Tętniak aorty – Rokowania, prognozy i postęp choroby

Tętniak aorty
Rokowania, prognozy i postęp choroby

Tętniak aorty, szczególnie aorty brzusznej (AAA), stanowi istotne wyzwanie kliniczne ze względu na ryzyko pęknięcia, które jest 13. przyczyną zgonów w USA. Wzrost tętniaka jest zwykle progresywny, co wymaga wczesnej oceny specjalistycznej i decyzji o leczeniu lub monitoringu. Nowoczesne narzędzia oparte na sztucznej inteligencji (AI) i uczeniu maszynowym (ML) wykazują wysoką skuteczność w przewidywaniu powikłań po otwartej naprawie AAA, z wartościami AUROC od 0,81 do 0,91 oraz niskim wynikiem Briera 0,03. Modele te potrafią prognozować m.in. zawał mięśnia sercowego, udar, zgon, reinterwencje i nieplanowane readmisje w ciągu 30 dni po zabiegu. Ponadto, klasyfikator prognostyczny tętniaka (APC) oparty na ML umożliwia stratyfikację pacjentów na grupy stabilne, wymagające naprawy lub z pękniętym AAA, co podkreśla znaczenie integracji danych klinicznych z ilościową oceną biomechaniczną i morfologiczną w obrazowaniu.

Prognoza w tętniaku aorty (Aortic aneurysm Prognosis)

Narzędzia do prognozowania wyników leczenia tętniaka aorty
Sztuczna inteligencja w prognozowaniu tętniaka aorty
Modele sieci neuronowych w prognozowaniu śmiertelności
Modele prognostyczne dla wewnątrznaczyniowej naprawy tętniaka
Biomarkery prognostyczne w tętniaku aorty
Predykcja wzrostu tętniaka aorty brzusznej
Rokowanie po leczeniu tętniaka aorty

Znaczenie prognostyki w tętniaku aorty

Kolejne rozdziały

Prognoza w tętniaku aorty (Aortic aneurysm Prognosis)

Tętniak aorty stanowi poważny problem kliniczny, charakteryzujący się postępującym poszerzeniem światła naczynia, które bez odpowiedniego leczenia może prowadzić do jego pęknięcia – zdarzenia będącego 13. przyczyną zgonów w USA. ¹ W większości przypadków tętniaki aorty wykazują tendencję do wzrostu w ciągu życia pacjenta i bez leczenia zwykle postępują w kierunku pęknięcia. Dlatego gdy tylko zostanie wykryte znaczące poszerzenie aorty, niezbędna staje się ocena specjalisty w celu określenia czy konieczne jest działanie zapobiegawcze, czy wystarczy odpowiedni monitoring. ²

Pacjenci poddawani zabiegom naprawy tętniaka aorty brzusznej (AAA) metodą otwartą mają wysokie ryzyko powikłań pooperacyjnych. Obecnie brakuje powszechnie stosowanych, wystandaryzowanych narzędzi do przewidywania ryzyka chirurgicznego w tej populacji, co podkreśla potrzebę opracowania lepszych i bardziej praktycznych narzędzi predykcyjnych dla pacjentów rozważających otwartą naprawę AAA. ³

Narzędzia do prognozowania wyników leczenia tętniaka aorty

W ostatnich latach nastąpił znaczący postęp w opracowywaniu zaawansowanych narzędzi prognostycznych opartych na sztucznej inteligencji i uczeniu maszynowym, które mogą pomóc chirurgom naczyniowym w podejmowaniu decyzji klinicznych dotyczących pacjentów z tętniakiem aorty. Badania wykazały, że zautomatyzowane algorytmy uczenia maszynowego mogą dokładnie przewidywać poważne zdarzenia sercowo-naczyniowe (MACE) w ciągu 30 dni po otwartym zabiegu naprawy AAA z wartością AUROC 0,90. ⁴

Opracowane algorytmy potrafią również przewidywać występowanie zawału mięśnia sercowego, udaru, zgonu, reinterwencji, innych powikłań, wypisów do miejsc innych niż dom oraz nieplanowanych readmisji w ciągu 30 dni z wartościami AUROC w zakresie od 0,81 do 0,91. Modele te są dobrze skalibrowane i osiągają wynik Briera 0,03, co świadczy o ich wysokiej jakości predykcyjnej. ⁵

Sztuczna inteligencja w prognozowaniu tętniaka aorty

Pojawienie się narzędzi sztucznej inteligencji (AI), głównie algorytmów uczenia maszynowego (ML), stwarza możliwość diagnostyki i ukierunkowania postępowania klinicznego w różnych chorobach, w tym w tętniaku aorty. Takie narzędzia oparte na AI/ML do prognostyki AAA mogą mieć szczególnie duży wpływ na praktykę kliniczną. ⁶

Jednym z nowatorskich rozwiązań jest klasyfikator prognostyczny tętniaka (APC) skonstruowany na podstawie modelu ML, który może stratyfikować wyniki pacjentów na stabilne, wymagające naprawy oraz pęknięte AAA. Model ten wykazał zdolność do rozróżniania pacjentów z AAA według wyników i stanowi potencjalnie ważny krok w kierunku stworzenia niezawodnego, nieinwazyjnego, obiektywnego narzędzia wspomagającego podejmowanie decyzji klinicznych w leczeniu tętniaka. ⁷

Co istotne, badania wykazały, że same wskaźniki kliniczne są niewystarczające do stratyfikacji wyników pacjentów, a obrazowanie oparte na ilościowej ocenie biomechanicznej i morfologicznej znacząco przyczynia się do skuteczności podejść opartych na ML. ⁸

