| Objective:Gadolinium(Gd)-based contrast agents(GBCAs)have been widely used in enhanced magnetic resonance imaging(MRI)examinations since the 1980s,and are the most commonly used contrast agents for MRI at present.They had been considered relatively safe due to the low incidence of serious adverse effects until several reports in2006 that associated nephrogenic systemic fibrosis(NSF)with multiple GBCA exposures.This discovery makes the safety of GBCAs widely concerned by scholars in medical radiology.Recent studies have revealed increased signal intensities in the deep cerebellar nucleus(DCN)and globus pallidus on unenhanced T1-weighted MR images following repeated intravenous administrations of GBCAs.Further studies detected Gd deposits in brain tissues,especially in patients and animals which received multiple linear GBCAs with lower intrinsic stability.To date,the potential risks of these deposits remain unclear,and the Food and Drug Administration(FDA)and the European Medicines Agency(EMA)attached great importance to GBCA-induced Gd retention.EMA even recommends that high-risk linear gadolinium contrast agents such as gadodiamide and Gadopentetate dimeglumine should be discontinued in clinical practice to avoid potential brain damage caused by Gd deposition.To date,the mechanism of GBCA accumulation in the brain remains unclear.One potential pathway of GBCA permeation into the brain is through the blood-brain barrier(BBB).Another possible route is infiltration from the blood into the cerebrospinal fluid(CSF).It was postulated that once permeated into the CSF space,the GBCA would enter the glymphatic system and deposite in the brain.Diabetes mellitus(DM)is one of the most challenging endocrine disorders across the world.As a common basic disease,it is often accompanied with other disorders,such as tumors,chronic liver diseases and vascular diseases,requiring repeated enhanced MRI examinations.DM has been shown to progressively disrupt BBB integrity and function.Diabetes is also associated with the impairment of the glymphatic system,which has recently been shown to eliminate potential neurotoxic waste products,including GBCAs from the interstitial space.On the other hand,DM is associated with numerous complications including diabetic nephropathy.Diabetic nephropathy is one of the major causes of renal failure,which may delay GBCA clearance.Therefore,investigating the Gd deposition and clearance in diabetic brains compared with healthy status is of particular significance for evaluating the safety of GBCAs in diabetic patients.As a metabolic disorder characterized by elevated blood sugar,diabetes can cause central nervous system complications,causing brain damage and cognitive dysfunction.Studies have shown that the pathological manifestations of brain tissue damage include neuronal degeneration,apoptosis and activation of glial cells.HE staining and Nissl staining are two commonly used methods of neural histological staining.Nissl staining could clearly show Nissl bodies in neurons.When neurons are damaged,it could show neural enlargement,cytoplasmic deep staining,nuclear vacuolar degeneration,and nuclear pyknosis or nuclear lysis in HE staining.Under normal physiological conditions,Nissl bodies are large and enough,indicating that neurons have strong functions in synthesizing various structural and functional proteins.Nissl staining can be manifested as a decrease or even disappearance of the Nissl bodies,with a lighter staining.Glial fibrillary acidic protein(GFAP)is a type III intermediate filamentous protein mainly distributed in astrocyte glial cells of the central nervous system,and is a marker of astrocyte activation.In the case of neuronal injury,GFAP expression is elevated,suggesting astrocyte gliosis.In the central nervous system,neuron specific enolase(NSE)is a soluble cytoplasmic protein that is continuously expressed by mature neurons and neurons-derived cells.It is a brain tissue-specific glycolytic enzyme and plays an important role in the process of cellular energy metabolism.In ischemic brain injury,elevated NSE expression can be detected.To date,there have been no studies on the effects of diabetes status on the deposition and clearance of Gd deposition in the brain and whether Gd deposition aggravates the damage of diabetic brain tissue.Therefore,in the first part of this paper,we established the normal rat model and type 1 diabetic rat model,and injected the rats with two linear GBCAs(gadodiamide and gadopentetate dimeglumine)and the macrocyclic GBCA gadoterate meglumine.MRI,inductively coupled plasma mass spectrometry(ICP-MS)and transmission electron microscopy(TEM)were used to study the deposition and clearance of Gd in various parts of the brain in diabetic status,in order to clarify the effects of diabetes status on the Gd deposition and clearance in the brain.In the second part of this paper,we established the normal rat model and type 1 diabetic rat model,and injected the rats with two linear GBCAs(gadodiamide and gadopentetate dimeglumine)and the macrocyclic GBCA gadoterate meglumine.HE staining,Nissl staining,GFAP and NSE immunohistochemical staining techniques were used to investigate whether the Gd deposition further aggravated the pathological damage of the type 1 diabetic brain,in order to contribute to evaluating the safety of GBCAs in diabetic patients.Methods:1.Experimental study of gadolinium-based contrast agent deposition and clearance of in the brain of type 1 diabetic rat model1.A rat model of type 1 diabetes was established.Diabetic rats(n=52)and normal rats(n=52)were routinely fed for 4 weeks.2.Diabetic rats and