Normal-pressure hydrocephalus (NPH) is a hydrocephalus syndrome characterized by the clinical triad of gait disturbance, cognitive decline, and incontinence, with ventricular enlargement and a normal cerebrospinal fluid (CSF) pressure. Clinically, it can be classified as idiopathic normal pressure hydrocephalus (INPH) or secondary normal pressure hydrocephalus (SNPH) based on whether the etiology is definite (Adams et al., 1965). At present, confirmation of INPH depends on the CSF tap test and shunt surgery (Nakajima et al., 2021), and clinical symptoms are treatable by permanent drainage (Relkin et al., 2005). However, a large number of patients with INPH are complicated with comorbidities, such as Alzheimer’s disease (AD), Parkinson’s-like diseases, extrapyramidal dyskinesia diseases (Allali et al., 2018), or similarities with normal aging (Agerskov et al., 2018), which may be the reason for a poor or even negatively impacting shunt response (Broggi et al., 2016; Bräutigam et al., 2019; Macki et al., 2020). With up to 40% of patients not responding, it remains difficult and essential to select patients who would benefit from surgery (Giordan et al., 2018). Therefore, by analyzing the relationship between baseline cerebral perfusion and changes in clinical symptoms after shunt, we attempted to infer the neural substrates underlying changes in clinical manifestations of INPH in order to screen the shunt surgery recipients and minimize unnecessary invasive procedures.

There is evidence that the pathological mechanism of INPH is complex and not only results from CSF circulation disorder (Lalou et al., 2018) but is also related to cerebrovascular self-regulation disorder and abnormal brain metabolism (Landau et al., 2014). Most studies have shown a significant decrease in regional cerebral blood flow (rCBF) in patients with INPH compared with healthy elderly individuals (Ziegelitz et al., 2014; Virhammar et al., 2017; Mattoli et al., 2020). Most studies have demonstrated that whole and regional cerebral blood flow (CBF) in patients with INPH were significantly lower than those in normal controls. Existing studies have not been consistent with the relationship between baseline CBF and change in clinical symptoms after shunt surgery. White matter CBF has been shown to gradually increase with greater distance from the ventricle, especially the lateral ventricle. In terms of gray matter, the frontal cortex and central gray matter were predominantly affected (Virhammar et al., 2014; Ziegelitz et al., 2016; Tuniz et al., 2017; Mattoli et al., 2020). Some studies suggest that hippocampal perfusion also decreased (Ziegelitz et al., 2014; Ziegelitz et al., 2015).

Arterial spin labeling (ASL) is a relatively new non-invasive perfusion imaging method that utilizes endogenous blood-based water molecules as tracers (without ionizing radiation) to visualize and quantify CBF (Jezzard et al., 2018). Based on the persistent stability of internal brain metabolic activity in the resting state, ASL can explore and monitor alterations in tissue perfusion in states of brain dysfunction (Soldozy et al., 2019).

The study aimed to determine the relationship between baseline CBF and preoperative clinical symptoms and the efficacy of shunt surgery. We hypothesized that baseline CBF may be associated with clinical changes after shunt surgery and help screen patients with INPH for shunt surgery clinically.

In this study, PASL, a non-invasive perfusion imaging technique, was used to compare cerebral perfusion in patients with INPH with that in healthy older individuals. We found that the whole-brain CBF of patients with INPH was significantly reduced, which was consistent with the results of previous studies (Virhammar et al., 2014; Ziegelitz et al., 2016; Mattoli et al., 2020), suggesting a significant decrease in the cerebral perfusion rate. Hypoperfusion of the brain could further lead to pathophysiological changes in brain tissue, particularly oxygen metabolism (including oxygen utilization and extraction) (Göttler et al., 2019).

Overall, significantly reduced CBF in the INPH group was found in the high convexity gray matter. According to the Guidelines for Management of Idiopathic Normal Pressure Hydrocephalus (Third Edition), disproportionately enlarged subarachnoid space hydrocephalus (DESH) can be observed in most patients with INPH, which may be a reliable imaging feature of subarachnoid CSF absorption disorder (Nakajima et al., 2021). Reduced perfusion in the high convexity brain may result in local vascular compression due to its greater compression than in other regions of the brain. In addition, the CSF in the narrow sulcus is significantly reduced compared with other parts, so the metabolic efficiency of the local brain tissue is decreased.

