BACKGROUND Ischaemic stroke accounts for 80% of all acute cerebrovascular events. Stroke is responsible for 10-12% of deaths (about 6 million) per year, making it the second leading cause of death worldwide and the first leading cause of disability in developed countries. 70% of all cerebral infarctions is secondary to middle cerebral artery occlusion. Within the hypoperfused tissue 3 regions are identified: - the infarct core: strictly hypoperfused and not more vital tissue; it progresses to necrosis even if reperfused (irreversibly damaged) - the ischaemic penumbra (around the ischemic core): hypoperfused but still vital tissue; it progressively evolves to necrosis if not reperfused (reversibly damaged) - the benign oligoemia (around the ischemic penumbra): moderately hypoperfused tissue and usually not at risk for necrosis; it recovers spontaneously also without reperfusion. Currently, the major acute phase therapies include intravenous thrombolysis and endovascular treatments. Aim of these therapies is to recanalise occluded vessels and to reperfuse the ischaemic tissue (penumbra), resulting in an improvement of neurological symptoms and prognosis. However, the main parameter to get access to reperfusion therapies is represented by the clinical onset time with a strict time window. Therefore, only a small percentage (15-20%) of people with acute ischemic stroke is actually eligible for this kind of therapies; moreover, despite recanalization, many of these patients (25-50%) do not reach a good clinical outcome; on the contrary 35% of them have a cerebral infarction growth. Therefore, it is clear that other factors (in addition to time) are involved in clinical outcome of acute ischemic stroke patients. In this context, the Neuroimaging, through the analysis of the most relevant pathophysiological aspects of acute cerebrovascular diseases, aims to contribute to the efficacy and safety of therapeutic strategies, improving and leading patients selection for reperfusion therapies. In order to meet these needs a multimodal CT protocol has been proposed. This protocol, by using non contrast CT scan (NCCT), CT angiography (CTA) and CT-perfusion (CTP), tries to provide informations about the absence of an haemorrhage (ICH), vessels patency, regional leptomeningeal collateral state (CTA) and haemodynamic consequences of vascular occlusion (CTP). This protocol is relatively fast and available in many Centers. CTP is the only able to describe the main aspects of cerebral hemodynamics. It studies the first transit of a contrast bolus through brain capillaries. Densitometric data are then analyzed according to a mathematical model called deconvolution. Through deconvolution is possible to obtain a density/time curve called Residual Function. The height of the peak of this curve quantifies the CBF (cerebral blood flow – ml blood/100 g tissue/min) and the area under the curve the CBV (cerebral blood volume - ml blood/100g tissue). MTT (mean transit time - seconds) is then calculated on the basis of the "central volume principle " (CBF = CBV / MTT). Based on deconvolution, finally we obtain the so-called parametric perfusion maps. "Classically": 1) Total hypoperfusion (core + penumbra) = extension of CBF and MTT injury 2) Core infarct size = CBV injury 3) ischaemic penumbra volume = CBF or MTT - CBV volume (CBF/MTT - CBV mismatch). AIMS OF THE STUDY Several new findings have questioned the validity of "classic" CTP approach to define cerebral haemodynamics and its reliability in order to describe ischaemic regions (core and penumbra) and to predict outcome of acute ischaemic stroke patients. Therefore, aim of this study was to verify: 1) if a visual analysis approach of “classical” CTP parameters is reliable to identify and describe cerebral hemodynamics in acute ischaemic stroke and