| 1.2.1 BackgroundThe aorta as a three-dimensional organ manifests both diameter and length.It has been noticed that the aorta elongates with aging.With the popularity of imaging reconstruction techniques,some studies have suggested that aortic elongation may be related to aortic dissection.However,little information is available regarding the longitudinal changes of the aneurysmal ascending aorta.1.2.2 Objectiveswe aim to outline the natural history of ascending thoracic aortic aneurysm(ATAA)based on ascending aortic length(AAL),and develop novel predictive models to further refine the management of patients with ATAA,utilizing in-depth statistical analysis and a much larger database compared to previous reports.1.2.3 MethodsThe ascending aortic diameters and lengths,and long-term aortic adverse events(AAE)(rupture,dissection,and death)of 522 ATAA patients were evaluated using comprehensive statistical approaches.1.2.4 ResultsThe mean age was 65.8±13.6 years and male gender was predominant(72.4%).Overall,the average aortic diameter was 4.8±0.7 cm(range 3.5-9 cm)and AAL was 11.2±1.3 cm(range 7.3-15.4 cm).During a mean follow-up of 42.0 months(range,5 days to 336 months),424(81.2%)patients were AAE-free and 98(18.8%)patients developed AAE.Specifically,64(12.2%)patients suffered type A aortic dissection,5(0.9%)patients suffered rupture,and 31(5.9%)died of ATAA(i.e.aortic deaths).21(4.0%)mortalities could be attributed to causes other than aortic aneurysm(i.e.non-aortic deaths).An AAL of>13 cm was associated with an almost 5-fold higher average yearly rate ofAAE compared to an AAL of<9 cm.Two AAL ’hinge points’ with a sharp increase in the estimated probability of AAE were detected between 11.5-12.0 cm and 12.5-13.0 cm,respectively.A length height index(LHI)of≥7.5 cm/m was associated with a more than 5-fold higher average yearly rate of AAE compared to an LHI of<5.5 cm/m.Multivariable Logistic regression showed that the odds of AAE were 12.4-fold greater in patients with AAL≥13 cm compared to AAL<9 cm(p=0.001)adjusted for age,bicuspid aortic valve,family history,smoking,dyslipidemia,and CKD.Similarly,multivariable Logistic regression revealed that the adjusted odds of AAE were 9.5-fold greater in patients with LHI>7.5 cm/m compared to LHI<5.5 cm(p=0.001).The mean estimated annual aortic elongation rate was 0.18 cm/year,and aortic elongation was age-dependent.Aortic diameter increased 18%due to dissection while AAL only increased by 2.7%.We classified the cohort into two groups based on the aortic height index(AHI)(AHI=diameter height index(DHI)+length height index(LHI))(’high AHI’ group:AHI≥10cm/m(n=112),and ’low AHI group’:AHI<10cm/m(n=410))There were more aortic adverse events in the high AHI group(35.7%,40/112)than in the low AHI group(14.1%,58/410)(p=0.001).We then performed a PSM analysis with 379 patients matched.After matching,there were no persisting differences in baseline characteristics between the high and low AHI groups(Online Figure 3).Matched samples still revealed more aortic adverse events in the high AHI group(35.8%,38/106)than in the low AHI group(17.5%,48/273)(p=0.003),lending further credence to the above results.There was a noticeable improvement in the discrimination of the Logistic regression model(AUC=0.810)due to the introduction of AHI.The AHIs<9.33,9.38 to 10.81,10.86 to 12.50,and≥12.57 cm/m were associated with a～4%,～7%,～12%,and～18%average yearly risk of AAE,respectively.1.2.5 ConclusionsThe study supports the following recommendations or conclusions:1.An aortic elongation of 11 cm serves as a potential intervention criterion for ATAA.2.Aortic length demonstrates a mean growth rate of 0.18cm annually.3.Aortic elongation is age-dependent,and relatively immune to dissection.4.Aortic-Height index(including both length and diameter)(easily discernible via modern imaging modalities)is more powerful than diameter alone in predicting AAE,with a significantly increased area under the ROC curve.The easy-to-use nomogram and 3-D plot provided,incorporating both aortic diameter and length,allow clinical application of this more advanced decision-making tool2.2.1 BackgroundThe normal diameter