| Background and ObjectivesOvarian hyperstimulation syndrome(OHSS),an iatrogenic disease that occurs after controlled ovarian hyperstimulation(COH),is one of the most common and serious complications during assisted reproductive technology(ART)and often develops kidney injury.Renal injury refers to abnormal renal function or structure(including abnormal blood,urine,histology,imaging and other examinations).Acute kidney injury(AKI)is a narrow definition focused on the assessment and prevention.Previous studies have shown that patients with OHSS often have kidney injury,which may be manifested as oliguria.Some patients may have nephrotic syndrome,obstructive nephropathy,etc.However,there is no relatively systematic study about AKI risk factors and early evaluation and diagnosis in patients with OHSS.At present,there is no specific diagnostic criteria for AKI in patients with ovarian hyperstimulation syndrome,and clinical practice still mainly refers to the AKI clinical practice guidelines developed by the Global Kidney Disease Outcome(KDIGO).The basis for AKI diagnostic criteria established by KDIGO is mainly derived from the non-pregnant population.For patients with ovarian hyperstimulation syndrome,due to the impact of physiological and pathophysiological changes caused by ovulation induction,pregnancy and other factors on the internal environment,there may be some differences in AKI criteria established with reference to KDIGO in theory;especially in the absence of monitoring the patient’s basal creatinine level and long-term urine volume,it is difficult to timely and accurately assess whether the patient has AKI according to AKI KDIGO diagnostic criteria in clinical practice,thus affecting early identification and early diagnosis.Therefore,it is necessary to carry out the study on single indicator or multi-indicator combined assessment and diagnosis of AKI in OHSS patients.There are relatively many research reports on early prediction or auxiliary diagnosis of AKI in patients with similar conditions in other fields,such as intensive care unit,after surgery,etc.However,by referring to domestic and foreign literatures,there is no relevant research report on early prediction or auxiliary diagnosis of AKI in patients with ovarian hyperstimulation syndrome.Based on the above factors,this topic uses a retrospective study method to collect the clinical data of patients with ovarian hyperstimulation syndrome admitted to our critical maternal treatment center,analyze the clinical characteristics of renal injury in patients with OHSS,explore the risk factors of acute kidney injury in patients with ovarian hyperstimulation syndrome,try to use statistical analysis methods,explore the auxiliary methods of multi-index combined assessment and diagnosis of AKI in patients with OHSS,in order to provide some basis and help for the early prevention,early identification and early diagnosis of AKI in patients with OHSS.Contents and MethodsThe clinical data of patients diagnosed with moderate to severe ovarian hyperstimulation syndrome during hospitalization in the Third Affiliated Hospital of Guangzhou Medical University from December 2012 to December 2020 were collected,ineligible cases were excluded,and eligible cases were included in the cohort to analyze the clinical features of their kidney injury.Then,hospitalized patients from December 2012 to December 2018 were selected from all study cases to form the study cohort.According to the presence or absence of acute kidney injury,the study cohort was divided into AKI group and non-AKI group;the risk factors of AKI were analyzed;finally,according to the risk factors of AKI in patients with ovarian hyperstimulation syndrome and laboratory tests during hospitalization,a statistical model was constructed.In addition,cases from January 2019 to December2020 will be included in the validation cohort to validate the model.Collect the clinical data of patients: including:(1)clinical data before ovulation induction: age,history of pregnancy and delivery,history of ovarian syndrome,basic biochemical indicators and anti-Müllerian hormone levels;(2)sex hormones on the