【作者】 杨杰；

【导师】 王佳；

【作者基本信息】 四川大学 ， 外科学（泌尿外科）（专业学位）， 2021， 博士

【摘要】 目的:肾结石是一种发病率、复发率高,经济负担重的疾病。预防肾结石形成及复发是现代泌尿外科面临的巨大挑战。虽然目前腔道泌尿外科手术治疗肾结石的效果很好,但我们仍缺少预防肾结石形成及复发的有效措施。有报道显示通过饮食干预可以降低肾结石的风险,比如地中海饮食依从性高的人群患肾结石风险降低。咖啡是全世界最流行的饮品之一,但饮用咖啡是否能降低肾结石发病风险尚无定论。因此,本研究目的是借助公共数据库的大数据样本和多种分析方法明确咖啡摄入和肾结石风险之间的关系。进一步的,我们将使用动物模型,基于肠道菌群和血清代谢物的变化,对咖啡抑制肾草酸钙结晶形成的机制进行探索。材料和方法:本研究分别从以下四个部分展开。1.第一部分:通过提取2007年至2018年美国国家健康与营养调查（National Health and Nutrition Examination Survey,NHANES）中的大样本数据,将是否患结石作为因变量,以咖啡作为自变量,采用多因素的加权Logistic回归分析咖啡摄入与肾结石风险之间的相关性。2.第二部分:系统性检索目前公开发表关于咖啡与肾结石风险相关性的原始研究文献,通过Meta分析方法定量合并效应量,再次评估咖啡与肾结石风险之间的相关性,并和我们的大数据分析结果相比较。3.第三部分:基于公开大样本全基因组关联研究（Genome-Wide Association Studies,GW AS）,提取和暴露因素（咖啡）及结局（结石）相关的单核苷酸多态性（single nucleotide polymorphisins,SNPs）作为工具,采用两样本孟德尔随机化（Two-sample Mendelian randomization,MR）方法,更进一步明确咖啡摄入与肾结石风险降低之间的因果关系。4.第四部分:基于乙二醇构建的大鼠肾草酸钙结石模型,观察咖啡干预是否能降低大鼠肾草酸钙结晶的形成;同时使用16s rRNA扩增测序分析大鼠肠道菌群的变化;基于液相色谱-质谱联用技术（Liquid Chromatography-Mass Spectrometry Technique,LC-MS）分析大鼠血清代谢物及代谢途径的差异,初步探索咖啡抑制肾草酸钙结晶形成的机制。结果:1.第一部分:NHANES 2007-2018年度参与调查的人群为59,842例,符合研究纳入、排除标准者共26,123例,经加权后相当于183,399,374例样本量被纳入研究。与不饮用咖啡人群相比,每日摄入咖啡1-2杯（OR:0.79,95%CI:0.66-0.93,p=0.006）和 2-3 杯（OR:0.68,95%CI:0.54-0.85,p=0.001）的人群患肾结石风险明显降低。亚组分析显示,咖啡摄入在某些特定人群中依然存在保护作用,在肥胖人群中发现每日摄入咖啡1-2杯（OR:0.73,95%CI:0.60-0.89,p=0.002）和 2-3 杯（OR:0.65,95%CI:0.50-0.83,p=0.001）的人群较非咖啡摄入人群患者结石风险降低;在糖尿病人群中发现,每日摄入咖啡大于2杯且少于3杯的人群较非咖啡摄入人群患肾结石风险降低约 45%（OR:0.54,95%CI:0.34-0.87,p=0.011）。以咖啡因摄入量的四分位间距分组,第四分位组人群患肾结石风险较第一分位组明显降低约 15%（OR:0.85,95%CI:0.72-0.99,p=0.037）。和非咖啡摄入人群比较,发现每日摄入脱咖啡因咖啡大于3杯的人群,肾结石发病风险降低约 76%（OR:0.24,95%CI:0.07-0.89,p=0.032）。2.第二部分:经过系统性的检索和纳入排除标准筛选,本次研究共纳入9项观察性研究,纳入研究总人群数为810316人,涵盖了欧洲、美洲、亚洲的人群。因纳入的原始研究的I2=81.1%,p<0.001,所以采用随机效应模型。Meta分析结果显示,摄入咖啡组比未摄入咖啡组的肾结石发病风险下降约 14%（ES:0.859,95%CI:0.800-0.924,p<0.001）。亚组分析结果显示,队列研究亚组中摄入咖啡人群较未摄入咖啡人群患肾结石发病风险低约 12%（ES:0.886,95%CI:0.824-0.953,p<0.001）;以男女性别为亚组分析显示,在男性（ES:0.856,95%CI:0.809-0.965,p<0.001）和女性（ES:0.893,95%CI:0.818-0.975,p<0.001）中,咖啡摄入与肾结石风险降低均相关;以咖啡是否含咖啡因进行亚组分析,和未摄入咖啡人群比较,摄入含咖啡因咖啡（ES:0.845,95%CI:0.752-0.950,p<0.001）与脱咖啡因咖啡（ES:0.887,95%CI:0.846-0.929,p<0.001）人群患肾结石风险均降低。采用漏斗图、Begg’s检验分析发现所纳入研究漏斗图基本对称,Begg’s检验p=0.171,表明并未有明显证据显示本研究具有发表偏倚。单项研究剔除法进行的敏感性分析显示总体结果基本稳定。3.第三部分:本部分研究基于已发表的大样本GWAS数据进行MR分析。咖啡的基因工具来源于英国生物样本库发表的关于苦味和甜味饮料消费GW AS研究（共375,833参与者,且全为欧洲血统）,经过连锁不平衡（设置连锁不平衡参数r2阈值为0.001,遗传距离为10,000kb）和多效性评估筛选出7个符合标准的工具变量SNPs;咖啡因的基因工具同样来源于使用英国生物样本库（407,072参与者）的GWAS研究,经过上述同样的标准筛选出12个符合标准的工具变量SNPs。肾结石的GW AS数据来源于2020年芬兰生物银行数据库（肾结石病例3,856例,非肾结石172,757例）。使用逆方差加权法（Inverse variance weighting,IVW）（OR:0.986,95%CI:0.975-0.998,p=0.019）和加权中位数法（Weighted median estimator,WME）（OR:0.987,95%CI:0.976-0.999,p=0.036）分析方法发现咖啡和肾结石风险具有因果关系。Cochran’s Q检验并未发现明显的异质性（p=0.247）,MR-Egger intercept检验p=0.581,MR-PRESSO分析并未发现明显离群点（p=0.344）,因此没有证据显示我们所选的工具变量存在多效性。同样的,IVW（OR:0.336,95%CI:0.156-0.724,p=0.005）和 WME（OR:0.386,95%CI:0.152-0.980,p=0.045）分析方法发现咖啡因和肾结石风险也具有因果关系。Cochran’s Q检验未发现明显的异质性检验（p=0.224）,MR-Egger intercept检验p=0.219,MR-PRESSO分析并未发现明显离群点p=0.282,因此没有证据显示我们所选的工具变量存在多效性。单项研究剔除法进行的敏感性分析显示总体结果基本稳定。4.第四部分:我们成功的使用乙二醇（1%v/v）喂养大鼠四周构建了草酸钙结晶动物模型（EG组）。和对照组相比,EG组的肾脏组织结晶沉积明显。咖啡加乙二醇（EG-Coffee）组肾脏组织病变明显较EG组轻,仅部分区域的肾小管内可见草酸钙结晶沉积,大约占20%,少有连成片。检测各个组别的血尿指标发现血肌酐有明显统计学差异,p<0.05,尿钙水平有统计学差异,p<0.01。此外,我们发现咖啡能改善EG造成的肠道菌群的破坏。和EG组比较,EG-Coffee组Alpha多样（丰富度Shannon指数及均匀度Shannoneven指数）明显较高,p<0.05;使用PCoA分析发现组间Beta多样性有明显差异（p=0.005）;在“属”水平上使用线性判别分析（linear discriminant analysis Effect Size,LEfSe）分析 EG 组和 EG-Coffee 组之间的显著差异菌（LDA>2.5,p<0.05）。