Study On Behavioral Improvement And Neural Mechanism Of MRI Associated With Compulsory Detoxification For Heroin Use Disorder

Part Ⅰ: Neural mechanisms underlying craving and cognitive changes in heroin addicts after compulsory detoxification based on MRIObjective: Previous studies have shown that there are structural and functional changes in the brain of individuals with heroin use disorder(HUD)compared to healthy controls(HCs).In China,the compulsory detoxification law is implemented for drug addiction.However,the impact of enforced abstinence for more than six months(referred to as "enforced abstinence")on the brain structure and function of chronic HUD patients,as well as the potential promoting effect on brain plasticity,remains unclear.This study aims to investigate the changes in craving,cognition,and impulsivity,the three core clinical features of HUD patients,after enforced abstinence,and their correlation with changes in brain function,thereby exploring the corresponding neural mechanisms.Method: This study included 108 patients with opioid/heroin use disorder(OUD/HUD)recruited from a drug rehabilitation center who met the diagnostic criteria for substance-related and addictive disorders in the Diagnostic and Statistical Manual of Mental Disorders-V(DSM-V).High-resolution T1-weighted structural magnetic resonance imaging(s MRI)and blood oxygen level-dependent(BOLD)-based resting-state functional magnetic resonance imaging(f MRI)data were collected from HUD patients at baseline(HU1 group).MRI examinations were conducted on 61 HUD patients at an average of 8 months after forced detoxification(HU2 group).Visual analog scale(VAS),Heroin Craving Questionnaire(HCQ),Trail Making Test-A(TMT-A)of the Psychomotor Cognitive Function Test,and the Barratt Impulsiveness Scale version 11(BIS-11)were used to evaluate craving,cognition,and impulsivity in both groups,and behavioral differences between the two groups were compared.Freesurfer software was used to compare cortical thickness differences between the 61 HUD patients before and after forced detoxification,and the cortical regions with significantly increased thickness were extracted as regions of interest(ROIs)and transformed to the Montreal Neurological Institute(MNI)standard space via ch2 space conversion.The resting-state functional connectivity(RSFC)between the ROIs and the whole brain was analyzed using tools such as AFNI(Analysis of Functional Neuro Images)and FSL(Functional MRI of the Brain’s Software Library)for the MRI data of the HU1 and HU2 groups.The paired t-test was used to compare the differences in behavioral scores and RSFC scores between the two groups.Finally,the Pearson correlation analysis was used to explore the longitudinal changes in behavioral scores and brain imaging.Results:(1)Compared with the HU1 group,the VAS and HCQ craving scores of the HU2 group were significantly reduced,and the completion time of the TMT cognitive test was significantly shortened(p< 0.05).There were no significant differences in BIS total scores and subscale scores between the two groups(p > 0.05).(2)Compared with the HU1 group,the cortical thickness of the bilateral superior frontal gyrus(SFG)was significantly increased in the HU2 group(p < 0.05,family-wise error correction,FWE correction).The cortical thickness of the left lingual gyrus,left posterior cingulate gyrus,left fusiform gyrus,left paracentral lobule,bilateral superior temporal gyrus,bilateral precuneus gyrus,and right precentral gyrus were significantly decreased in the HU2 group(p < 0.05,FWE correction).(3)Compared with the HU1 group,the RSFC between the left SFG and bilateral inferior frontal gyrus(IFG)was significantly increased in the HU2 group(p < 0.05,FWE correction).(4)Compared with the HU1 group,the RSFC between the left SFG and bilateral insula,bilateral thalamus,and left IFG,left insula,left orbital frontal cortex,right IFG,and right insula were increased in the HU2 group(p < 0.005,uncorrected).