A Clinical Study Of Hip Arthrography Combined With MRI To Assess Acetabular Development After Early Treatment Of Ddh In Children

Background: Developmental dysplasia of the hip(DDH)is one of the common skeletal muscle disorders in children,and even after successful early conservative treatment,there is a high risk of residual acetabular dysplasia(RAD)in the affected hip thus requiring secondary surgery.The current study suggests some predictors for clinical reference,but if the hip with the poor future outcome can be identified at an earlier stage for earlier secondary surgery by some predictors,the greater the developmental potential left for the child.Purpose: One of the objectives of this study was to analyze the concordance between the measurements of the same cartilaginous coverage index in arthrogram and magnetic resonance imaging(MRI)to explore the alternative or complementary implications of the two imaging modalities;Second,the cartilaginous acetabular index(CAI)and cartilaginous center-edge angle(CCE)were mentioned in the literature as two different principles of measurement,and we performed a Meta-analysis to compare their diagnostic and prognostic efficacy;Third,We used a combination of hip arthrography and MRI and a combination of multiple indicators to predict RAD after early treatment to find a more stable and accurate early RAD prediction model.Patients and methods: The clinical data of 25 children with developmental dysplasia of the hip(DDH)who underwent closed reduction and cast fixation after hip arthrography at Pediatric Orthopaedic Hospital,Honghui Hospital,Xi’an Jiaotong University,Xi’an from December 2020 to May 2022 were included.Based on the exclusion criteria,57 data sets from MRI and arthrogram were finally included for analysis.The acetabular index(AI),labral acetabular index(L-AI),center-edge angle(CEA),labral center-edge angle(L-CEA),and femoral head coverage(FHC)were measured in the intraoperative arthrogram and MRI before or after surgery within 3days.The intraclass correlation coefficient(ICC)was used to assess the concordance of the same index measured in MRI and arthrogram.Pearson correlation analysis was used to verify the relationship between bony coverage and cartilaginous coverage indicators.A Meta-analysis was performed to compare the diagnostic performance of the cartilaginous acetabular index(CAI)and cartilaginous center-edge angle(CCE)measures.We defined CAI and L-AI as the CAI group and CCE and L-CEA as the CCE group.Databases such as Pub Med,Embase,Cochrane,and Web of Science were searched to screen the literature.The quality of the literature was assessed by the QUADAS-2 tool.A quantitative synthesis of the literature was performed based on extracted data.For quantitative synthesis studies,the sensitivity,specificity,diagnostic odds ratio(DOR),and summary receiver operating characteristic(SROC)curves with corresponding confidence intervals were calculated.Thirteen children who met the inclusion criteria between December 2020 and December 2021 were included,with a total of 21 hips(8 with bilateral DDH),and the final remaining 15 hips were statistically analyzed according to the exclusion criteria.The acetabular cartilaginous angle(ACA)and femoral head coverage(FHC)were measured in the initial intraoperative arthrogram,the cartilaginous acetabular index(CAI)and cartilaginous center-edge angle(CCE)was measured in the MRI after the initial closed reduction.The AI in the one-year follow-up x-ray was compared with the corresponding normal AI range in Chinese children and divided into satisfactory and unsatisfactory groups.ICC was used to assess intra-and inter-observer consistency on the same index.Independent samples t-test and Fisher’s exact test were used to compare the measurement data and count data between the two groups,respectively.The difference was considered statistically significant at P < 0.05.The prognostic efficacy of CAI,CCE,ACA,FHC,and different combinations of each index was evaluated by the area under the curve(AUC)of the receiver operating characteristic(ROC)curve.Results: In MRI and arthrogram,the ICC values of AI and CEA were high,0.765(95 % CI: 0.768 ～ 0.912)and 0.765(95 % CI: 0.628 ～ 0.852),respectively.The ICC values of L-AI,L-CEA,and FHC were moderate,0.719(95 % CI: 0.562 ～ 0.822),0.714(95% CI: 0.606 ～ 0.842)and 0.738(95% CI: 0.590 ～ 0.835).Pearson correlation analysis showed a positive correlation between AI and L-AI in both MRI(r = 0.472 >0;P < 0.001)and arthrogram(r = 0.544 > 0;P < 0.001).CEA and L-CEA were also positively correlated in both MRI(r = 0.672 > 0;P < 0.001)and arthrogram(r = 0.747 >0;P < 0.001).Meta-analysis results showed that for the CAI group,the combined values of sensitivity,specificity,and DOR were 0.80(95% CI=0.54-0.93),0.73(95% CI=0.57-0.84),and 10.62(95% CI=3.96-28.53),respectively.The corresponding values for the CCE group were 0.71(95% CI=0.57-0.82),0.78(95% CI=0.66-0.87),and 8.64(95%CI=3.08-24.25).The area under the curve(AUC)for the SROC curve in the CAI and CCE groups was 0.82(95% CI = 0.78-0.85)and 0.80(95% CI = 0.76-0.83).The CAI group had higher sensitivity,DOR,and AUC than the CCE group.In the analysis of arthrography combined with MRI to assess developmental regression after closed reduction,both intra-and inter-observer consistency was good for the same index(ICC > 0.75).Independent sample t-test results for the satisfactory and unsatisfactory groups showed the difference between age was not statistically significant(t=-1.144,P=0.273 > 0.05);the difference between follow-up time was not statistically significant(t=-0.533,P=0.603 > 0.05);the difference between AI at the last follow-up was statistically significant(t=-3.454,P < 0.01);the difference between CAI was not statistically significant(t=-0.953,P= 0.358 > 0.05);the difference between CCE was not statistically significant(t= 0.169,P= 0.868 > 0.05);the difference between ACA was statistically significant(t=-3.099,P < 0.01);the difference between FHC was not statistically significant(t=1.494,P=0.159>0.05).The results of the Fisher’s exact test between the satisfied and unsatisfied groups showed that the difference between gender was not statistically significant(P=1.000>0.05);the difference between disease sides was not statistically significant(P=1.000>0.05);the difference between Tonnis typing was not statistically significant(P=0.608>0.05).The ROC curves of each early warning indicator showed that when assessing short-term prognosis with individual cartilaginous indicators,CAI and CCE had AUC values between 0.5 and 0.7,which were in the low level;while ACA and FHC had AUC values between 0.7 and 0.9,which were in the medium level.While the combination of indicators can improve the diagnostic efficacy,in the analysis of the combination of two indicators,CAI combined with ACA has the highest AUC value of 0.893,which belongs to the medium level.When the three indicators were combined,the combination of CAI+CCE+ACA had the highest AUC value of 0.946,which is a high level.And the combination of all four indicators had the maximum AUC value of 0.982,reaching a high level of diagnostic efficacy.Conclusions: There is a good concordance between MRI and arthrography in the observation of bony coverage and cartilaginous coverage,which is sufficient to demonstrate the concordance between the two in general observation.Both can achieve substitution compensation for qualitative observations of cartilage and soft tissue,but their moderately slightly upper concordance is not sufficient to achieve accurate substitution of quantitative information such as angulation.For both CAI group and CCE group measurements,they have relatively good diagnostic accuracy,and CAI/LAI has a slight advantage over CCE/L-CEA.However,their application as separate indications for RAD orthopedic surgery is still unreliable.It is worth mentioning that the combination of multiple indicators and multiple imaging examinations can form a comprehensive and systematic prediction model,which can significantly improve the accuracy of single indicator prediction,and our validated "ACA+FHC+CAI+CCE" is a reliable prediction model.This model can provide early warning of acetabular developmental regression after closed reduction and guide possible secondary orthopedic surgery.