There was reduced perfusion in the temporal lobe. One possible reason is the temporal lobe compression due to the temporal horn dilation, and another potential cause may be the AD comorbidity, which may also explain the precuneus hypoperfusion observed in this study (Ishii, 2020). Consistent with the findings of Ziegelitz’s two studies on regional cerebral perfusion (Ziegelitz et al., 2014; Ziegelitz et al., 2015), CBF in the frontal lobe and periventricular white matter in patients with INPH was significantly lower than that in controls. Combined with previous studies (Ziegelitz et al., 2014; Virhammar et al., 2017), we speculated that the edema in the paraventricular white matter led to local compression of small vessels and metabolic disorder of vasoactive metabolites. This study found that thalamic CBF significantly decreased, which may be due to the decrease in N-acetyl-aspartate levels in the thalamus of patients with INPH due to impaired local metabolism (Miyamoto et al., 2007; Lundin et al., 2011), and downregulated circuits between cortical and subcortical structures, which may be related to the occurrence of clinical symptoms.

Many previous studies (Calamante, 2010; Ziegelitz et al., 2014, 2015) assumed that occipital cortex perfusion was less affected by diseases and used the occipital lobe as an internal reference to evaluate the perfusion in other brain regions. However, the results of this study showed that the occipital lobe perfusion in patients with INPH was significantly reduced compared with that in HCs, suggesting that there may be some deviation between the assumptions of previous studies and the real situation.

In this study, there was a significant correlation between the baseline CBF in the frontal lobe, temporal lobe, basal ganglia, thalamus, and cingulate gyrus and poor gait manifestation. Elderly individuals have more brain regions involved in motor control than younger ones. Previous studies found that the CBF in the frontal periventricular white matter was significantly correlated with gait (Jurcoane et al., 2014). The prefrontal cortex, which receives almost all information from the sensory system and is preferentially connected to motor information processing structures, plays a core role in cognitive control of motor performance, and therefore, elderly individuals are more dependent on activation of the bilateral frontal cortex during exercise. The frontal periventricular corticobasal ganglia-thalamocortical pathways are constituted by the fibers of the frontal lateral ventricle connected with the supplementary motor cortex, basal ganglia, and thalamus, which are involved in gait and body balance control (Virhammar et al., 2014). The thalamus and cingulate gyrus play important roles in this pathway, so reduced local perfusion may lead to downregulation of pathway function, which can manifest as motor dysfunction (Ziegelitz et al., 2015). The thalamus is the main structure regulating basal ganglia function and plays an important role in motor function. These structures influence each other and contribute to the occurrence and outcome of gait disorders in the course of diseases.

In the study, preoperative CBF in the frontal lobe and basal ganglia was positively correlated with the severity of preoperative cognitive impairment associated with INPH and negatively correlated with the change of postoperative MMSE score. Decreased perfusion and metabolic disturbance lead to impaired ventricular white matter and then influence the function of the frontal-subcortical pathways and the frontal periventricular corticobasal ganglia-thalamocortical pathways, which further affects cognitive function. Previous studies have proven that this circuit is positively correlated with cognitive and psychological scores (Ziegelitz et al., 2016).

The improvement in motor function after shunting was negatively correlated with preoperative perfusion in the parietal lobe, occipital lobe, and temporal lobe. The occipital lobe is connected to the precuneus and frontal and temporal lobes through the cingulate tract and is involved in the regulation of spatial relations and visual attention of body movement (Tanglay et al., 2021). Some studies have shown that parietal occipital white matter damage is associated with gait disorder (de Laat et al., 2011). We hypothesized that it may be possible to predict the recovery of motor function after shunting by the perfusion damage observed in the parietal and occipitotemporal lobes before surgery.

The results showed that there was a moderate positive correlation between reduced anterior cingulate perfusion and improved postoperative urination. The higher brain’s net effect on micturition is thought to be inhibitory. Moreover, the micturition reflex passes through the dorsolateral frontal cortex, anterior cingulate cortex, and hypothalamus (Tish and Geerling, 2020) as a part of the frontal-subcortical pathway and is closely related to the urinary control function. Preoperative perfusion in the cingulate gyrus being associated with recovery of bladder function is likely to predict the functional restoration of the descending cingulate pathway or cingulate cortex after shunt surgery.