day of human chorionic gonadotropin injection: estradiol on the day of HCG,luteinizing hormone on the day of HCG,progesterone on the day of HCG and other hormone indicators;(3)whether there is embryo transfer and the number of oocytes retrieved;(4)relevant clinical data after admission: including height,weight,urine volume,blood type and other basic information,blood routine,coagulation blood routine,biochemical indicators and urinalysis data.The data of each cohort were statistically analyzed.Independent sample t test was used for measurement data conforming to normal distribution(data were expressed as mean ± standard deviation);chi-square test was used for enumeration data or ordinal categorical variables;Wilcoxon signed rank test was used for measurement data not conforming to normal distribution(data were expressed as median and interquartile range);P < 0.05 indicated statistically significant difference.The ROC curve(receiver operating characteristic curve)will be constructed for the quantitative data with statistical difference,and the cutoff value will be determined.Univariate analysis and multivariate logistic regression analysis were performed on the clinical data,ROC curve was drawn,and logistic model equation was derived.This equation was used as an auxiliary diagnostic model,and the calculated value of the model was named OHHS-AKI auxiliary diagnostic score(OHHS-AKI-ADS).The efficacy of the auxiliary diagnostic model was assessed by the area under the curve.The auxiliary diagnostic model was substituted into the validation cohort for external validation,and the auxiliary diagnostic efficacy of this model for AKI in patients with ovarian hyperstimulation syndrome was again assessed.Study results1 Case screening results and cohort setting1.1 Total cohort: A total of 818 patients diagnosed with moderate to severe ovarian hyperstimulation syndrome were admitted from December 2012 to December2020 for screening,and 566 eligible cases who met the criteria were included in the cohort.1.2 Study cohort: 496 eligible cases who met the criteria between December2012 and December 2018 were included as the study cohort.1.3 Validation cohort: 70 eligible cases who met the criteria were included in the validation cohort between January 2019 and December 2020.2 Analysis of main clinical manifestations of renal injury in total corps2.1 There were 144 cases(25.4%)of AKI,69 cases(12.2%)of hematuria,and46 cases(8.1%)of proteinuria in total cohort.2.2 There were 105 cases(18.6%)of simple AKI,39 cases(6.9%)of simple hematuria,11 cases(1.9%)of simple proteinuria,15 cases(2.7%)of AKI with hematuria,20 cases(3.5%)of AKI with proteinuria,4 cases(1.0%)of AKI with hematuria and proteinuria,and 11 cases(1.9%)of proteinuria with hematuria.2.3 131 cases(23.1%)met the AKI diagnostic criteria for decreased urine volume,27 cases(4.8%)met the diagnostic criteria for elevated creatinine,and 14cases(2.5%)met the diagnostic criteria for both elevated creatinine and decreased urine volume.3 Analysis of risk factors for AKI in study cohort and construction of auxiliary diagnostic model3.1 Univariate analysis of AKI related risk factors in the study cohort3.1.1 128 patients in the AKI group,with a mean age of(30.19 ± 3.42)years;368 patients in the non-AKI group,with a mean age of(30.27 ± 3.73)years,t = 0.225,P = 0.822,and there was no significant difference in age between the two groups.3.1.2 There were 47 cases of AKI with adverse pregnancy history and 128 cases of non-AKI,and there was no significant difference between the two groups(chi-square test,chi-square value = 0.156,P = 0.693).Mann-Whitney rank test was performed for gravidity and parity between AKI group and non-AKI group,and the difference was not statistically significant.3.1.3 There are 21 cases had a history of PCOS in AKI group and 34 cases in non-AKI group.The chi-square test showed that the chi-square value = 4.948 P =0.026,and the difference was statistically significant.3.1.4 There were 62 cases with embryo transfer in AKI group and 125 cases in non-AKI group.The chi-square test was performed for whether there was embryo transfer.The chi-square value