if it could be still used in clinical practice 2) if the classical mismatch (CBF/MTT – CBV) the so-called "penumbral hypothesis" can still has a value to predict patients outcome. For this reason we decided to measure CBF, CBV and MTT absolute value and the volume of the whole ischaemic lesion, infarct core and penumbra in a broad number of patients with acute ischaemic stroke. These measures were than coupled with: -the clot burden and the status of regional leptomeningeal collaterals (onset conditions and outcome predictors at the same time) -recanalization and reperfusion (conditions related to therapy issues) -final infarct volume and disability (mRS) (long-term outcome measures). As, infarct core was measured on CBV map and ischemic penumbra by the MTT - CBV mismatch, our study aims to verify whether a qualitative approach may still has a value to predict the fate of ischaemic tissue and patients prognosis. METHODS Patient selection and study design We retrospectively evaluated 200 patients having middle cerebral artery acute stroke. All patients performed pre-treatment multimodal CT protocol including non-enhanced CT (NECT), CT Angiography (CTA) and CTP. Informed consent was obtained from the patient/next of kin and local ethics committee approved the protocol. The National Institute of Health Stroke Scale (NIHSS) was recorded at admission and at 3 months. The modified Rankin scale (mRS) was recorded at 3 months; mRS ≤ 2 and > 2 were classified as good and poor outcomes, respectively. Imaging protocol CTP studies were performed with a dynamic first-pass bolus-tracking methodology on a 64-section CT scanner (Lightspeed VCT 64, GE Helthcare) according to a one-phase imaging protocol consisting of an acquisition of 50-seconds continuous (cine) scans reconstructed at 0.5 second intervals to produce a series of 99 sequential images for each of 8 sections which covered a total of 4 cm from the basal ganglia to the lateral ventricles. Cine CTP scanning (80 kVp; 100 mAs; matrix size = 512 x 512; FOV = 25-cm; scan type = cine full 1 sec.; rotation = 0.5 sec.; total scan time 50 sec) was initiated 5 seconds after the automatic injection of 40 ml of non-ionic contrast agent (Iomeron 300 mg/ml, Bracco Imaging SpA) at the rate of 4 ml/sec. Data processing All CTP scans were assessed with a deconvolution-based delay sensitive algorithm by using an imaging workstation (Advantage Windows; GEHelthcare). CBF, CBV and MTT CTP maps were generated for each patient. CBF, CBV and MTT values were expressed in ml/100g/min, ml/100g and seconds, respectively. After identification by visual inspection on MTT and CBV maps, total ipoperfusion, infarct core and ischemic penumbra volumes were calculated by manually drawing MTT, CBV and MTT-CBV (CTP mismatch) lesions, respectively. Clot Burden Score (CBS) and regional LeptoMeningeal Collateral (rLMC) score were calculated on CTA at onset. CBS ≤ 7 and > 7 (range 0-10) and rLMC ≤ 10 and > 10 (range 0-20) were classified as poor and good, respectively. Recanalization was assessed on 24 hour CTA according to TICI criteria. Reperfusion was defined as a percentage reduction > 80% in the baseline-24 hour MTT lesion. The final infarct volume was manually outlined on follow-up 7 days unenhanced CT (NECT). Statistical analysis The normality of each variable was checked by using the Kolmogorov-Smirnov test. When normality of data distribution was found in all variables, statistical analysis was performed by a parametric approach. Conversely, when normality of data distribution was rejected in several variables, a non-parametric analysis was applied. Accordingly, continuous variables were compared by t-test, whereas their correlation were assessed by linear regression or Spearman's rank correlation coefficient, respectively. A value of p < 0.05 was accepted