Modele sieci neuronowych w prognozowaniu śmiertelności

Pęknięty tętniak aorty brzusznej (rAAA) wiąże się z wysoką śmiertelnością, nawet przy szybkim transporcie do ośrodka medycznego. Badania wykazały, że modelowanie sztucznych sieci neuronowych (ANN) może stanowić skuteczne narzędzie do oceny prawdopodobieństwa śmiertelności wewnątrzszpitalnej przy przyjęciu z powodu rAAA, wykorzystując łatwo dostępne informacje o pacjencie. ⁹

W badaniach zidentyfikowano pięć przedoperacyjnych czynników będących istotnymi niezależnymi predyktorami śmiertelności wewnątrzszpitalnej w analizie wieloczynnikowej: zaawansowany wiek, choroby nerek, utrata przytomności, zatrzymanie krążenia i wstrząs. Sekwencyjne gromadzenie od zera do czterech z tych czynników ryzyka stopniowo zwiększało ogólny wskaźnik śmiertelności, od 11% do 16% do 44% do 76% do 89% (wiek ≥70 uznany za czynnik ryzyka). ¹⁰

Model predykcyjny oparty na ANN okazał się najbardziej różnicującym spośród porównywanych modeli, przewyższając model regresji logistycznej i ustalony wynik GAS. Sieć neuronowa osiągnęła najwyższą wartość AUC i wartość kwadratu r² Pearsona, z najlepszą dokładnością (98%), czułością (94%), swoistością (100%), wartością predykcyjną dodatnią (100%) i wartością predykcyjną ujemną (97%). ¹¹

Modele prognostyczne dla wewnątrznaczyniowej naprawy tętniaka

Wewnątrznaczyniowa naprawa tętniaka (EVAR) ma wyraźną krótkoterminową przewagę nad otwartą naprawą chirurgiczną w leczeniu tętniaków aorty brzusznej, jednak ta korzyść zanika w perspektywie długoterminowej. Obecny trend w kierunku medycyny stratyfikowanej doprowadził do powstania różnorodnych modeli prognostycznych i systemów punktacji dla EVAR. Modele te mogą działać jako narzędzia wspomagające podejmowanie decyzji, wykorzystujące czynniki związane z pacjentem i operacją w celu poprawy długoterminowych wyników. ¹²

Ocenę takich modeli przeprowadza się z wykorzystaniem wytycznych do krytycznej oceny i ekstrakcji danych dla systematycznych przeglądów badań modelowania predykcyjnego oraz wytycznych PRISMA dla przeglądów systematycznych. Do krytycznej oceny włączonych badań stosuje się narzędzie oceny ryzyka błędu modelu predykcyjnego (Prediction model Risk of Bias Assessment Tool) oraz listę kontrolną przejrzystego raportowania wielozmiennego modelu predykcyjnego do indywidualnej prognozy lub diagnozy (TRIPOD). ¹³

Biomarkery prognostyczne w tętniaku aorty

Tętniak aorty brzusznej (AAA) to złożona choroba angażująca różne szlaki i procesy biologiczne, co sprawia, że zrozumienie progresji choroby może być trudne. Znalezienie predykcyjnych biomarkerów prognozy AAA w momencie diagnozy może być korzystne dla stratyfikacji ryzyka pacjentów, ściślejszego monitorowania osób z większym ryzykiem powikłań i zapewnienia im mniej inwazyjnych strategii chirurgicznych, jeśli są wymagane. ¹⁴

Badania zidentyfikowały łącznie 45 białek jako potencjalne biomarkery prognostyczne dla AAA, przewidujące występowanie AAA, pęknięcie AAA, wzrost AAA, przeciek i śmiertelność po zabiegu chirurgicznym. ¹⁵ Identyfikacja biomarkerów prognostycznych dla AAA pozwala lepiej zrozumieć mechanizmy inicjujące tę chorobę oraz stanowi podstawę dla przyszłych badań, które mogą umożliwić dalszą walidację tych markerów do wykorzystania w warunkach klinicznych. ¹⁶

Kluczowym elementem jest możliwość zintegrowania historii medycznej z tymi markerami, aby lepiej dostosować modele do pacjenta. Obecnie brakuje jednak badań, które badałyby zdolność wyłącznie biomarkerów do przewidywania wyników. Przyszłe badania mogą zbadać nowe markery lub wykorzystanie kombinacji tych markerów w celu dalszego zwiększenia dokładności przewidywania niekorzystnych zdarzeń sercowo-naczyniowych w tej populacji pacjentów. ¹⁷

Predykcja wzrostu tętniaka aorty brzusznej

Opracowanie nowych metod przewidywania wzrostu AAA jest uznawane za priorytet badawczy. Dokładne przewidywanie wzrostu AAA u pacjentów może umożliwić optymalizację odstępów między badaniami kontrolnymi i lepiej informować o czasie operacji. ¹⁸

Badania wykazały, że cechy geometryczne AAA mogą przewidywać jego przyszły wzrost. Ta metoda może być stosowana do rutynowych klinicznych skanów CT uzyskiwanych od pacjentów podczas ich ścieżki nadzoru AAA. ¹⁹