was 8.466,P = 0.004,and the difference was statistically significant.3.1.5 Wilcoxon signed-rank test was performed for alanine aminotransferase,aspartate aminotransferase,urea nitrogen,creatinine,uric acid,blood glucose,total cholesterol,and triglyceride levels before ovulation induction between the AKI and non-AKI groups,and there were no significant differences(p > 0.05).3.1.6 The median AMH level(interquartile range)was 9.14(5.86,14.31)ng/ml in AKI group and 6.01(3.63,9.59)ng/ml in non-AKI group;Z =-4.644,p < 0.001 by signed-rank test between the two groups,and there was significant difference in AMH level between AKI group and non-AKI group.AMH was included in the ROC receiver operating characteristic curve,and its AUC was 0.668,(95%CI 0.611-0.711),Z value = 4.933,p< 0.0001;the optimal cutoff value was > 8.4 ng/m L,when the maximum Youden index was 0.2837,sensitivity was 56.52%,and specificity was71.85.3.17 The median(interquartile range)of estradiol on HCG day,luteinizing hormone on HCG day,and progesterone on HCG day were ≥ 18350(14172,≥ 18350)pmol/l,0.81(0.59,1.10)U/L,and 2.70(1.9,3.9)nmol/l in the AKI group and 15745(11613,≥ 18350),0.88(0.63,1.22),and 2.3(1.7,3.6)in the non-AKI group,respectively.There was a significant difference in estradiol on Hcg day between the two groups,Z =-3.75,p value < 0.0001,and there was no significant difference in luteinizing hormone on h CG day and progesterone on h CG day(p > 0.05),and estrogen levels on the day of h CG were included in the ROC receiver operating characteristic curve,and its AUC is 0.616(95% CI 0.568-0.662),Z = 4.012,p <0.0001;the optimal cutoff value was > 17958 pmol/l,when the maximum Youden index was 0.2142,the sensitivity was 60.04%,and the specificity was 61.41%.3.1.8 The number of oocytes retrieved was 22(15,31)in AKI group and 18(13,24)in non-AKI group,with statistically significant difference between the two groups(Z =-2.579,p = 0.01).The number of oocytes retrieved was included in the ROC receiver operating characteristic curve,with an area under the curve of 0.602(95%CI0.542-0.661),Z = 2.534,p = 0.0113;the optimal cutoff value was > 21,when the maximum Youden index was 0.1928,the sensitivity was 52.78%,and the specificity was 66.50%.3.2 Construction of auxiliary diagnostic model related to AKI in study cohort3.2.1 Construction of auxiliary diagnostic model: The above indicators with statistical difference were included in the binary logistic regression equation,and the regression equation was established by the stepwise method.After simplification,the model equation was finally obtained: OHHS-AKI auxiliary diagnostic score =UA-5.75 ALT + 18.5TBIL + 14.75WBC-316AMH-11.25WBC(AMH > 8.4 ng/ml was marked as 1,otherwise it was marked as 0).3.2.2 Internal validation of auxiliary diagnostic model: test this model :H-L test X2 = 4.558,P = 0.804(> 0.05).The area under the ROC curve for the auxiliary diagnostic model was 0.793(95%CI 0.761-0.822),Z=14.481,p < 0.0001.When the OHHS-AKI auxiliary diagnostic score was > 391,the sensitivity of diagnosing AKI in ovarian hyperstimulation syndrome was 73.60% and the specificity was 70.17%.4 External validation of auxiliary diagnostic modelThe validation cohort of 70 cases(54 non-AKI patients + 16 AKI patients)was used for discrimination assessment of the model.By plotting the ROC curve,the AUC was calculated as 0.809,p < 0.0001.This model had a sensitivity of 68.8% and a specificity of 74.1% in the validation cohort.Conclusion1 The clinical manifestations of renal injury in moderate to severe ovarian hyperstimulation syndrome are mainly three basic types: acute kidney injury,hematuria,and proteinuria,which may occur alone or at the same time.2 History of ovarian syndrome,embryo transfer,number of oocytes retrieved,AMH,and estrogen on the day of HCG are all risk factors for acute kidney injury in patients with ovarian hyperstimulation syndrome,respectively.3 The multi-index combined auxiliary diagnostic model constructed in this study has an early warning effect on the early identification and early diagnosis of acute kidney injury in ovarian hyperstimulation syndrome. |
PDF Full Text Request