as statistically significant. RESULTS CTP volumetric and absolute values detected in the three different ischemic ROI analyzed, showed that the absolute values of CBF and CBV measured in core and penumbra were in agreement with those reported in previous studies. Furthermore, a CTP mismatch indicative of salvageable penumbra was detected at onset in all patients. MTT total hypoperfusion, CBV infarct core and (MTT/CBF-CBV) penumbra volumes were smaller (P<0.0001) in good than in poor CBS. Total hypoperfusion and infarct core were smaller in in good than in poor rlMC (p < 0.05 e p < 0.001, respectively). No differences were detected between MTT-CBV penumbra volumes in patients with good and poor rLMC. Total hypoperfusion, infarct core and penumbra volumes were smaller (p < 0.001, p < 0.01 e p < 0.001, respectively) recanalized than occluded patients. Penumbra volumes were smaller (p < 0.02) in reperfused than not reperfused patients. No differences were detected between total hypoperfusion and infarct core volumes in reperfused and not reperfused. Total hypoperfusion, infarct core and penumbra volumes were smaller (p < 0.0001, p < 0.0001 e p < 0.01, respectively) in good than poor outcome patients. These results were confirmed by correlations observed between CTP volumes and physiopathological, clinical and radiological parameters. DISCUSSION Our results suggest that: -absolute values measured in infarct core and penumbra were in agreement with data previously reported, indicating that the identification of these two regions performed by visual inspection was reliable. -CTP represents an useful tool to predict prognosis in acute stroke patients; the extent of MTT (total ipoperfusion) and CBV (infarct core) volumes is indeed strongly related to physiopathological, clinical and radiological outcome measures; in particular, the CBV map seems to be the most important determinant of outcome. This is in agreement with some previous studies. Thus, “classical” CTP maps (MTT, CBF and CBV) analysis by visual inspection (and particularly the CBV map) still represent a valuable method to describe cerebral haemodynamics in acute ischaemic stroke. This is most important from a daily clinical practice point of view in the aim to implement the number of treated patients and to predict their outcome.

INTRODUZIONE L’ictus ischemico rappresenta l’80% di tutti gli eventi cerebrovascolari acuti; è responsabile del 10-12% dei decessi (circa 6 milioni) per anno, rappresentando la seconda causa di morte a livello mondiale e la principale causa di disabilità nei paesi industrializzati. Il 70% di tutti gli infarti cerebrali è secondario ad occlusione dell’arteria cerebrale media. Nell’ambito del tessuto ipoperfuso si distinguono 3 regioni: - core infartuale (al centro dell’area ischemica): severamente ipoperfuso e non più vitale; evolve sempre verso l’infarto anche se viene riperfuso (danneggiato in modo irreversibile) - penombra ischemica (attorno al core): severamente ipoperfusa ma ancora vitale; evolve progressivamente verso l’infarto se non viene riperfusa (danneggiata in modo reversibile) - oligoemia benigna (attorno alla penombra ischemica): moderatamente ipoperfusa e di solito non a rischio di infarto; di solito recupera spontaneamente anche in assenza di riperfusione. Attualmente, le principali terapie di fase acuta comprendono la trombolisi endovenosa ed i trattamenti endovascolari e sono volti a ricanalizzare il vaso occluso e riperfondere il tessuto ischemico (penombra), con conseguente miglioramento dei sintomi neurologici e della prognosi. Tuttavia, il principale parametro su cui si basa la possibilità di accedere alle terapie di riperfusione è attualmente rappresentato dal tempo intercorso dall’esordio clinico con una finestra di intervento molto rigida. Pertanto, solo una ridotta