Wykorzystując średnicę przednio-tylną (APD), wskaźnik nierówności (UI) i promień krzywizny (RC) jako 3 zmienne wejściowe, pole pod krzywą charakterystyki odbiornika (ROC) dla przewidywania powolnego wzrostu (≤2,5 mm/rok) lub szybkiego wzrostu (≥5 mm/rok) w ciągu 12 miesięcy wynosi odpowiednio 0,80 i 0,79. ²⁰ Przewidywanie tempa wzrostu mieści się w granicy błędu 2 mm w 87% przypadków. ²¹

Zaobserwowano istotne dodatnie korelacje między rozmiarem AAA (Spearman r = 0,25, P≤0,05) i UI (Spearman r = 0,38, P≤0,001) z rocznym tempem wzrostu AAA. Znaczącą ujemną korelację zaobserwowano między minimalnym RC a rocznym tempem wzrostu AAA (Spearman r = -0,53, P≤0,001). ²²

Różne kombinacje cech wejściowych (APD, UI i RC) były używane do trenowania wielu modeli regresji logistycznej. Model składający się z 3 zmiennych znacznie przewyższa stosowanie samej średnicy AAA jako predyktora (P≤0,01). Przewidywania z tego modelu były znacząco skorelowane (r = 0,61, P≤0,001) i bliższe (RMSE: 1,32±1,44 mm) obserwowanym pomiarom niż przewidywania z innych modeli. ²³

Rokowanie po leczeniu tętniaka aorty

Po naprawie tętniaka pacjenci zazwyczaj pozostają stabilni przez resztę życia i rzadko wymagają kolejnej operacji na tym samym odcinku aorty. ²⁴

Rozwarstwienie aorty jest poważnym powikłaniem, ale bezpośrednie rokowanie zależy od miejsca jego wystąpienia. Rozwarstwienia dotyczące aorty wstępującej lub łuku aorty zwykle wiążą się z natychmiastowym ryzykiem zgonu, ponieważ mogą zagrażać ważnym strukturom anatomicznym w okolicy (tętnice wieńcowe, tętnice szyjne, zastawka aorty), a także powodować ryzyko poszerzenia i pęknięcia. ²⁵

Rozwarstwienia zlokalizowane w zstępującej aorcie piersiowej (poza lewą tętnicą podobojczykową) mają z kolei tendencję do lepszego rokowania z znacznie niższym ryzykiem powikłań zagrażających życiu, ale ich prawidłowe leczenie nadal wymaga intensywnego postępowania medycznego. ²⁶

Znaczenie prognostyki w tętniaku aorty

Postęp w dziedzinie sztucznej inteligencji i uczenia maszynowego otwiera nowe możliwości w prognozowaniu wyników leczenia tętniaka aorty. Opracowane modele prognostyczne mogą pomóc chirurgom naczyniowym w podejmowaniu bardziej świadomych decyzji dotyczących leczenia pacjentów z tętniakiem aorty, co może prowadzić do poprawy wyników i zmniejszenia kosztów związanych z powikłaniami, reinterwencjami i ponownymi przyjęciami do szpitala. ²⁷²⁸

Dalsze badania w dziedzinie biomarkerów i modelowania matematycznego mogą przyczynić się do opracowania jeszcze dokładniejszych narzędzi prognostycznych, które umożliwią personalizację leczenia tętniaka aorty w zależności od indywidualnych cech pacjenta i charakterystyki choroby. ²⁹³⁰

Kolejne rozdziały

Zapraszamy do dalszego czytania naszego leksykonu.

Wybierz kolejny rozdział z menu poniżej, aby otworzyć nową podstronę kompedium wiedzy i uzyskać szczegółowe informację o leku, substancji lub chorobie.