of the descending thoracic aorta is 2.5±0.2cm in males and 2.2±0.2cm in females,with a 1-2mm and 3-4mm decrease in size in both genders at the diaphragm and infrarenal portion,respectively.Elucidating critical aortic diameters at which natural complications occur(rupture,dissection,death)is paramount to guide timely surgical intervention.However,natural history knowledge for descending thoracic or thoracoabdominal aortic aneurysms(DTTAA)is sparse.Our small early studies recommended repairing DTTAAs before a critical diameter of 7.0 cm.2.2.2 ObjectivesWe aim to do the same for descending and thoracoabdominal aortic aneurysms as we have done for ascending thoracic aortic aneurysm(ATAA).We focus exclusively on a large number of DTTAA followed over time,thereby enabling a more detailed analysis with greater granularity across a range of aortic sizes.2.2.3 MethodsAnthropometric,radiologic,and clinical data were manually accrued retrospectively from individual electronic medical records and hospital charts.All aortic diameter measurements were doubly confirmed by two senior investigators.Aortic diameters and long-term complications of 907 patients with DTTAA were reviewed.Growth rates(instrumental variables approach),yearly complication rates,5-year event-free survival(Kaplan-Meier),and risk of complications as a function of diameter-height index(DHI)(aortic diameter[cm]/height[m])(Competing-risks regression)were calculated.2.2.4 ResultsThe median DTTAA size in centimeters prior to dissection,operation,aortic death,and rupture was 4.1(IQR 3.7-4.7),5.8(IQR 4.6-6.6),6.0(IQR 4.7-7.0),and 6.6(IQR 5.7-6.9),respectively.Estimated mean growth rate of DTTAAs was 0.19 cm/year,increasing with increasing aortic size.Multivariable linear regression analysis of the factors affecting growth rate revealed that patient age and COPD were associated with a higher growth rate,whereas male gender and bovine aortic arch were associated with a slower growth rate(P<.05).There were a total of 289 all-cause deaths among the 907 patients,of which 87 were DTTAA-related(46 definite,41 possible),27 were deaths due to other aortic causes(ascending and abdominal aortic deaths),and 175 were non-aortic,as confirmed by medical records and death certificates.Median size at acute type B dissection was 4.1cm.80%of dissections occurred below 5cm,whereas 93%of ruptures occurred above 5cm.DTTAA diameter>6cm was associated with a 19%yearly rate of rupture,dissection,or death.5-year complication free survival progressively decreased with increasing DHI.Larger DHIs were associated with decreased AAE-free survival probability:the 1 and 5-year freedom from AAE was 98.5%,87.5%,69.3%,84.3%,52.6%,and 49%,and 95.9%,87.1%,64.3%,56.5%,27.1%,and 26.1%for DHIs(cm/m)<1.8,1.8-2.3,2.4-2.9,3-3.5,3.6-4.1,and≥4.2,respectively.The 1 and 5-year freedom from a composite end-point of rupture and aortic death was 99.5%,98.7%,87.9%,90.5%,59.6%,and 50%,and 96.9%,98.2%,81.6%,60.6%,30.5%,and 26.7%for DHIs(cm/m)<1.8,1.8-2.3,2.4-2.9,3-3.5,3.6-4.1,and≥4.2,respectively.Hazard of complications showed 6-fold increase at DHI>4.2 compared to DHI 3.0-3.5(P<.05).Probability of fatal complications(aortic rupture or death)increased sharply at 2 hinge points:6.0 and 6.5 cm.2.2.5 ConclusionsThis study of the natural history of descending thoracic and thoracoabdominal aortic aneurysm permits the following conclusions:1.The descending thoracic and thoracoabdominal aorta grows slowly at 0.19 cm/year.2.Descending aortic dissection and rupture occur primarily at very different aortic sizes:Dissection at dimensions in the 4 cm range,and rupture at dimensions above 5 cm.3.The aortic size thresholds for operative repair in current guidelines would not be expected to afford protection from dissection of the descending thoracic and thoracoabdominal aorta.4.The natural risk of rupture/aortic death based on aortic size increases sharply at two‘hinge points’:6.0cm and 6.5cm.5.We recommend intervention on the descending or thoracoabdominal aorta at 5.0 to 5.5 cm,to prevent aortic related rupture and consequent death.3.2.1 BackgroundAccurate elucidation of the natural history of TAA is crucial for optimal