among the multiple sclerosis patients, we offered that if the study extended to more patients with MS we may have more precise areas as defected regions. In addition, the aforementioned region was demonstrated on our prior studies that had an important role in status of olfactory. It may amend to do a follow up study in MS patients [33, 36].

Although our study didn’t show a significant clinical benefit from HBOT on the MS patient, the treatment significantly improved the patients’ cerebral perfusion. Another study showed similar results in a clinical setting [34]. In a small, randomized, placebo-controlled trial of HBOT, Martin et al. showed a transient mild improvement in CNS functions, including mobility, coordination, and bladder control, and a reduction in fatigability that is associated with changes of two or more points on the Kurtzke disability status scale [34]. A study by Barnes et al. showed that HBOT in 120 patients with chronic MS was not associated with improvement on the Kurtzke disability status scale [37]. Kindwall et al. didn’t show any benefit from HBOT in MS patients compared to the control group, except for a temporary improvement in bladder control which was reported by some patients [38]. Furthermore, two other main studies did not demonstrate a beneficiary from HBOT in MS patients [9, 38].

The clinical imaging dissociation may originate from the object that old lesions with severe tissue injury in a specific region of the CNS, such as the brainstem and spinal cord, are not affected by HBOT and consequently are not beneficial from the treatment. However, to our knowledge, no functional brain imaging research exists addressing this question in MS subgroups.

The mechanism of the HBOT effect is uncertain. One possibility is an immunosuppressive effect. Another possibility is that ischemic areas of perivenular plaques, which come from an associated venous occlusion, are transiently resolved by HBOT [8]. At the cellular level, HBOT seems to ameliorate neuronal and glia cells’ mitochondrial function and eventually cellular metabolism, reduces oxidative stress, augments neurotrophins and nitric oxide concentrations, and up-regulates axon guidance agents. The treatment may also induce neurogenesis and reduce apoptosis [38]. Another possible mechanism is that hyperoxia acts as a neuronal stimulant [39]. Hyperoxia decreases membrane conductance and affects ion channel closures, presumably decreasing outward (K) and/or inward (Cl) currents that eventually causing depolarization and stimulated to fire rate [39].

All of the mechanisms can promote neuronal activity and metabolism, and eventually increase cerebral perfusion [40–43]. However, there are ongoing studies on HBOT with promising results at least based on imaging findings indicating it will be a open argument in various neurological disorders [15, 16, 44, 45].

In summary, the data may depict diminished cerebral perfusion in SP-MS patients with a moderate to severe disability score and its remarkable association with clinical parameters. Because of its accessibility, rather low price, practical ease, and possibility to acquire objective quantitative information, brain perfusion SPECT can be complementary to other diagnostic modalities in disease surveillance and monitoring, even though additional neurophysiological and imaging investigations are needed. Besides, it might depict the effect of hyperbaric oxygen therapy in the improvement of cerebral perfusion in MS patients but not in clinical outcome. However, it could keep the physiological debate open in which this approach may have some value. The literature on this important issue is scarce, and follow up studies are required to assess these preliminary results.

Finally, it should be mentioned that the current study has a number of limitations. One of the most important limitations is the relatively low number of participants; however, it was quite homogenous in terms of the disease severity of patients. Another limitation of the study is that we didn’t assess the possible association of the cognitive status of MS patients with the improvement in brain perfusion following HBOT mostly due to retrospective nature of this study. Furthermore, most brain perfusion studies showed improvement following HBOT without a substantial effect on clinical subjective and objective evaluation in the short-term follow up. Extended monitoring in a more prolonged time course should be taken into account in future studies.

This study demonstrated decreased cerebral perfusion in SP-MS subjects with a moderate to severe disability score and its association with clinical parameters. Because of its accessibility, rather low price, practical ease, and being objective quantitative data, brain perfusion SPECT could be complementing to other diagnostic modalities such as MRI and clinical examinations in disease surveillance and monitoring, even though additional neurophysiological and imaging investigations are highly required.

CNS, central nervous system; EDSS, expanded disability status scale score;HBOT, hyperbaric oxygen therapy; MS, multiple sclerosis; PP, primary progressive; PR, progressive-relapsing; RR, relapsing remitting; SP, secondary progressive; SPM, statistical parametric mapping; Tc-99 m-ECD SPECT, Tc-99 m ethyl cysteinate dimer single-photon emission computed tomography.