Materiały źródłowe

#1 An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes | Scientific Reports
https://www.nature.com/articles/s41598-024-53459-5
Abdominal aortic aneurysms (AAA) have been rigorously investigated to understand when their clinically-estimated risk of rupturean event that is the 13th leading cause of death in the USexceeds the risk associated with repair. […] In this study, we train and assess machine learning models using clinical, biomechanical, and morphological indices from 381 patients to develop an aneurysm prognosis classifier to predict one of three outcomes for a given AAA patient: their AAA will remain stable, their AAA will require repair based as currently indicated from the maximum diameter criterion, or their AAA will rupture. […] The APC model therefore represents a potential clinical tool to striate specific patient outcomes using machine learning models and patient-specific image-based (biomechanical and morphological) and clinical data as input. Such a tool could greatly assist clinicians in their management decisions for patients with AAA.
#2 Prognosis for aortic aneurysm
https://www.clinicbarcelona.org/en/assistance/diseases/aortic-aneurysm/evolution-of-the-disease
In most cases, aortic aneurysms tend to grow throughout the patients lifetime and, if left untreated, usually progress towards rupture. Therefore, as soon as significant dilatation of the aorta is detected, an assessment by a specialist becomes essential to determine if preventive action is necessary to avoid complications or, alternatively, whether to continue with appropriate monitoring. […] Once the aneurysm has been repaired, patients tend to remain stable for the rest of their lives and rarely require another operation on the same section of aorta. […] Aortic dissection is a serious complication, but the immediate prognosis depends on where it occurs. As such, dissections affecting the ascending aorta or aortic arch usually entail an immediate risk of death because they may compromise important anatomical structures in the area (coronary arteries, carotid arteries, aortic valve), as well as the risk of dilatation and rupture. Dissections located in the descending thoracic aorta (beyond the left subclavian artery), on the other hand, tend to have a better prognosis with a considerably lower risk of life-threatening complications, but their correct treatment still requires intensive medical management.
#3 Predicting outcomes following open abdominal aortic aneurysm repair using machine learning | Scientific Reports
https://www.nature.com/articles/s41598-025-98573-0
Patients undergoing open surgical repair of abdominal aortic aneurysm (AAA) have a high risk of post-operative complications. […] However, there are no widely used tools to predict surgical risk in this population. […] Currently, there are no standardized tools to predict adverse outcomes following open AAA repair. […] Therefore, there is an important need to develop better and more practical risk prediction tools for patients being considered for open AAA repair. […] Our automated ML algorithm can help guide risk-mitigation strategies for patients being considered for open AAA repair to improve outcomes. […] In this study, we used data from the ACS NSQIP targeted AAA files between 2011 and 2021 consisting of 3,620 patients who underwent open AAA repair to develop ML models that accurately predict 30-day MACE with an AUROC of 0.90.
#4 Predicting outcomes following open abdominal aortic aneurysm repair using machine learning | Scientific Reports
https://www.nature.com/articles/s41598-025-98573-0
Patients undergoing open surgical repair of abdominal aortic aneurysm (AAA) have a high risk of post-operative complications. […] However, there are no widely used tools to predict surgical risk in this population. […] Currently, there are no standardized tools to predict adverse outcomes following open AAA repair. […] Therefore, there is an important need to develop better and more practical risk prediction tools for patients being considered for open AAA repair. […] Our automated ML algorithm can help guide risk-mitigation strategies for patients being considered for open AAA repair to improve outcomes. […] In this study, we used data from the ACS NSQIP targeted AAA files between 2011 and 2021 consisting of 3,620 patients who underwent open AAA repair to develop ML models that accurately predict 30-day MACE with an AUROC of 0.90.
#5 Predicting outcomes following open abdominal aortic aneurysm repair using machine learning | Scientific Reports
https://www.nature.com/articles/s41598-025-98573-0
Our algorithms also predicted 30-day MI, stroke, death, re-intervention, other morbidity, non-home discharge, and unplanned readmission with AUROCs ranging from 0.81 to 0.91. […] Our model was well calibrated and achieved a Brier score of 0.03. […] Overall, we have developed robust ML-based prognostic models with excellent predictive ability for perioperative outcomes following open AAA repair, which may help guide clinical decision-making to improve outcomes and reduce costs from complications, reinterventions, and readmissions.
#6 An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes | Scientific Reports
https://www.nature.com/articles/s41598-024-53459-5
These considerations highlight the deficiencies of the more than 60-year-old one-size-fits-all maximum diameter criterion to reliably assess AAA patient prognosis. […] The advent of artificial intelligence (AI) tools, mainly machine learning (ML) algorithms, provide the possibility of diagnosis and guidance on clinical management for various diseases. […] Therefore, such AI/ML-based tools for AAA prognostics in particular could have a very big impact. […] In this study, a novel aneurysm prognosis classifier (APC) was constructed based on a ML model that was to striate patient outcomes for stable, repair, and ruptured AAA. […] The models were then trained using all three sets of indices along with known patient outcomes (stable, repair, and rupture). […] The APC model demonstrated the ability to striate AAA patients according to outcomes and represents a potentially important step towards the creation of a reliable, noninvasive, objective clinical decision support tool for aneurysm management. […] Throughout training of the hierarchical levels targeting every combination of categories, it was found that clinical indices alone are insufficient to striate patient outcomes and that imaging-based biomechanical and morphological quantification contributes significantly to ML approaches.
#7 An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes | Scientific Reports
https://www.nature.com/articles/s41598-024-53459-5