发现:和EG-Coffee组相比,EG组的肠道菌群中与损害肠道粘膜屏障及参与炎症介导的代谢性疾病相关的菌属增加,如Corvnebacterium,Ralstonia等。而 EG-Coffee 组中,以下有益菌:Brevibacterium、norank_fMuribaculaceae>UCG₀03、Candidatus_Soleaferrea、norank_fButyricicoccaceae 比例升高,其参与保护肠道粘膜屏障功能及减轻炎症反应、短链脂肪酸产生等相关。代谢组学分析,发现:EG-Coffee组与EG组在血清代谢物水平上差异明显,主要集中在以下几个方面:色氨酸代谢通路（Tryptophan metabolis）,咖啡因代谢通路（Caffeine metabolism）,D-谷氨酰胺和D-谷氨酸代谢通路（D-Glutamine and D-glutamate metabolism）,鞘脂代谢通路（Sphingolipid metabolism）,甘氨酸、丝氨酸和苏氨酸代谢通路（Glycine,Serine and Threonine metabolism）,苯丙氨酸代谢通路（Phenylalanine metabolism）。结论:1.利用NHANES数据库进行大数据分析发现咖啡摄入和肾结石风险具有负相关。同时对已发表文章结果进行Meta分析,二者的结果一致。这表明咖啡摄入和降低肾结石风险之间具有相关性。2.明确咖啡摄入和肾结石风险为负相关之后,我们利用大样本GWAS数据,采用两样本孟德尔随机化分析,规避了传统研究潜在的混杂因素和逆相关风险,证实了咖啡摄入能降低肾结石风险,二者具有因果关系。3.证明咖啡摄入和肾结石风险下降有因果关系之后,我们又继续基于乙二醇诱导的肾草酸钙大鼠模型,再次证实了咖啡摄入可以减少肾结石结晶的形成。同时,我们还发现咖啡能显著改善被乙二醇破坏的肠道菌群结构和功能。4.咖啡除了影响菌群,还能影响大鼠血清代谢物,变化主要集中在氨基酸、咖啡因、鞘脂相关功能代谢方面。更多还原

【Abstract】 Objective:Renal stone is a disease with a high incidence,a high recurrence rate,and a heavy economic burden.Preventing the formation and recurrence of renal stone is a huge challenge facing modern urology.Although the effect of the current Endourology satisfactory surgical treatment of kidney stones,we still lack effective measures to prevent the formation and recurrence of renal stones.It has been reported to reduce the risk of kidney stones by dietary intervention,such as the Mediterranean diet high compliance to reduce risk of people suffering from renal stones.Coffee is one of the most popular drinks in the world,but it is still uncertain whether drinking coffee can reduce the risk of renal stones.Therefore,the purpose of this study was to identify the relationships between coffee consumption and renal stone risk using large data samples from public databases and multiple analysis methods.Furthermore,we will use animal models to conduct preliminary exploration of its mechanism based on and serum metabolites.Materials and Methods:This research is carried out from the following four parts.1.The first part:We extracted large sample data from the National Health and Nutrition Examination Survey（NHANES）2007-2018,taked stone or not as the independent variable and coffee as the independent variable.We used factor weighted Logistic regression analysis of the relationship between coffee intake and the risk of renal stones.2.The second part:Systematic retrieval of the current published original research literature on the correlation between coffee and kidney stone risk.Meta-analysis was used to quantify the combined effect size to evaluate the correlation between coffee and renal stone risk.3.The third part:We Used the open large-sample Genome-Wide Association Studies（GWAS）,using the two-sample Mendelian randomization（MR）method to further clarify the relationship between coffee intake and kidney The causal relationship between the risk reduction of stones..4.The forth part:We used EG to develop renal CaOx stones model rats,observed whether coffee intervention can reduce the formation of rat kidney calcium oxalate crystals;at the same time,16s rRNA amplification and sequencing are used to analyze the intestinal flora of rats Changes;Based on the Liquid Chromatography-Mass Spectrometry Technique（LC-MS）analysis of rat serum metabolites and differences in metabolic pathways,preliminary exploration of the mechanism by which coffee inhibits the formation of