| ObjectiveWe aim to investigate the exposure of urinary antibiotics and develop a method to measure the levels of short-chain organic acids in the elderly.To investigate the association between antibiotic exposure,short-chain organic acids levels and geriatric depression and to analyze the mediating effect of short-chain organic acids in the association between antibiotic exposure and geriatric depression.MethodsThe study based on the Environmental and Health Controllable Factors in Older Adults Cohort in Lu’an City.According to the inclusion and exclusion criteria,984 subjects were finally included.A structured questionnaire was self-designed to investigate the basic demographic characteristics,lifestyle habits,cognitive level and depressive symptoms of the elderly.Morning urine and fasting blood samples were collected,packaged and frozen.Ultra-high performance liquid chromatography tandem mass spectrometry was used to detect the levels of urinary antibiotics and short chain organic acids in serum.The concentration of antibiotics or short-chain organic acids in the undetected subjects was replaced by 1/2 limit of detection(LOD).The level of internal exposure to antibiotics in the elderly was described by the concentration of antibiotics corrected with urine creatinine.The Chi-square test was used to analyze the related factors of depressive symptoms in the elderly.And the associations between antibiotic exposure,short-chain organic acids and geriatric depression were analyzed by multiple linear regression.Binary logistic regression was used to explore the association between antibiotic exposure level and the risk of depression in the elderly,and restricted cubic splines were used to analyze the dose-response relationship between antibiotic exposure or the level of short-chain organic acids and the risk of geriatric depression.The mediating effect of short-chain organic acids in the association between antibiotic exposure and geriatric depression was explored by the mediation analysis.ResultsThe data of 984 older adults were analyzed in the study,including 500(50.9%)elderly with ages ≥ 70 years and 447(45.4%)males.A total of 273(27.7%)elderly with depressive symptoms were detected.The results of the Chi-square test indicated that gender,physical exercise,drinking,education level,living alone,cognitive function,marital status,and diet were associated with depressive symptoms in the elderly.The detection rates of individual antibiotics in subjects’ urine ranged from 0.2% to 35.5%,and antibiotics with detection rates ≥ 10.0% were azithromycin,penicillin V,sulfaclozine,trimethoprim,tetracycline,oxytetracycline,doxycycline,ofloxacin,ciprofloxacin,norfloxacin,enrofloxacin and florfenicol.Multiple linear regression models found positive association between sulfaclozine(β = 0.456,95% CI: 0.034,0.878)or veterinary antibiotics(VAs,β = 0.666,95% CI: 0.002,1.329)with geriatric depression scale-30(GDS-30)scores.Binary logistic regression model found that exposure of high concentrations of sulfaclozine(OR=1.557,95% CI: 1.059,2.291),tetracyclines(OR=1.468,95% CI: 1.008,2.138)or VAs(OR=1.477,95% CI: 1.028,2.123)was associated with an increased risk of geriatric depression.The results of the restricted cubic spline analysis showed a significant non-linear association between sulfaclozine exposure and the risk of depression in the elderly.Eleven short-chain organic acids were examined in the study,including acetic acid(geometric mean: 26147.13 nmol/L),propionic acid(245.21 nmol/L),crotonic acid(12.14 nmol/L),iso-butyric acid(325.31 nmol/L),butyric acid(62.85 nmol/L),iso-valeric acid(473.90 nmol/L),valeric acid(8.48 nmol/L),iso-caproic acid(0.88nmol/L),caproic acid(706.58 nmol/L),lactic acid(3.68 mmol/L),and β-hydroxybutyric acid(BHB,94.21 μmol/L).Multiple linear regression model found each unit increase in the log-transformed iso-caproic acid level was related to the decrease of the GDS-30(β= 1.758,95%CI: 0.250,3.266).However,each unit increase in log-transformed BHB concentration was correlated with the increase of the GDS-30 score(β = 1.812,95%CI:0.190,3.434).In addition,the association between antibiotic exposure and short-chain organic acids was analyzed.Acetic acid,lactate and BHB had positive association with the exposure of sulfaclozine or doxycycline.Conversely,ofloxacin exposure was negatively associated with iso-butyric or isocaproic acid concentrations,and each unit increase in urinary log-transformed tetracycline concentration was associated with decrease in log-transformed caproic acid(β = 0.058,95%CI: 0.017,0.098),iso-butyric(β = 0.061,95% CI: 0.006,0.117)or isocaproic acid(β = 0.061,95% CI: 0.031,0.091).Classificationally,sulfonamides exposure was correlated with increases in acetic acid,lactic acid and BHB,and the concentrations of urinary tetracyclines were positively associated with lactic acid and the increase of log-transformed BHB levels was relative to the increase of VAs.Differently,the negative correlation between the concentration of urinary tetracyclines or preferred as veterinary antibiotics(PVAs)and iso-caproic acid concentrations was statistically significant.The mediation analysis revealed the positive mediating effect of serum BHB in the association of the GDS-30 scores with sulfaclozine(8.6%),sulfonamides(27.2%),or VAs(7.5%)in the elderly,but the mediating effects of BHB was no statistically significant.Similarly,the mediating effect of serum iso-caproic acid in the association of geriatric depression with tetracyclines and PVAs was not statistically significant.However,the mediating effect of serum iso-caproic acid in the association of geriatric depression with ofloxacin or tetracycline was statistically significant,with mediating effects of 25.3% and 46.3%,respectively.ConclusionsAntibiotic exposure can increase the risk of geriatric depression.The increase of GDS-30 scores was related with the exposure of sulfaclozine,VAs or broad-spectrum antibiotics.Antibiotic exposure can similarly affect the levels of short-chain organic acids in the elderly,and the levels of short-chain organic acids were significantly associated with GDS-30 scores.Ofloxacin or tetracycline exposure decreased the concentration of iso-caproic acid in serum,and result in the geriatric depression.And,the mediating effect of iso-caproic acid on the association of geriatric depression with ofloxacin and tetracycline was 25.3% and 46.3%,respectively. |