These considerations highlight the deficiencies of the more than 60-year-old one-size-fits-all maximum diameter criterion to reliably assess AAA patient prognosis. […] The advent of artificial intelligence (AI) tools, mainly machine learning (ML) algorithms, provide the possibility of diagnosis and guidance on clinical management for various diseases. […] Therefore, such AI/ML-based tools for AAA prognostics in particular could have a very big impact. […] In this study, a novel aneurysm prognosis classifier (APC) was constructed based on a ML model that was to striate patient outcomes for stable, repair, and ruptured AAA. […] The models were then trained using all three sets of indices along with known patient outcomes (stable, repair, and rupture). […] The APC model demonstrated the ability to striate AAA patients according to outcomes and represents a potentially important step towards the creation of a reliable, noninvasive, objective clinical decision support tool for aneurysm management. […] Throughout training of the hierarchical levels targeting every combination of categories, it was found that clinical indices alone are insufficient to striate patient outcomes and that imaging-based biomechanical and morphological quantification contributes significantly to ML approaches.
#8 An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes | Scientific Reports
https://www.nature.com/articles/s41598-024-53459-5
These considerations highlight the deficiencies of the more than 60-year-old one-size-fits-all maximum diameter criterion to reliably assess AAA patient prognosis. […] The advent of artificial intelligence (AI) tools, mainly machine learning (ML) algorithms, provide the possibility of diagnosis and guidance on clinical management for various diseases. […] Therefore, such AI/ML-based tools for AAA prognostics in particular could have a very big impact. […] In this study, a novel aneurysm prognosis classifier (APC) was constructed based on a ML model that was to striate patient outcomes for stable, repair, and ruptured AAA. […] The models were then trained using all three sets of indices along with known patient outcomes (stable, repair, and rupture). […] The APC model demonstrated the ability to striate AAA patients according to outcomes and represents a potentially important step towards the creation of a reliable, noninvasive, objective clinical decision support tool for aneurysm management. […] Throughout training of the hierarchical levels targeting every combination of categories, it was found that clinical indices alone are insufficient to striate patient outcomes and that imaging-based biomechanical and morphological quantification contributes significantly to ML approaches.
#9 Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network
https://pmc.ncbi.nlm.nih.gov/articles/PMC4484301/
Ruptured abdominal aortic aneurysm (rAAA) carries a high mortality rate, even with prompt transfer to a medical center. The goal of this study was to effectively use ANN modeling to provide vascular surgeons a discriminant adjunct to assess the likelihood of in-hospital mortality on a pending rAAA admission using easily obtainable patient information from the field. Of the 125 patients, 53 (42%) did not survive to discharge. Five preoperative factors were significant (P .05) independent predictors of in-hospital mortality in multivariate analysis: advanced age, renal disease, loss of consciousness, cardiac arrest and shock, though renal disease was excluded from the models. The sequential accumulation of zero to four of these risk factors progressively increased overall mortality rate, from 11% to 16% to 44% to 76% to 89% (Age 70 considered a risk factor). An ANN-based predictive model may represent a simple, useful and highly discriminant adjunct to the vascular surgeon in accurately identifying those patients who may carry a high mortality risk from attempted repair of rAAA, using only easily definable preoperative variables. The ANN model represented the most discriminant of the three. The data from our ANN allow prediction of in-hospital mortality for surgical rAAA patients with greater discriminant ability than a multiple logistic regression model or the established GAS score. The ANN had the highest AUC and Pearson r2 square value of all models, with the best accuracy (98%), sensitivity (94%), specificity (100%), positive predictive value (100%) and negative predictive value (97%). This allowed the development of an internally validated, user-friendly aid for risk assessment. As this ANN model reliably predicted the presence of in-hospital mortality, it may function as an evidence-based adjunct to the vascular surgeons clinical judgment in estimation of mortality of a rAAA patient, on the basis of easily obtained preoperative factors.
#10 Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network
https://pmc.ncbi.nlm.nih.gov/articles/PMC4484301/
Ruptured abdominal aortic aneurysm (rAAA) carries a high mortality rate, even with prompt transfer to a medical center. The goal of this study was to effectively use ANN modeling to provide vascular surgeons a discriminant adjunct to assess the likelihood of in-hospital mortality on a pending rAAA admission using easily obtainable patient information from the field. Of the 125 patients, 53 (42%) did not survive to discharge. Five preoperative factors were significant (P .05) independent predictors of in-hospital mortality in multivariate analysis: advanced age, renal disease, loss of consciousness, cardiac arrest and shock, though renal disease was excluded from the models. The sequential accumulation of zero to four of these risk factors progressively increased overall mortality rate, from 11% to 16% to 44% to 76% to 89% (Age 70 considered a risk factor). An ANN-based predictive model may represent a simple, useful and highly discriminant adjunct to the vascular surgeon in accurately identifying those patients who may carry a high mortality risk from attempted repair of rAAA, using only easily definable preoperative variables. The ANN model represented the most discriminant of the three. The data from our ANN allow prediction of in-hospital mortality for surgical rAAA patients with greater discriminant ability than a multiple logistic regression model or the established GAS score. The ANN had the highest AUC and Pearson r2 square value of all models, with the best accuracy (98%), sensitivity (94%), specificity (100%), positive predictive value (100%) and negative predictive value (97%). This allowed the development of an internally validated, user-friendly aid for risk assessment. As this ANN model reliably predicted the presence of in-hospital mortality, it may function as an evidence-based adjunct to the vascular surgeons clinical judgment in estimation of mortality of a rAAA patient, on the basis of easily obtained preoperative factors.
#11 Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network
https://pmc.ncbi.nlm.nih.gov/articles/PMC4484301/