(5)The changes in RSFC between the left SFG and bilateral IFG,bilateral caudate nucleus,left globus pallidus,and right insula between the HU1 and HU2 groups were positively correlated with changes in TMT-A completion time.Conclusion: The results of this study indicate that there is partial recovery of the prefrontal-limbic brain circuit in the HUD group after forced abstinence,and forced abstinence may have a promoting effect on the improvement of cognitive function and the reduction of drug craving levels.This study provides new evidence for the partial recovery of brain functions related to neurocognition and craving after forced abstinence.Part Ⅱ: Study on the Mechanism of Craving and Changes in Functional Connectivity of Insula Subregions in Heroin Use Disorder After Compulsory AbstinencePurpose: The insular cortex plays a crucial role in the core reward circuit of substance use disorder(SUD)and is closely related to craving.Recent studies have shown functional differences among various subregions of the insula.This study aimed to explore the effects of abstinence on the resting-state functional connectivity(RSFC)of insula subregions in heroin-dependent individuals(HUD).Methods: This study recruited 108 HUD patients who met the DSM-V criteria for substance-related and addictive disorders and collected high-resolution T1-weighted structural MRI and blood oxygen level-dependent(BOLD)f MRI data at baseline(HU1 group).MRI scans were also performed on 61 HUD patients at a follow-up time point(HU2group)after an average of 8 months of enforced abstinence.The craving level of HUD patients was evaluated using the Heroin Craving Questionnaire(HCQ).The insula was divided into three subregions: the posterior insular(PI),dorsal anterior insular(DAI),and ventral anterior insular(VAI).Using bilateral insula subregions as seed points(6 regions in total),Analysis of Functional Neuro Images(ANFI)and FMRIB Software Library(FSL)software were used to perform whole-brain RSFC analysis of the f MRI data of the HU1 and HU2 groups.Paired t-tests were used to compare the differences in HCQ and RSFC between the two groups.Pearson correlation analysis was used to explore the correlation between longitudinal changes in craving and RSFC of insula subregions before and after abstinence.Results:(1)Compared with the HU1 group,the HCQ score of the HU2 group was significantly reduced(p<0.05).(2)Compared with the HU1 group,the RSFC between the right DAI and right superior frontal gyrus(SFG)/bilateral anterior cingulate gyrus(ACG)in the HU2 group was significantly increased.The RSFC between the right PI and bilateral ACG/SFG in the HU2 group was significantly increased,as well as the RSFC between the right VAI and bilateral SFG,and the left DAI and bilateral ACG/bilateral SFG(all p<0.05,FWE corrected).(3)The changes in HCQ scores and RSFC of insula subregions were negatively correlated in HUD patients(right DAI-right SFG: r=-0.409,p=0.0052;right DAI-left SFG: r=-0.461,p=0.0015;right PI-right SFG: r=-0.491,p=0.0006;right PI-left SFG: r=-0.456,p=0.0016).Conclusion: The results of this study indicate that there are changes in the functional connectivity of insula subregions in HUD patients after abstinence and provide new evidence for the correlation between craving changes and changes in the neural circuits of insular cortex subregions after abstinence.This study reveals the possibility of long-term neuroplasticity after abstinence and provides a reference for the withdrawal treatment and rehabilitation program of future drug use disorder patients.Part Ⅲ: MRI Biomarkers of Prefrontal-Striatal Circuitry Implicated in Impulsive Behavior in Heroin Use DisorderObjective: Building on the findings from Part One,where impulsivity levels did not significantly change in HUD patients after forced abstinence,which may be related to innate impulsive traits in HUD patients,this part aims to explore the differences in impulsive behavior and brain imaging between HUD patients and HCs at baseline,with the aim of investigating brain imaging markers of impulsive behavior related to prefrontal-striatal structural and functional circuits in HUD.Methods: This study recruited 51 HUD patients and 50 matched HCs who met the diagnostic criteria for substance-related and addictive disorders in the DSM-V;T1-weighted high-resolution structural imaging,resting-state BOLD imaging,and diffusion tensor imaging(DTI)MRI data were collected.The volume of the subcortical nucleus accumbens(NAc)was analyzed using Free Surfer in both groups,and RSFC analysis was performed on f MRI data using AFNI/FSL tools.Probabilistic tractography based on striatal seed points was used to detect changes in the strength of white matter fiber connections between the striatum and10 pre-defined brain regions.Two-sample t-tests were used to compare BIS scores,subcortical NAc volume,RSFC,and white matter fiber