A prospective longitudinal study published in Frontiers in Neurology provides preliminary evidence that cardiopulmonary bypass (CPB) is associated with dynamic and coordinated remodeling of brain structure and function within the first two weeks after surgery. The research team, led by Bo Liu and colleagues from Xinqiao Hospital and other institutions in China, scanned 68 patients undergoing open-heart surgery requiring CPB both before the procedure and 15 days after. By comparing these within-subject scans, the study aims to characterize the early trajectory of neural alterations that previous cross-sectional studies have only partially identified.
The study enrolled 68 patients, 34 male and 34 female, with a mean age of 54.83 years, who were scheduled for cardiac surgery at Xinqiao Hospital between November 2023 and November 2024. The cohort included patients with rheumatic heart disease, cardiac valve disease, and other lesions. All participants underwent CPB under hypothermic circulatory arrest, with a mean surgical duration of 268.01 minutes and a mean CPB duration of 147.32 minutes. The 15-day postoperative time point was specifically chosen to capture early neural changes that might precede longer-term cognitive outcomes.
Each participant underwent two identical MRI sessions on a 3.0 T scanner: one within 72 hours before surgery and another 15 days after clinical stability was achieved. The imaging protocol included resting-state functional MRI (fMRI) to assess local neural activity and high-resolution T1-weighted structural MRI to examine brain morphology. The functional scans used an echo-planar imaging sequence, while the structural scans used a 3D fast spoiled gradient recalled acquisition in the steady state sequence to achieve high spatial resolution.
To quantify functional changes, the researchers calculated regional homogeneity (ReHo), a metric that measures the similarity of the time series of a voxel to its 26 nearest neighbors, reflecting local neural synchronization. They also constructed functional brain networks using the Power-264 atlas to analyze global and nodal network properties, including efficiency, clustering coefficient, and betweenness centrality. For structural analysis, surface-based morphometry was used to estimate cortical metrics such as gray matter volume, cortical thickness, sulcal depth, fractal dimension, and gyrification index.
The functional results revealed a coordinated pattern of brain reorganization. ReHo values decreased significantly in hubs of the default mode network (DMN), specifically the posterior cingulate cortex and the right angular gyrus. Conversely, ReHo increased in regions of the fronto-parietal network (FPN), including the bilateral inferior frontal gyri, and in the right temporal lobe. For instance, the mean ReHo in the posterior cingulate cortex dropped from 1.24 preoperatively to 1.08 postoperatively, a change associated with a medium effect size.
Network analysis indicated a shift in the balance between functional segregation and integration. Postoperative scans showed reduced clustering coefficient and local efficiency, metrics associated with local specialization, alongside increased global efficiency and betweenness centrality. This suggests a potential network-level reorganization where the brain may be altering its integrative capacity. The nodes showing these changes were predominantly located in the bilateral temporal and frontal lobes, the bilateral precuneus, the right angular gyrus, and the right anterior cingulate cortex.
Structurally, the functional shifts were paralleled by widespread cortical alterations. Gray matter volume was notably reduced in the bilateral frontal and temporal lobes. Cortical thinning was observed in the bilateral superior frontal gyri, with the right superior frontal gyrus showing a particularly large effect size. Sulcal depth was shallower in the bilateral middle frontal gyrus, temporal lobes, left supramarginal gyrus, and right precuneus. Additionally, the postoperative cohort exhibited reduced fractal dimension, indicative of less complex cortical folding, in the bilateral superior temporal lobes, middle frontal lobes, and left lateral occipital gyrus.
The convergence of these findings suggests a potential mechanistic pathway where structural changes may affect the antagonistic influence of the FPN on the DMN, thereby altering their dynamic balance. The DMN is recognized for its role in internally directed processes such as self-awareness and episodic memory retrieval, while the FPN is critical for executive control. The observed imbalance between these networks, along with distinct alterations in the right temporal lobe, may represent candidate neural correlates for postoperative cognitive dysfunction, although this remains a hypothesis requiring direct validation.
The study has several important limitations that must be considered when interpreting the results. The sample size was relatively small, and the lack of long-term follow-up data precludes the evaluation of trends in brain changes over time. Furthermore, the absence of a healthy control group scanned at comparable intervals means that potential time effects or scan-rescan reliability issues cannot be ruled out. The study also did not account for variations in intraoperative variables or disease duration, which could influence the observed outcomes.
A significant limitation is the lack of cognitive assessment data, such as the Attention Network Test, due to poor postoperative cooperation among patients. Consequently, the direct clinical and cognitive implications of the imaging findings remain uncertain. The authors acknowledge that their interpretation of the network-level alterations is based on indirect evidence and requires direct validation in future studies incorporating neuropsychological assessments. More comprehensive modeling approaches, such as linear mixed-effects models, could provide additional insights but were not feasible due to the limited availability of clinical covariates in the present dataset.