normal rats were randomly divided into 4 groups:gadodiamide group,gadopentetate dimeglumine group,gadoterate meglumine group,and control group(0.9%physiological saline).For each group of rats,0.6 mmol/kg(1.2 ml/kg)GBCA or an equal amount of physiological saline was injected into the tail vein for 5weeks and 4 days per week.3.According to the different time of sacrifice in rats,each group was divided into two subgroups:7 d subgroup(killed 7 days after the last injection of GBCAs,n=6)and42 d subgroup(killed 42 days after the last injection of GBCAs,n=7).For the 7 d subgroup,6 rats were executed and ICP-MS was performed to analyze the Gd concentration in different parts of the brain.For the 42 d subgroup,6 rats were subjected to the brain T1W MRI scan once a week until the end of observation period,and ICP-MS was performed after execution.One rat was subjected to TEM in order to observe the microscopic distribution of Gd granules after execution.4.T1W MRI scan in the brain and image quantitative analysis:For the 42 d subgroup,6 rats were subjected to the brain T1W MRI scan and image quantitative analysis during the injection of GBCAs and during the observation period.The region of interests(ROIs)were placed in the bilateral DCN and cerebellar parenchyma,and the DCN signal intensity ratio was calculated:DCN(the higher value of the two sides)/cerebellar parenchyma.5.Microwave digestion and ICP-MS experiments were carried out to analyze the Gd concentrations in the brain of 7 d subgroup and 42 d subgroup.6.TEM was conducted to analyze the microscopic distribution of the Gd forci in the brain of 42 d subgroup.2.Experimental study of histopathological features in the brain of type 1 diabetic rat model after administrations of gadolinium-based contrast agents1.A rat model of type 1 diabetes was established.Diabetic rats(n=24)and normal rats(n=24)were routinely fed for 4 weeks.2.Diabetic rats and normal rats were randomly divided into 4 groups:gadodiamide group,gadopentetate dimeglumine group,gadoterate meglumine group,and control group(0.9%physiological saline).For each group of rats,0.6 mmol/kg(1.2 ml/kg)GBCA or an equal amount of saline was injected into the tail vein for 5 weeks and 4 days per week.3.The rats were sacrificed 42 days after the last injection of GBCAs.The brain tissue was removed after 4%paraformaldehyde perfusion.4 mm typical brain coronal sections including DCN and hippocampal CA3 area were isolated and immersed in 4%paraformaldehyde for 72 h.4.The well-fixed brain tissues were dehydrated,transparent,and embedded in paraffin.5.HE staining was carried out to observe the presence of neuronal degeneration,necrosis and gliosis.The the number of normal neurons per 3.8×10~5μm~2 was counted.6.Nissl staining was performed to observe whether the Nissl body was decreased,the atrophy of neurons and a high degree of basophilicity.The number of injured neurons per 3.8×10~5μm~2 was counted.7.GFAP immunohistochemical staining was performed for the brain,and the number of GFAP-positive cells per 9.6×10~4μm~2 was counted.8.NSE immunohistochemical staining was performed for the brain,and the number of NSE positive cells per 9.6×10~4μm~2 was counted.Results:1.Experimental study of gadolinium-based contrast agent deposition and clearance of in the brain of type 1 diabetic rat modelBefore the animals were sacrificed,the average blood glucose level of diabetic rats was significantly higher than that of normal rats(24.6±2.3 vs 5.2±0.4 mmol/L)(p<0.001).There was no significant difference in the total amount of injected GBCA between normal and diabetic rats(3.44±0.26 vs 3.41±0.31 mmol,p=0.988).In normal rats and diabetic rats,a significant T1 enhancement of DCN was observed in the gadodiamide group from the third week of the injection period(week 7)until the end of the study(week 14,p<0.05),compared with the saline group.Enhancement in the gadopentetate dimeglumine group was more progressively compared with the gadodiamide group during the injection period and the treatment-free period.No such enhancement was observed in the macrocyclic gadoterate meglumine group compared with the saline group(p>0.05).Compared with normal rats,the T1 signal intensity ratios of DCN/cerebellum in diabetic rats treated with gadodiamide were significantly lower during the injection period and the treatment-free period(p<0.001).Similarly,the signal ratios in diabetic rats treated with gadopentetate dimeglumine were relatively lower from the eighth week until the end of the study,compared with normal rats(p<0.001).In the gadoterate meglumine group,no significant difference of the T1 signal intensity ratio of DCN/cerebellum was found in diabetic rats and normal rats,both at the baseline level(p>0.05)In both time-point subgroups of the 2 linear groups and 1 macrocyclic group,the average Gd retentions in the brain of diabetic rats were significantly lower than in normal rats.In normal rats and diabetic rats,the deposition of sputum in different parts of the brain showed similar distribution characteristics.The highest Gd deposition occurred in the olfactory bulb,DCN and striatum after multiple intravenous injections of the linear gadodiamide.In both the normal rats and diabetic rats,the highest average residual Gd depositions were measured in the gadodiamide group,followed by gadopentetate dimeglumine group.By contrast,the Gd retentions with macrocyclic gadoterate meglumine were significantly lower.In normal rats,compared with residual Gd concentrations 7 days after the injection period,42 days after injections,the average Gd concentrations were significantly decreased in both the linear GBCAs