connection strength of the striatal circuit between the two groups.The Barratt Impulsiveness Scale version 11(BIS-11)was used to assess impulsive behavior in both groups,and Pearson correlation analysis was used to explore the correlation between brain imaging and behavioral measures in the HUD group.Results:(1)Compared to the HC group,the HUD group had significantly higher scores on the BIS motor subscale(t=2.347,p=0.0253);(2)Compared to the HC group,the HUD group had significantly larger right NAc volume(t=4.719,p=0.036),and the RSFC between the right NAc and bilateral SFG in both groups was correlated with BIS motor scores;(3)Compared to the control group,the white matter fiber connection strength between the left striatum and medial orbitofrontal cortex(mOFC)was significantly reduced in HUD patients(Bonferroni-corrected,p<0.05/20=0.0025);(4)Compared to male HCs,male HUD patients had significantly higher total BIS scores,attention,motor,and non-planning scores(Bonferroni-corrected,p<0.05/4=0.0125);(5)In the HUD group,the white matter fiber connection strength between the left striatum and mOFC was negatively correlated with total BIS scores,attention,and non-planning scores(r1=-0.410,p1=0.005;r2=-0.432,p2=0.003;r3=-0.506,p3<0.001).Conclusion: Compared to the normal control group,the HUD group exhibited higher impulsivity,manifested as an increase in the volume of the right NAc and a reduction in the structural connectivity of the striatal-mOFC circuit.The RSFC between the right NAc and bilateral SFG and the white matter fiber connection strength of the left striatal-mOFC circuit may be brain imaging markers for identifying impulsivity levels in HUD patients.Part Ⅳ: Connectome-Based Model Can Predict Individual Treatment Effect After Compulsory Detoxification in individuals of Heroin Use DisorderObjective: The results of Parts I and II indicate that there are individual differences in the efficacy of forced abstinence treatment for heroin use disorder(HUD),as measured by changes in craving levels.Therefore,this section aims to use machine learning models,specifically the Connectome-Based Prediction Model(CPM),which incorporates MRI data and connectivity-based group analysis,to predict individual craving levels in HUD patients after FW treatment,using baseline f MRI data,with the goal of achieving personalized treatment efficacy prediction.Methods: The study recruited 69 HUD patients who met the DSM-V criteria for substance-related and addictive disorders from a drug rehabilitation center.T1 WI high-resolution s MRI and f MRI data were collected at the baseline(HU1 group),and the craving levels of the HU1 group and the follow-up HU2 group,which occurred 8 months after FW treatment,were assessed using the Heroin Craving Questionnaire(HCQ).The BOLD data was preprocessed and analyzed using tools such as SPM12 and GRETNA to construct the whole-brain functional connectivity matrix,which was then input into the CPM model.Leave-one-out cross-validation was used to determine the relevant brain networks that had the potential to predict HCQ scores and HCQ changes(HCQ follow-up minus HCQ baseline)in the HU2 group.Results: The CPM model was able to predict changes in craving after FW treatment,predicting HCQ scores(negative correlation network:r=0.417,p=0.0002)and HCQ change values(negative correlation network: r= 0.334,p= 0.0022;positive correlation network: r= 0.233,p=0.0384)at the 8-month follow-up after forced abstinence treatment.The predicted craving-related brain networks identified by CPM included the Somato-Motor Network(SMN),Salience Network(SALN),Default Mode Network(DMN),Medial Frontal Network(MFN),Visual Network(VN),and Auditory Network(AN).The connections between the Frontal-Parietal Network(FPN)and SMN,DMN,and SALN decreased,while the connections between the Subcortical Network(SCN)and DMN,SALN,and SMN increased,showing a positive correlation with changes in craving levels.Conclusion: The results of this study suggest that CPM has potential applications in predicting the effect of FW treatment for HUD patients.The study’s findings are helpful for identifying specific populations that are not sensitive to forced abstinence treatment’s efficacy through early brain imaging screening,and for developing personalized treatment plans for them.Moreover,the specific craving-related brain networks identified by CPM may serve as target brain regions for future novel treatments.