Despite these limitations, the study offers a more dynamic view of the early neurobiological sequelae of CPB than previous cross-sectional designs. By identifying specific concurrent alterations in the DMN, FPN, and right temporal lobe, the work provides a preliminary basis for developing perioperative strategies to preserve brain network equilibrium. The findings extend previous cross-sectional work by revealing a specific pattern of network-level dysfunction and localized structural changes, highlighting the interconnected nature of brain reorganization following CPB.
Future research should expand the sample size, conduct long-term follow-up investigations, and enroll a healthy control group for comparative research. Controlling for more variables and incorporating comprehensive neuropsychological testing will be essential to validate these observations and elucidate the effects of CPB on brain structure and function. This study contributes to the growing body of evidence on the neurological sequelae of cardiac surgery and underscores the need for targeted neuroprotective strategies to enhance postoperative recovery.
The research was approved by the institutional ethics committee, and all participants provided written informed consent in accordance with the Declaration of Helsinki. The data supporting the conclusions of this article will be made available by the authors without undue reservation. This open-access article is distributed under the terms of the Creative Commons Attribution License, allowing for use, distribution, or reproduction in other forums provided the original authors and copyright owners are credited.
The clinical motivation behind tracking these early postoperative shifts stems from the substantial burden imposed by postoperative cognitive dysfunction, a documented complication that involves memory loss, altered personality, and reduced learning capacity. Published studies indicate that between 30 and 80 percent of surgical patients exhibit measurable cognitive deficits upon hospital discharge, with 20 to 40 percent remaining impaired six months later. These cognitive disturbances frequently prolong hospital stays, raise post-discharge admissions to specialized care facilities, and elevate overall patient morbidity and mortality. Investigators hypothesize that the underlying etiology is multifaceted, driven primarily by cerebral micro-embolic showers, hypoperfusion-reperfusion injuries, oxidative stress, and the systemic inflammatory cascade triggered by cardiopulmonary bypass. Additionally, the intraoperative hemodilution and altered blood composition inherent to bypass circulation can critically impair cerebral oxygen delivery and metabolic homeostasis, predisposing patients to acute and protracted neurological injury.
To isolate the specific impact of cardiopulmonary bypass, the investigators enforced stringent eligibility criteria, excluding individuals with left-handedness, pre-existing structural brain lesions, acute psychiatric conditions, substance dependency, or a history of head trauma. Patients were also barred if they presented with chronic systemic diseases that could impair cognition, including uncontrolled diabetes, active malignancy, rheumatoid arthritis, or hypertension complicated by end-organ damage. Scans displaying motion artifacts above 1.5 millimeters of translation or 1.5 degrees of rotation were removed from evaluation. Across the cohort, surgical operations required a mean aortic cross-clamp time of 101.78 minutes and a mean hypothermic circulatory arrest period of 112.00 minutes under standardized total intravenous anesthesia using propofol and sufentanil. Patient education averaged 8.0 years, and head motion did not differ significantly between scanning sessions, with mean framewise displacement measuring 0.0840 millimeters preoperatively and 0.0789 millimeters postoperatively.
The imaging pipeline incorporated rigorous computational controls to safeguard data fidelity. Preprocessing of functional series discarded the initial ten volumes to permit signal stabilization, followed by slice-timing adjustments, rigid-body realignment, and DARTEL spatial normalization to Montreal Neurological Institute coordinate space. To minimize non-neural artifacts without removing global baseline trends, the researchers avoided global signal regression, choosing instead to regress white matter and cerebrospinal fluid signals alongside Friston-24 motion parameters. Topological brain graphs generated from the Power-264 parcellation were evaluated across sparsity thresholds from 0.05 to 0.50, using area under the curve calculations. False Discovery Rate adjustments were strictly enforced across the brain at the voxel level for functional synchronization and across the cortical surface at the vertex level for structural morphometry. Beyond widespread gray matter reductions, these surface assessments confirmed localized reductions in cortical complexity, including a lower gyrification index in the right rostral middle frontal lobe with a large Cohen’s d effect size of 0.92, alongside marked thinning in both the right and left superior frontal gyri.
Subject of Research: Longitudinal fMRI study of brain structural and functional alterations following cardiopulmonary bypass
Article Title: Brain structural and functional alterations in patients following cardiopulmonary bypass: a short-term longitudinal fMRI study
Article References: Liu, B., Zhao, W., Fan, W., Yuan, J., Zhang, D., & Zhang, S. (2026). Brain structural and functional alterations in patients following cardiopulmonary bypass: a short-term longitudinal fMRI study. Frontiers in Neurology, 17, Article 1853171. https://doi.org/10.3389/fneur.2026.1853171
Image Credits: AI Generated
DOI: 10.3389/fneur.2026.1853171
Keywords: cardiopulmonary bypass, neuroimaging, functional MRI, structural MRI, default mode network, fronto-parietal network, regional homogeneity, surface-based morphometry, cognitive dysfunction, brain reorganization, postoperative care, neuroplasticity, cardiac surgery, cortical thickness, gray matter volume, network topology, longitudinal study, neurological sequelae, brain connectivity, postoperative recovery
News Source: Ophelia Keating. (November 14, 2026). Brain Scans Show Early Structural and Functional Shifts After Heart Surgery. Scienmag.