percentuale (15-20%) dei pazienti con ictus ischemico acuto risulta eleggibile, per il momento, a questo tipo di terapie. E’ stato osservato inoltre, come, nonostante la ricanalizzazione, molti di questi pazienti (25-50%) non raggiungano comunque un buon outcome clinico, anzi, nel 35% di essi si abbia addirittura un’espansione dell’infarto cerebrale. E’ quindi evidente come esistano altri fattori (oltre al dato temporale) potenzialmente in grado di condizionare la prognosi dei pazienti con ictus ischemico acuto. In tale contesto, il Neuroimaging, mediante l’analisi degli aspetti fisiopatologici più rilevanti della patologia cerebrovascolare acuta, si pone come obiettivo quello di contribuire all’implementazione dell’efficacia e della sicurezza delle strategie terapeutiche, guidando e migliorando la selezione dei pazienti candidati alle terapie di riperfusione. Al fine di rispondere a queste esigenze è stato proposto pertanto un protocollo TC multimodale che, avvalendosi dell’utilizzo di TC standard senza mezzo di contrasto (NCCT), angio-TC (CTA) e TC-perfusionale (CTP), è in grado di fornire informazioni relativamente alla presenza/assenza di emorragie (ICH), alla pervietà dei vasi, allo stato della circolazione collaterale di supplenza alla regione ischemica (CTA) ed alle conseguenze emodinamiche dell’occlusione vasale (CTP). Tale protocollo è relativamente rapido e disponibile in molti centri. La CTP è l’unico strumento in grado di descrivere i principali aspetti dell’emodinamica cerebrale studiando il primo passaggio attraverso i capillari cerebrali di un bolo di mezzo di contrasto (m.d.c.) iodato somministrato per via endovenosa (e.v.). I dati densitometrici sono quindi analizzati secondo un modello matematico definito di deconvoluzione attraverso cui è possibile ricavare una curva densità/tempo definita Funzione Residua dove l’altezza del picco quantifica il CBF (flusso ematico cerebrale espresso in ml sangue/100 g di tessuto/min) e l’area sotto la curva rappresenta il CBV (volume ematico cerebrale a livello capillare-tissutale espresso in ml di sangue/100 g di tessuto). L’MTT (il tempo di transito medio capillare, ovvero il tempo intercorso tra l’immissione arteriosa nel letto capillare ed il deflusso venoso del m.d.c., espresso in secondi) è quindi calcolato sulla base del “principio del volume centrale” (CBF=CBV/MTT) dal rapporto CBV/CBF. Sulla base dei calcoli di deconvoluzione, il software genera infine le cosiddette mappe parametriche perfusionali pixel per pixel su scala colorimetrica. “Classicamente”: 1) ipoperfusione totale (core + penombra) = estensione della lesione CBF o MTT 2) core infartuale = dimensioni della lesione CBV 3) penombra ischemica = volume CBF o MTT - volume CBV (mismatch CBF o MTT – CBV). PRESUPPOSTI ED OBIETTIVI DELLO STUDIO Recenti evidenze hanno messo in discussione la validità del “classico” approccio CTP nel definire i parametri emodinamici cerebrali e quindi la sua affidabilità nel descrivere le caratteristiche delle regioni in cui notoriamente è suddivisa l’area ischemica e nel fornire indicazioni sulla prognosi dei pazienti con ictus ischemico acuto. Complessivamente pertanto, scopo di questo studio è stato quello di verificare: 1) se un approccio basato sull’analisi visiva dei classici parametri CTP sia accurato nell’identificare e nel descrivere i principali elementi dell’emodinamica cerebrale in corso di ictus ischemico acuto e pertanto ancora utilizzabile nella pratica clinica 2) se il classico mismatch tra CBF o MTTe CBV (the “penumbral hypothesis”) possa ancora avere un valore nel predire l’outcome di questi pazienti. Per questo motivo abbiamo deciso di misurare e descrivere i valori assoluti di CBF, CBV ed MTT ed il volume della lesione ischemica totale, del core infartuale e della penombra ischemica in una larga popolazione