and macrocyclic gadoterate meglumine.In diabetic rats,compared with the values 7 days after the injection period,the average Gd concentrations 42 days after injections in the 2 linear GBCAs showed no significant difference.By contrast,for the macrocyclic gadoterate meglumine group,42 days after injections,the average Gd concentrations was significantly decreased.During TEM evaluation,electron dense granules were observed in the olfactory bulbs of both the linear gadodiamide group and the linear gadopentetate dimeglumine group of the normal rats.The larger number of the Gd deposits was detected in normal rats treated with gadodiamide.Electron dense granules were also observed in the olfactory bulbs of the linear gadodiamide group in the diabetic rats.The amoun of granules around the vessels in diabetic rats of gadodiamide group was significantly less than that in normal rats.2.Experimental study of histopathological features in the brain of type 1 diabetic rat model after administrations of gadolinium-based contrast agentsThere was no statistically significant difference in the number of normal neurons in the DCN area of normal rats(p=0.069).In the diabetic rats,the cytoplasm of some neurons in the DCN region was deeply stained,and some of the nucleus disappeared.There was no statistically significant difference in the number of normal neurons in the DCN area among the diabetic groups(p=0.458).There was no statistically significant difference in the number of normal neurons in the hippocampal CA3 area of normal rats(p=0.057).In the hippocampal CA3 area of diabetic rats,a large number of neurons were atrophied,the cytoplasm was deeply stained,and some of the nucleus disappeared.There was no significant difference in the number of normal neurons in the hippocampal CA3 area among the diabetic groups(p=0.222).The structure and morphology of neurons in the DCN area of normal rats were normal,and the Nissl staining was clear.There was no statistically significant difference in the number of abnormal neurons in the normal groups(p=0.550).In the DCN area of diabetic rats,the number of Nissl bodies in some neurons was reduced and the staining was blurred.There was no statistically significant difference in the number of abnormal neurons in the DCN area among the diabetic groups(p=0.876).Nissl staining was evident in the hippocampal CA3 region of normal rats,and a small amount of neuronal cytoplasm was deeply stained.There was no statistically significant difference in the number of abnormal neurons in each normal group(p=0.336).A large number of neurons were atrophied in the hippocampal CA3 area of diabetic rats,and the cytoplasm was deeply stained,and some of the nucleus disappeared.There was no statistically significant difference in the number of abnormal neurons in the hippocampal CA3 area among the diabetic groups(p=0.400).The morphology of GFAP-positive astrocytes in the DCN of normal rats was normal,and there was no significant difference in the number of GFAP-positive glial cells among the normal groups(p=0.554).The number of GFAP-positive astrocytes in the DCN area of diabetic rats was increased,showing typical dendritic changes.There was no statistically significant difference in the number of GFAP-positive glial cells in the DCN area among the diabetic group(p=0.440).The morphology of GFAP-positive astrocytes in the hippocampal CA3 area of normal rats was normal,and there was no statistically significant difference in the number of GFAP-positive glial cells among the normal groups(p=0.456).The number of GFAP-positive astrocytes in the hippocampal CA3 area of diabetic rats was increased,showing typical dendritic changes.There was no statistically significant difference in the number of GFAP-positive glial cells in the hippocampal CA3 area among the diabetic groups(p=0.992).The structure and morphology of NSE-positive neurons in the DCN area of normal rats were normal.There was no statistically significant difference in the number of NSE-positive cells in each group(p=0.420).The number of NSE-stained positive neurons in the DCN area of diabetic rats was increased.There was no statistically significant difference in the number of NSE-stained neurons in the DCN area of each diabetic group(p=0.839).The structure and morphology of NSE-positive neurons in the hippocampal CA3area of normal rats were normal.There was no statistically significant difference in the number of NSE staining positive cells in each group(p=0.323).The number of NSE-stained positive neurons in the hippocampal CA3 area of diabetic rats was increased.There was no statistically significant difference in the number of NSE-stained neurons in the hippocampal CA3 area among the diabetic groups(p=0.834).Conclusions:1.Compared with normal rats,diabetic status decreased the residual Gd concentrations in the brain after multiple intravenous administrations of linear(gadodiamide and gadopentetate dimeglumine)and macrocyclic(gadoterate meglumine)GBCAs.2.The concentrations of deposited Gd in the brain of diabetic rats with multiple intravenous injections of linear GBCAs were significantly higher than that of diabetic rats with multiple intravenous injections of macrocyclic GBCA.3.Diabetic status delayed the Gd clearance of linear gadodiamide and gadopentetate dimeglumine from the brain,but did not delay the Gd clearance of macrocyclic gadoterate meglumine from the brain.4.No histological damages were detected in the brain of normal rats treated with either linear or macrocyclic GBCA.5.The histopathological changes were not aggravated by the deposited Gd in the type 1 diabetic brain. |
PDF Full Text Request