Ruptured abdominal aortic aneurysm (rAAA) carries a high mortality rate, even with prompt transfer to a medical center. The goal of this study was to effectively use ANN modeling to provide vascular surgeons a discriminant adjunct to assess the likelihood of in-hospital mortality on a pending rAAA admission using easily obtainable patient information from the field. Of the 125 patients, 53 (42%) did not survive to discharge. Five preoperative factors were significant (P .05) independent predictors of in-hospital mortality in multivariate analysis: advanced age, renal disease, loss of consciousness, cardiac arrest and shock, though renal disease was excluded from the models. The sequential accumulation of zero to four of these risk factors progressively increased overall mortality rate, from 11% to 16% to 44% to 76% to 89% (Age 70 considered a risk factor). An ANN-based predictive model may represent a simple, useful and highly discriminant adjunct to the vascular surgeon in accurately identifying those patients who may carry a high mortality risk from attempted repair of rAAA, using only easily definable preoperative variables. The ANN model represented the most discriminant of the three. The data from our ANN allow prediction of in-hospital mortality for surgical rAAA patients with greater discriminant ability than a multiple logistic regression model or the established GAS score. The ANN had the highest AUC and Pearson r2 square value of all models, with the best accuracy (98%), sensitivity (94%), specificity (100%), positive predictive value (100%) and negative predictive value (97%). This allowed the development of an internally validated, user-friendly aid for risk assessment. As this ANN model reliably predicted the presence of in-hospital mortality, it may function as an evidence-based adjunct to the vascular surgeons clinical judgment in estimation of mortality of a rAAA patient, on the basis of easily obtained preoperative factors.
#12 Prognostic prediction models for endovascular abdominal aortic aneurysm repair: protocol for a scoping review – PubMed
https://pubmed.ncbi.nlm.nih.gov/36307155/
Endovascular aneurysm repair (EVAR) has a marked short-term advantage over open surgical repair in managing abdominal aortic aneurysms (AAA); however, this benefit is lost in the long term. The current trend towards stratified medicine has given rise to diverse prognostic prediction models and scoring systems for EVAR. These models could act as decision support tools that employ patient and operative factors, to improve long-term outcomes. Past literature evaluated and compared model performance for predicting one outcome, for example, mortality. None were deemed competent for clinical application. The proposed study will use a scoping review approach to capture literature on prognostic modelling in EVAR for all predictable outcomes. The results are anticipated to inform future research, identify knowledge gaps, and assist in determining the potential of models for clinical use.
#13 Prognostic prediction models for endovascular abdominal aortic aneurysm repair: protocol for a scoping review – PubMed
https://pubmed.ncbi.nlm.nih.gov/36307155/
Data will be abstracted using a charting form based on the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies guidelines and PRISMA guidelines for systematic reviews. The Prediction model Risk of Bias Assessment Tool and the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis checklist will be used to critically appraise included studies.
#14 Current Prognostic Biomarkers for Abdominal Aortic Aneurysm: A Comprehensive Scoping Review of the Literature
https://www.mdpi.com/2218-273X/14/6/661
Abdominal aortic aneurysm (AAA) is a progressive dilatation of the aorta that can lead to aortic rupture. […] A total of 45 proteins were found to be potential prognostic biomarkers for AAA, predicting incidence of AAA, AAA rupture, AAA growth, endoleak, and post-surgical mortality. […] This review outlines the wide pathophysiological processes that are implicated in AAA disease progression. […] Abdominal aortic aneurysm (AAA) is a complex disease that involves various biological pathways and processes, and understanding the progression of the disease can be difficult. […] Therefore, finding predictive biomarkers for the prognosis of AAA at the time of diagnosis may be beneficial to risk stratify patients and monitor more closely those who are at a greater risk of complications and provide them with less invasive surgical strategies if required.
#15 Current Prognostic Biomarkers for Abdominal Aortic Aneurysm: A Comprehensive Scoping Review of the Literature
https://www.mdpi.com/2218-273X/14/6/661
Abdominal aortic aneurysm (AAA) is a progressive dilatation of the aorta that can lead to aortic rupture. […] A total of 45 proteins were found to be potential prognostic biomarkers for AAA, predicting incidence of AAA, AAA rupture, AAA growth, endoleak, and post-surgical mortality. […] This review outlines the wide pathophysiological processes that are implicated in AAA disease progression. […] Abdominal aortic aneurysm (AAA) is a complex disease that involves various biological pathways and processes, and understanding the progression of the disease can be difficult. […] Therefore, finding predictive biomarkers for the prognosis of AAA at the time of diagnosis may be beneficial to risk stratify patients and monitor more closely those who are at a greater risk of complications and provide them with less invasive surgical strategies if required.
#16 Current Prognostic Biomarkers for Abdominal Aortic Aneurysm: A Comprehensive Scoping Review of the Literature
https://www.mdpi.com/2218-273X/14/6/661
By identifying prognostic biomarkers for AAA, we hope to further understand the mechanisms behind the initiation of this disease, as well as set the groundwork for future research which may allow for further validation of these markers to be used in a clinical setting. […] This review has outlined some of the single proteins and their ability to predict outcomes. […] The key element is the ability to integrate past medical history in combination with these markers to better tailor models to the patient. […] There is currently, however, a lack of studies that investigate the ability of biomarkers only to predict outcomes. […] This review has outlined the currently available data on prognostic markers for AAA. Future studies may investigate new markers or using a combination of these markers to further increase the accuracy of predicting adverse cardiovascular events in this patient population.
#17 Current Prognostic Biomarkers for Abdominal Aortic Aneurysm: A Comprehensive Scoping Review of the Literature
https://www.mdpi.com/2218-273X/14/6/661
By identifying prognostic biomarkers for AAA, we hope to further understand the mechanisms behind the initiation of this disease, as well as set the groundwork for future research which may allow for further validation of these markers to be used in a clinical setting. […] This review has outlined some of the single proteins and their ability to predict outcomes. […] The key element is the ability to integrate past medical history in combination with these markers to better tailor models to the patient. […] There is currently, however, a lack of studies that investigate the ability of biomarkers only to predict outcomes. […] This review has outlined the currently available data on prognostic markers for AAA. Future studies may investigate new markers or using a combination of these markers to further increase the accuracy of predicting adverse cardiovascular events in this patient population.
#18
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