di pazienti affetti da ictus cerebrale acuto e di metterli in relazione con: - il carico trombotico e lo stato dei circoli collaterali leptomeningei di compenso (condizioni presenti all’esordio e contemporaneamente predittori di outcome), - la ricanalizzazione e la riperfusione (condizioni legate all’esito delle terapie messe in atto) - il volume dell’infarto finale e la disabilità (mRS) (indicatori di outcome a lungo termine). Dato che, il core infartuale è stato misurato sulla mappa CBV e, la penombra ischemica mediante il mismatch MTT - CBV, il nostro studio si propone di verificare se in effetti un approccio qualitativo di questo tipo può essere ancora utilizzabile ed utile nel predire il destino del tessuto ischemico e la prognosi dei pazienti. MATERIALI E METODI Selezione dei pazienti e disegno dello studio Abbiamo valutato retrospettivamente 200 pazienti con ictus ischemico nel territorio dell’arteria cerebrale media. Tutti i pazienti hanno eseguito un protocollo TC multimodale pre-trattamento comprendente: TC standard senza m.d.c. (NCCT), angio-TC (CTA) e CTP. E’ stato ottenuto il consenso informato dal paziente/familiari e il comitato etico locale ha approvato il protocollo. La scala NIHSS (National Institute of Health Stroke Scale) è stata calcolata al momento del ricovero e dopo 3 mesi. La scala di Rankin modificata (mRS) è stata calcolata a 3 mesi; un punteggio mRS ≤ 2 e > 2 è stato classificato come buono e cattivo outcome clinico, rispettivamente. Protocollo di imaging Gli studi CTP sono stati condotti secondo il modello del tracciante cinetico, analizzando il primo passaggio di m.d.c. a livello tissutale con un’apparecchiatura TC a 64 banchi (Lightspeed VCT 64, GE Helthcare) in accordo con il seguente protocollo di acquisizione: una scansione continua (cine) di 50 secondi ricostruita ad intervalli di 0.5 secondi al fine di ottenere una sequenza di 99 immagini per ciascuna delle 8 sezioni in cui è suddiviso lo spessore totale di 4 cm (dai gangli della base ai ventricoli laterali). La scansione CTP cine (80 kVp; 100 mAs; matrix size = 512 x 512; FOV = 25-cm; scan type = cine full 1 sec.; rotation = 0.5 sec.; total scan time 50 sec) è stata avviata 5 secondi dopo l’iniezione automatica di 40 ml di mezzo di contrasto non ionico (Iomeron 300 mg/ml, Bracco Imaging SpA) alla velocità di 4 ml/sec. Elaborazione dei dati Tutte le scansioni CTP sono state valutate con un algoritmo di deconvoluzione sensibile al ritardo utilizzando una stazione di elaborazione delle immagini (Advantage Windows; GEHelthcare). Sono state generate mappe CTP (CBF, CBV ed MTT) per ogni paziente. I valori di CBF, CBV e MTT sono stati espressi in ml/100g/min, ml/100g e secondi rispettivamente. Dopo averle identificate mediante analisi visiva sulle mappe MTT e CBV, l’ipoperfusione totale, il core infartuale e la penombra ischemica sono state calcolate disegnando manualmente tali lesioni rispettivamente sulle mappe MTT, CBV ed MTT-CBV (mismatch CTP). Il punteggio relativo al carico trombotico (clot burden score – CBS) ed allo stato dei circolai collaterali di compenso (regional leptomeningeal collaterals – rLMC) sono stati calcolati sulla CTA all’esordio. Un CBS ≤ 7 e> 7 (range 0-10) ed un rLMC ≤ 10 e> 10 (range 0-20) sono rispettivamente stati classificati come poveri e buoni. La ricanalizzazione è stata valutata a 24 sulla CTA secondo i criteri TICI. La riperfusione è stata definita come riduzione percentuale > 80% della lesione MTT basale rispetto a quella documentata a 24 ore. Il volume dell'infarto finale è stato delineato manualmente sulla TC di controllo (NCCT) eseguita a 7 giorni. Analisi statistica La distribuzione dei dati di ciascuna variabile è stata verificata attraverso il test di Kolmogorov-Smirnov. In caso di distribuzione normale dei dati, le differenze fra i valori medi di