timing of surgical intervention,which affords subsequent protection from adverse aortic events(AAE),including dissection,rupture,and death).A limitation of previous ascending thoracic aortic aneurysm(ATAA)natural history studies is inclusion of patients who eventually underwent elective repair,barring observations of the ’natural’ course of the disease.3.2.2 Objectiveswe aim to trace the fate of non-operated ATAAs with a robust data set,and to develop a multivariable model incorporating both aortic size and other risk factors,to achieve a better,evidence-based clinical decision-making algorithm.3.2.3 MethodsAnthropometric,radiologic,and clinical data were retrospectively accrued and entered into a computerized database from individual electronic medical records and hospital charts,which were cross-referenced with Connecticut State death certificates.For the purpose of our current study,all radiologic studies have been re-verified,re-read,and re-analyzed in a standardized manner.All aortic diameter measurements were doubly confirmed by two senior investigators.Aortic diameters and long-term complications of 636 unoperated patients with ascending aortic aneurysm were reviewed.The aortic complications within one year,cumulative risk of complications,average yearly risk of aortic complications were calculated to show the impact of aortic size on AAE.Aortic growth rate estimates were performed utilizing the instrumental variables(IV)approach.Kaplan-Meier analysis with log-rank test was used to assess the survival.Variables showing p<0.05 in the univariable competing-risk regression were entered into the multivariable competing-risk model.We developed a graphic representation(nomogram)of the prediction model based on regression coefficients.The nomogram was then validated in two steps:calibration plot and ROC curve.3.2.4 ResultsDuring follow-up,560(88.1%)patients were AAE-free and 76(11.9%)patients developed AAE.Specifically,29(4.6%)patients suffered type A aortic dissection,5(0.8%)patients suffered rupture,51(8.0%)patients died of aortic causes(9 definite aortic deaths;42 possible aortic deaths),and 115(18.0%)patients died of non-aortic reasons.Of these 636 patients,442 were male and 194 were female.The average age of the cohort was 67.8±13.7 years(Male:66.6±13.4 years;Female:70.4±13.9 years,p=0.001),with a range of 18 to 92 years.The mean and median sizes of the ascending aorta were 4.73±0.75 cm and 4.6(IQR 4.20-5.10)cm,respectively.In patients who suffered a rupture/acute dissection,aortic death,or overall AAE,the median aortic sizes were 5.20(IQR 4.90-5.80),5.38(IQR 4.97-6.05),and 5.30(IQR 5.00-5.90)cm,respectively.ATAAs grow slowly--at 0.20±0.02 cm/year.At aortic sizes of 3.5-3.9,4.0-4.4,4.5-4.9,5.0-5.4,5.5-5.9,and above 6.0 cm,the risk of AAE within one-year follow-up was 0.0%,1.01%,1.60%,12.50%,15.38%,and 28.57%,respectively(p<0,001).The cumulative risk was 4.08%,3.76%,5.67%,19.23%,37.14%,and 45.95%,respectively(p<0.001).The average yearly risk was 0.75%,0.75%,1.10%,4.05%,5.74%,11.19%,respectively(p<0.001).Kaplan-Meier analysis demonstrated a marked decrease in survival rate when aortic size exceeded 5 cm(p<0.001).Competing risk regression identified aortic size and sex as the independent predictors for AAE.Female sex was more likely to develop AAE(subdistribution hazard ratio:2.04;95%confidence interval:1.26-3.22).A nomogram model incorporating size and sex was built.Calibration plot showed that the predicted and observed AAE-free survival closely approximates the ideal predictions.ROC analysis showed a good predicative accuracy,with an Area Under the Curve(AUC)of 0.8283.2.5 ConclusionsThe study supports the following conclusions and recommendations:1.Size matters:Clinical prediction based on aortic size and patient gender accurately portends adverse aortic events(AAEs)with a high degree of accuracy.A simple predictive nomogram is provided.2.The "hinge point" for AAE is seen earlier(at a smaller size)than in earlier studies.So,prophylactic intervention at an aortic dimension of 5 cm is recommended at experienced centers.3.Sex matters:Ascending aortic aneurysms fare worse in females. |
PDF Full Text Request