renal calcium oxalate crystals.Results:1.The first part:NHANES 2007-2018 participated in the investigation of 59,842 cases,in accordance with the research,excluding the standards of 26,123 cases,and the sample was incorporated in the same amount of 183,399,374 after weighting.Compared with the non-drinking coffee crowd,1-2 cups per day（OR:0.79,95%CI:0.66-0.93,P=0.006）and 2-3 cups（OR:0.68,95%CI:0.54-0.85,P=0.001）The risk of stone is significantly reduced.Subgroup analysis shows that coffee intake still has a protective effect in certain specific people,and found 1-2 cups per day in the obese population（OR:0.73,95%CI:0.60-0.89,P=0.002）And 2-3 cups（OR:0.65,95%CI:0.50-0.83,P=0.001）Crowd of patients with non-coffee in patients with decreased stones;found in diabetic populations,daily intake of coffee is greater than 2 cups and Less than 3 cups of crowds were reduced by kidney stones in crowds by kidney stones（OR:0.54,95%CI:0.34-0.87,P=0.011）.The quadrant spacing of caffeine intake,the fourth group of people suffers from kidney stones,and the first zip group is significantly reduced by 15%（OR:0.85,95%CI:0.72-0.99,P=0.037）.Compared with non-coffee drinkers,those who consumed three or more cups of decaffeinated coffee a day had a 76 percent lower risk of developing renal stones（OR:0.24,95%CI:0.07-0.89,P=0.032）.2.The second part:After systematic retrieval and screening of inclusion and exclusion criteria,this study included a total of 9 observational studies.The total number of people included in the study was 810,316,including populations from Europe,America,and Asia.There is heterogeneity among the studies,so the random effects model is used for data merging analysis.Meta-analysis showed that the risk of kidney stones in the coffee-intake group was about 14%lower than that in the non-coffee-intake group（ES=0.859,95%CI:0.800-0.924,P<0.001）.The results of the subgroup analysis showed that the coffee consuming people in the cohort study subgroup had a 12%lower risk of kidney stones than those who did not consume coffee（ES=0.886,95%CI:0.824-0.953,P<0.001）;in the gender subgroup classification,compared with men who did not consume coffee,the risk of kidney stones in men who consumed coffee was reduced by 14%（ES=0.856,95%CI:0.809-0.965,P<0.001）.Compared with women who did not consume coffee,the risk of kidney stones in women who consumed coffee was reduced by 11%（ES=0.893,95%CI:0.818-0.975,P<0.001）;The subgroup analysis of caffeine-containing coffee intake found that compared with people who did not consume coffee,the consumption of caffeine-containing coffee（ES,0.845,95%CI:0.752-0.950,P<0.001）and decaffeinated Coffee（ES,0.887,95%CI:0.846-0.929,P<0.001）people have a lower risk of kidney stones.Funnel plot and Begg’s test analysis were used to test whether the included studies had publication bias.The results showed that the funnel plot of the included studies was basically symmetrical.The Begg’s test P value=0.171.Sensitivity analysis showed that except for one of the included literature,there was no risk of bias literature that greatly affects the stability of the combined results.3.The third part:This part studies MR analysis based on the published large sample GWAS data.Coffee gene tools are derived from the British Biological Sample Library GW AS Research（375,833 Participants,and all European blood）,through the chain imbalance（set the chain unbalance