ciascuna variabile (valori assoluti di CBF, CBV, MTT e volume di lesione totale, core e penombra) sono state valutate mediante il t test di Student per campioni indipendenti. Nel caso invece di distribuzione non normale, i valori medi di ogni variabile sono stati confrontati utilizzando il Mann-Whitney U test. Dato che almeno una variabile presentava una distribuzione non normale, le correlazioni fra il volume della lesione totale, del core e della penombra e il Clot Burden Score (CBS), il Collateral Score (rLMC), il volume della lesione finale (calcolato dalla TC standard eseguita a 7 giorni) e l’outcome del paziente (calcolato mediante la mRS somministrata a 3 mesi) sono state valutate mediante il test del coefficiente di correlazione di Spearman.E’ stato considerato statisticamente significativo, un valore di p < 0.05. RISULTATI I valori CTP assoluti e volumetrici rilevati in tre differenti ROI analizzate, hanno mostrato che i valori assoluti di CBF e CBV misurati in core e penombra erano in accordo con quelli riportati in studi precedenti. Inoltre, un mismatch CTP indicativo della presenza di penombra ischemica è stato rilevato all'esordio in tutti i pazienti. Il volume della lesione totale, del core infartuale e della penombra ischemica erano significativamente inferiori (p < 0.0001) nei pazienti con basso carico trombotico rispetto a quelli con alto carico trombotico. Le dimensioni della lesione totale e del core infartuale si dimostravano significativamente più piccole nei pazienti con buona circolazione collaterale di compenso rispetto a quelli con scarsi circoli collaterali (p < 0.05 e p < 0.001, rispettivamente). Nessuna differenza statistica compariva per il volume della penombra in queste due categorie di pazienti. Il volume della lesione totale, del core infartuale e della penombra ischemica erano significativamente inferiori (p < 0.001, p < 0.01 e p < 0.001, rispettivamente) nei pazienti ricanalizzati rispetto a quelli con occlusione persistente. Le dimensioni della penombra risultavano significativamente più piccole (p < 0.02) nei pazienti riperfusi rispetto a quelli non riperfusi. Nessuna differenza statistica compariva per il volume della lesione totale e del core infartuale in queste due categorie di pazienti. Le dimensioni della lesione totale, del core infartuale e della penombra ischemica si dimostravano significativamente più piccole (p < 0.0001, p < 0.0001 e p < 0.01, rispettivamente) nei pazienti con outcome favorevole rispetto a quelli con outcome sfavorevole. Questi risultati sono stati confermati dalle correlazioni osservate tra i volumi CTP ed in parametri fisiopatologici, clinici e radiologici analizzati. DISCUSSIONE I nostri risultati suggeriscono che: -i valori assoluti misurati nel core infartuale e nella penombra sono sostanzialmente in accordo con i dati precedentemente riportati, indicando che l'identificazione di queste due regioni mediante analisi visiva è affidabile. -la TC perfusionale rappresenta uno strumento utile per predire la prognosi dei pazienti con ictus acuto; l'estensione dei volumi MTT (ipoperfusione totale) e CBV (core infartuale) è infatti fortemente legata ai parametri fisiopatologici ed alle misure di outcome clinico e radiologico; in particolare, la mappa CBV sembra essere il principale indicatore prognostico in accordo con alcuni studi precedenti. Complessivamente pertanto, la valutazione delle “classiche” mappe CTP (MTT, CBF e CBV) ed in particolare della mappa CBV, mediante un’analisi visiva rappresenta ancora un metodo affidabile per descrivere i principali aspetti dell’emodinamica cerebrale nell'ictus ischemico acuto. Questo è particolarmente importante nella pratica clinica quotidiana nell’ottica di implementare il più possibile il numero di pazienti trattati e di predire il loro esito funzionale a lungo termine.