Some studies have suggested that preoperative whole-brain CBF is related to clinical outcomes, and patients with lower CBF show clinical improvements after shunt surgery, indicating that preoperative CBF may contribute to predicting clinical outcomes after shunt surgery (Klinge et al., 2002). The slightly lower significance of our results compared with previous studies may be related to the small sample size and heterogeneity among the subjects, as well as differences in research methods and evaluation criteria of clinical symptoms. The results of previous studies have been discrepant, and whether baseline CBF can predict the outcome of shunt surgery has not been determined. Therefore, it is necessary to expand the sample size and conduct more in-depth research on this issue.

In some patients, motor function significantly improved in the short term after shunt surgery, but the duration was short, with the shortest maintenance of nearly a week. Specific analysis of individual patients revealed that a long course of the disease was a common feature. Combined with a CT perfusion (CTP) study (Ziegelitz et al., 2016), we speculated that massive CSF drainage could improve periventricular perfusion in the short term, but there might be permanent damage to white matter, which may interfere with the improvements in gray matter perfusion through the frontal periventricular corticobasal ganglia-thalamocortical pathways.

Positron emission tomography perfusion imaging, the only technique that can intrinsically quantify perfusion, is considered the gold standard in the cerebral perfusion evaluation with imaging (Mattoli et al., 2020). Previous studies have validated the CBF value of PASL against PET, proving that PASL has high repeatability in HCs and patients with AD (Xu et al., 2010). The advantage of the study lies in the quantitative analysis of CBF based on the whole-brain voxel level. On the one hand, this method directly analyzes the original data and does not involve a priori assumption of artificially defined ROI, so it is not subject to the subjective influence of researchers. On the other hand, the object of statistical analysis is each voxel in the CBF map, so the statistical result is not affected by volume. The spatial normalization process using the individuals’ structural phase enables a voxel-based statistical comparison of brain images with different morphologies.

Compared with previous studies that take the occipital cortex as an internal reference to delineate regions of interest (ROIs), this study solved the limitations of manually drawing ROIs and avoided the influence of anatomical artifacts and volume on research results, resulting in higher accuracy. Therefore, voxel-based analysis has the advantages of automaticity, comprehensiveness, objectivity, and repeatability.

There were some limitations to this study that need to be considered. First, the post-labeling delay in the PASL technique used in this study was short (1,290 ms), which led to hypoperfusion artifacts in some subjects. Second, a single TI PASL was used in this study, which is less efficient and more dependent on model assumptions for the arterial transit time, resulting in slightly poorer accuracy. Considering that the subjects involved in this study could not undergo a long-time MRI scan due to cognitive impairment and old age, we chose the short post-labeling delay single TI PASL sequence as a trade-off. In the following study, we will continue to increase the sample size of the INPH group and strengthen follow-up, make a regression analysis of preoperative perfusion and clinical symptoms, and try to establish a prediction model of shunt efficacy to predict the efficacy through preoperative perfusion and help patients, their families, and healthcare professionals involved in treating INPH.

In this study, we measured brain perfusion in the patients with diagnosed INPH before and 1 month after shunt surgery to investigate the relationship between preoperative CBF and postoperative CBF changes and clinical symptoms. The current findings suggest that the HCs and the patients with diagnosed INPH exhibited CBF differences in the whole cerebrum, especially in the high convexity, temporal gyrus, and frontal white matter. The perfusion in different brain regions in the patients with INPH was correlated with clinical symptoms, and improvements in clinical symptoms after shunting were affected by the preoperative CBF. This study demonstrates that widely distributed and intercorrelated cortical and subcortical pathways are involved in the development of INPH symptoms. The pathogenesis of hypoperfusion and its specific effects on disease development need to be further explored in combination with other imaging techniques and molecular studies.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Institutional Review Board of Huadong Hospital affiliated with Fudan University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

WH, XF, SL, RM, CY, WL, and GL made a substantial contribution to the concept and design, acquisition of data or analysis, and interpretation of data. WH, XF, and SL drafted the manuscript and revised it critically for relevant intellectual content. WH and XF performed the MR examination and follow-up of patients. All authors made a substantial contribution to the concept and design, acquisition of data or analysis, interpretation of data, and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.