We investigated the utility of geometric features for future AAA growth prediction. […] Novel methods for growth prediction of AAA are recognized as a research priority. […] Geometric features of an AAA can predict its future growth. This method can be applied to routine clinical CT scans acquired from patients during their AAA surveillance pathway. […] Methods for the prediction of AAA growth is considered as a priority for research in the opinions of vascular and endovascular surgeons. […] Accurate prediction of AAA growth in patients can allow for the optimization of surveillance intervals and better inform the timing for surgery. […] Using APD, UI, and RC as 3 input variables, the area under receiver operating characteristics curve for predicting slow growth (2.5 mm/yr) or fast growth (5 mm/yr) at 12 months are 0.80 and 0.79, respectively.
#19
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
We investigated the utility of geometric features for future AAA growth prediction. […] Novel methods for growth prediction of AAA are recognized as a research priority. […] Geometric features of an AAA can predict its future growth. This method can be applied to routine clinical CT scans acquired from patients during their AAA surveillance pathway. […] Methods for the prediction of AAA growth is considered as a priority for research in the opinions of vascular and endovascular surgeons. […] Accurate prediction of AAA growth in patients can allow for the optimization of surveillance intervals and better inform the timing for surgery. […] Using APD, UI, and RC as 3 input variables, the area under receiver operating characteristics curve for predicting slow growth (2.5 mm/yr) or fast growth (5 mm/yr) at 12 months are 0.80 and 0.79, respectively.
#20
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
We investigated the utility of geometric features for future AAA growth prediction. […] Novel methods for growth prediction of AAA are recognized as a research priority. […] Geometric features of an AAA can predict its future growth. This method can be applied to routine clinical CT scans acquired from patients during their AAA surveillance pathway. […] Methods for the prediction of AAA growth is considered as a priority for research in the opinions of vascular and endovascular surgeons. […] Accurate prediction of AAA growth in patients can allow for the optimization of surveillance intervals and better inform the timing for surgery. […] Using APD, UI, and RC as 3 input variables, the area under receiver operating characteristics curve for predicting slow growth (2.5 mm/yr) or fast growth (5 mm/yr) at 12 months are 0.80 and 0.79, respectively.
#21
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
The prediction or growth rate is within 2 mm error in 87% of cases. […] There were significant positive correlations between AAA size (Spearman r = 0.25, P 0.05) and UI (Spearman r = 0.38, P 0.001) with annual AAA growth rate. […] A significant negative correlation between minimum RC and annual AAA growth rate was observed. (Spearman r=-0.53, P 0.001). […] Different combinations of input features (APD, UI, and RC) were used to train multiple logistic regression models. […] The area under receiver operation curve for predicting slow growth (2.5 mm/yr) and prediction fast growth (5mm/yr) is 0.80 and 0.79, respectively. […] This model comprising of 3 variables significantly outperforms the use of AAA diameter alone as the predictor (P 0.01). […] Predictions from this model were significantly correlated (r = 0.61, P 0.001) and closer (RMSE: 1.32 1.44 mm) to that of observed measurements than that of the other models. […] This method can be applied to historic scans acquired during the routine clinical pathways of each AAA patient.
#22
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
The prediction or growth rate is within 2 mm error in 87% of cases. […] There were significant positive correlations between AAA size (Spearman r = 0.25, P 0.05) and UI (Spearman r = 0.38, P 0.001) with annual AAA growth rate. […] A significant negative correlation between minimum RC and annual AAA growth rate was observed. (Spearman r=-0.53, P 0.001). […] Different combinations of input features (APD, UI, and RC) were used to train multiple logistic regression models. […] The area under receiver operation curve for predicting slow growth (2.5 mm/yr) and prediction fast growth (5mm/yr) is 0.80 and 0.79, respectively. […] This model comprising of 3 variables significantly outperforms the use of AAA diameter alone as the predictor (P 0.01). […] Predictions from this model were significantly correlated (r = 0.61, P 0.001) and closer (RMSE: 1.32 1.44 mm) to that of observed measurements than that of the other models. […] This method can be applied to historic scans acquired during the routine clinical pathways of each AAA patient.
#23
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
The prediction or growth rate is within 2 mm error in 87% of cases. […] There were significant positive correlations between AAA size (Spearman r = 0.25, P 0.05) and UI (Spearman r = 0.38, P 0.001) with annual AAA growth rate. […] A significant negative correlation between minimum RC and annual AAA growth rate was observed. (Spearman r=-0.53, P 0.001). […] Different combinations of input features (APD, UI, and RC) were used to train multiple logistic regression models. […] The area under receiver operation curve for predicting slow growth (2.5 mm/yr) and prediction fast growth (5mm/yr) is 0.80 and 0.79, respectively. […] This model comprising of 3 variables significantly outperforms the use of AAA diameter alone as the predictor (P 0.01). […] Predictions from this model were significantly correlated (r = 0.61, P 0.001) and closer (RMSE: 1.32 1.44 mm) to that of observed measurements than that of the other models. […] This method can be applied to historic scans acquired during the routine clinical pathways of each AAA patient.
#24 Prognosis for aortic aneurysm
https://www.clinicbarcelona.org/en/assistance/diseases/aortic-aneurysm/evolution-of-the-disease
In most cases, aortic aneurysms tend to grow throughout the patients lifetime and, if left untreated, usually progress towards rupture. Therefore, as soon as significant dilatation of the aorta is detected, an assessment by a specialist becomes essential to determine if preventive action is necessary to avoid complications or, alternatively, whether to continue with appropriate monitoring. […] Once the aneurysm has been repaired, patients tend to remain stable for the rest of their lives and rarely require another operation on the same section of aorta. […] Aortic dissection is a serious complication, but the immediate prognosis depends on where it occurs. As such, dissections affecting the ascending aorta or aortic arch usually entail an immediate risk of death because they may compromise important anatomical structures in the area (coronary arteries, carotid arteries, aortic valve), as well as the risk of dilatation and rupture. Dissections located in the descending thoracic aorta (beyond the left subclavian artery), on the other hand, tend to have a better prognosis with a considerably lower risk of life-threatening complications, but their correct treatment still requires intensive medical management.