parameter R2 threshold is 0.001,genetics The distance is 10,000 kB）and a multi-effect assessment screening 7 compliant tool variables SNPS;caffeine gene tools are also from GWAS research using the British Biological Sample Library（407,072 Participants）,after the above-mentioned standard screening 12 tool variables that meet the standard SNPS.The GWAS data of renal stones is derived from the 2020 Finnish Biological Bank Database（3,856 renal stones,172,757 cases of non-kidney stones）.Using the Inverse Variance Weighting,IVW)（OR:0.986,95%CI:0.975-0.998,P=0.019）and Weighted Median Estimator,WME（OR:0.987,95%CI）:0.976-0.999,P=0.036)Analysis method found that coffee and kidney stones risk had a causal relationship.Cochran’S Q did not find a significant heterogeneity（P=0.247）,MR-EGGER Intercept test P=0.581,MR-PRESSO analysis did not find significant out of group（P=0.344）,so there is no evidence to show that we selected There is a lot effect in tool variables.Similarly,IVW（OR:0.336,95%CI:0.156-0.724,P=0.005）and WME（OR:0.386,95%CI:0.152-0.980,P=0.045）Analysis method found that caffeine and kidney stones Causal relationship.Cochran’S Q test（P=0.224）did not find a significant heterogeneity.MR-Egger Intercept test（P=0.219）and MR-PRESSO analysis（P=0.282）did not find significant off-group,so there is no evidence to show the tools we selected has multiple effect.Sensitive analysis of single study rejection method showed that the overall results were basically stable.4.The forth part:Our successful use of ethylene glycol（1%V/V）feeding rats constructed a calcium oxalate crystalline animal model（EG group）.Compared with the control group,the crystallization of the kidney tissue of the EG group was obvious.The kidney tissue lesion of Coffee and EG-Coffee group is significantly lower than that of the EG group,and the deposition of rhizophilic acid calcium acid is considered to be found in the renal tubule in part,about 20%,and there are less film.It was found that the blood creatinine had a significant statistical difference in blood creatinine,P<0.01.In addition,we found coffee to improve the destruction of the intestinal flora caused by EG.Compared with the EG group,the EG-Coffee group Alpha Diverse（Shannon Index and Shannoneven Index）is significantly higher,P<0.05;We found that the BETA diversity of the group is significantly different from the PCOA analysis（P=0.005）;We used Linear Discriminant Analysis Effect Size（LEFSE）to analyze significant differences between the EG group and the EG-Coffee group（LDA>2.5,P<0.05）.The number of bacteria related to intestinal mucosal barrier damage and metabolic diseases mediated by inflammation increased in EG group,such as Corynebacterium,Ralstonia.In the EG-Coffee group,the following beneficial bacteria:Brevibacterium、norank_fMuribaculaceae、UCG₀03、Candidatus_Soleaferrea、norank_fButyricicoccaceae rising,which participated in the protection of intestinal mucosal barrier function and reduced inflammatory response,short chain fatty acid production.Metabolic syndrome analysis,found that the difference between the EG-Coffee group and the EG group was obvious in serum metabolite,mainly in the following aspects:D-Glutamine and D-glutamate metabolism,tryptopter metabolis,caffeine metabolism,sphingolipid metabolism,glycine,serine and threonine metabolism（Glycine,Serine and Threonine metabolism）,Phenylalanine metabolism.更多还原