#25 Prognosis for aortic aneurysm
https://www.clinicbarcelona.org/en/assistance/diseases/aortic-aneurysm/evolution-of-the-disease
In most cases, aortic aneurysms tend to grow throughout the patients lifetime and, if left untreated, usually progress towards rupture. Therefore, as soon as significant dilatation of the aorta is detected, an assessment by a specialist becomes essential to determine if preventive action is necessary to avoid complications or, alternatively, whether to continue with appropriate monitoring. […] Once the aneurysm has been repaired, patients tend to remain stable for the rest of their lives and rarely require another operation on the same section of aorta. […] Aortic dissection is a serious complication, but the immediate prognosis depends on where it occurs. As such, dissections affecting the ascending aorta or aortic arch usually entail an immediate risk of death because they may compromise important anatomical structures in the area (coronary arteries, carotid arteries, aortic valve), as well as the risk of dilatation and rupture. Dissections located in the descending thoracic aorta (beyond the left subclavian artery), on the other hand, tend to have a better prognosis with a considerably lower risk of life-threatening complications, but their correct treatment still requires intensive medical management.
#26 Prognosis for aortic aneurysm
https://www.clinicbarcelona.org/en/assistance/diseases/aortic-aneurysm/evolution-of-the-disease
In most cases, aortic aneurysms tend to grow throughout the patients lifetime and, if left untreated, usually progress towards rupture. Therefore, as soon as significant dilatation of the aorta is detected, an assessment by a specialist becomes essential to determine if preventive action is necessary to avoid complications or, alternatively, whether to continue with appropriate monitoring. […] Once the aneurysm has been repaired, patients tend to remain stable for the rest of their lives and rarely require another operation on the same section of aorta. […] Aortic dissection is a serious complication, but the immediate prognosis depends on where it occurs. As such, dissections affecting the ascending aorta or aortic arch usually entail an immediate risk of death because they may compromise important anatomical structures in the area (coronary arteries, carotid arteries, aortic valve), as well as the risk of dilatation and rupture. Dissections located in the descending thoracic aorta (beyond the left subclavian artery), on the other hand, tend to have a better prognosis with a considerably lower risk of life-threatening complications, but their correct treatment still requires intensive medical management.
#27 Predicting outcomes following open abdominal aortic aneurysm repair using machine learning | Scientific Reports
https://www.nature.com/articles/s41598-025-98573-0
Our algorithms also predicted 30-day MI, stroke, death, re-intervention, other morbidity, non-home discharge, and unplanned readmission with AUROCs ranging from 0.81 to 0.91. […] Our model was well calibrated and achieved a Brier score of 0.03. […] Overall, we have developed robust ML-based prognostic models with excellent predictive ability for perioperative outcomes following open AAA repair, which may help guide clinical decision-making to improve outcomes and reduce costs from complications, reinterventions, and readmissions.
#28 An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes | Scientific Reports
https://www.nature.com/articles/s41598-024-53459-5
These considerations highlight the deficiencies of the more than 60-year-old one-size-fits-all maximum diameter criterion to reliably assess AAA patient prognosis. […] The advent of artificial intelligence (AI) tools, mainly machine learning (ML) algorithms, provide the possibility of diagnosis and guidance on clinical management for various diseases. […] Therefore, such AI/ML-based tools for AAA prognostics in particular could have a very big impact. […] In this study, a novel aneurysm prognosis classifier (APC) was constructed based on a ML model that was to striate patient outcomes for stable, repair, and ruptured AAA. […] The models were then trained using all three sets of indices along with known patient outcomes (stable, repair, and rupture). […] The APC model demonstrated the ability to striate AAA patients according to outcomes and represents a potentially important step towards the creation of a reliable, noninvasive, objective clinical decision support tool for aneurysm management. […] Throughout training of the hierarchical levels targeting every combination of categories, it was found that clinical indices alone are insufficient to striate patient outcomes and that imaging-based biomechanical and morphological quantification contributes significantly to ML approaches.
#29 Current Prognostic Biomarkers for Abdominal Aortic Aneurysm: A Comprehensive Scoping Review of the Literature
https://www.mdpi.com/2218-273X/14/6/661
By identifying prognostic biomarkers for AAA, we hope to further understand the mechanisms behind the initiation of this disease, as well as set the groundwork for future research which may allow for further validation of these markers to be used in a clinical setting. […] This review has outlined some of the single proteins and their ability to predict outcomes. […] The key element is the ability to integrate past medical history in combination with these markers to better tailor models to the patient. […] There is currently, however, a lack of studies that investigate the ability of biomarkers only to predict outcomes. […] This review has outlined the currently available data on prognostic markers for AAA. Future studies may investigate new markers or using a combination of these markers to further increase the accuracy of predicting adverse cardiovascular events in this patient population.
#30
https://journals.lww.com/annalsofsurgery/fulltext/2023/01000/prediction_of_abdominal_aortic_aneurysm_growth.51.aspx
The prediction or growth rate is within 2 mm error in 87% of cases. […] There were significant positive correlations between AAA size (Spearman r = 0.25, P 0.05) and UI (Spearman r = 0.38, P 0.001) with annual AAA growth rate. […] A significant negative correlation between minimum RC and annual AAA growth rate was observed. (Spearman r=-0.53, P 0.001). […] Different combinations of input features (APD, UI, and RC) were used to train multiple logistic regression models. […] The area under receiver operation curve for predicting slow growth (2.5 mm/yr) and prediction fast growth (5mm/yr) is 0.80 and 0.79, respectively. […] This model comprising of 3 variables significantly outperforms the use of AAA diameter alone as the predictor (P 0.01). […] Predictions from this model were significantly correlated (r = 0.61, P 0.001) and closer (RMSE: 1.32 1.44 mm) to that of observed measurements than that of the other models. […] This method can be applied to historic scans acquired during the routine clinical pathways of each AAA patient.