Affidabilità delle mappe TC perfusionali nel definire il core infartuale e la penombra ischemica in pazienti con ictus ischemico acuto

TAMBORINO, Carmine
2016

Abstract

BACKGROUND Ischaemic stroke accounts for 80% of all acute cerebrovascular events. Stroke is responsible for 10-12% of deaths (about 6 million) per year, making it the second leading cause of death worldwide and the first leading cause of disability in developed countries. 70% of all cerebral infarctions is secondary to middle cerebral artery occlusion. Within the hypoperfused tissue 3 regions are identified: - the infarct core: strictly hypoperfused and not more vital tissue; it progresses to necrosis even if reperfused (irreversibly damaged) - the ischaemic penumbra (around the ischemic core): hypoperfused but still vital tissue; it progressively evolves to necrosis if not reperfused (reversibly damaged) - the benign oligoemia (around the ischemic penumbra): moderately hypoperfused tissue and usually not at risk for necrosis; it recovers spontaneously also without reperfusion. Currently, the major acute phase therapies include intravenous thrombolysis and endovascular treatments. Aim of these therapies is to recanalise occluded vessels and to reperfuse the ischaemic tissue (penumbra), resulting in an improvement of neurological symptoms and prognosis. However, the main parameter to get access to reperfusion therapies is represented by the clinical onset time with a strict time window. Therefore, only a small percentage (15-20%) of people with acute ischemic stroke is actually eligible for this kind of therapies; moreover, despite recanalization, many of these patients (25-50%) do not reach a good clinical outcome; on the contrary 35% of them have a cerebral infarction growth. Therefore, it is clear that other factors (in addition to time) are involved in clinical outcome of acute ischemic stroke patients. In this context, the Neuroimaging, through the analysis of the most relevant pathophysiological aspects of acute cerebrovascular diseases, aims to contribute to the efficacy and safety of therapeutic strategies, improving and leading patients selection for reperfusion therapies. In order to meet these needs a multimodal CT protocol has been proposed. This protocol, by using non contrast CT scan (NCCT), CT angiography (CTA) and CT-perfusion (CTP), tries to provide informations about the absence of an haemorrhage (ICH), vessels patency, regional leptomeningeal collateral state (CTA) and haemodynamic consequences of vascular occlusion (CTP). This protocol is relatively fast and available in many Centers. CTP is the only able to describe the main aspects of cerebral hemodynamics. It studies the first transit of a contrast bolus through brain capillaries. Densitometric data are then analyzed according to a mathematical model called deconvolution. Through deconvolution is possible to obtain a density/time curve called Residual Function. The height of the peak of this curve quantifies the CBF (cerebral blood flow – ml blood/100 g tissue/min) and the area under the curve the CBV (cerebral blood volume - ml blood/100g tissue). MTT (mean transit time - seconds) is then calculated on the basis of the "central volume principle " (CBF = CBV / MTT). Based on deconvolution, finally we obtain the so-called parametric perfusion maps. "Classically": 1) Total hypoperfusion (core + penumbra) = extension of CBF and MTT injury 2) Core infarct size = CBV injury 3) ischaemic penumbra volume = CBF or MTT - CBV volume (CBF/MTT - CBV mismatch). AIMS OF THE STUDY Several new findings have questioned the validity of "classic" CTP approach to define cerebral haemodynamics and its reliability in order to describe ischaemic regions (core and penumbra) and to predict outcome of acute ischaemic stroke patients. Therefore, aim of this study was to verify: 1) if a visual analysis approach of “classical” CTP parameters is reliable to identify and describe cerebral hemodynamics in acute ischaemic stroke and if it could be still used in clinical practice 2) if the classical mismatch (CBF/MTT – CBV) the so-called "penumbral hypothesis" can still has a value to predict patients outcome. For this reason we decided to measure CBF, CBV and MTT absolute value and the volume of the whole ischaemic lesion, infarct core and penumbra in a broad number of patients with acute ischaemic stroke. These measures were than coupled with: -the clot burden and the status of regional leptomeningeal collaterals (onset conditions and outcome predictors at the same time) -recanalization and reperfusion (conditions related to therapy issues) -final infarct volume and disability (mRS) (long-term outcome measures). As, infarct core was measured on CBV map and ischemic penumbra by the MTT - CBV mismatch, our study aims to verify whether a qualitative approach may still has a value to predict the fate of ischaemic tissue and patients prognosis. METHODS Patient selection and study design We retrospectively evaluated 200 patients having middle cerebral artery acute stroke. All patients performed pre-treatment multimodal CT protocol including non-enhanced CT (NECT), CT Angiography (CTA) and CTP. Informed consent was obtained from the patient/next of kin and local ethics committee approved the protocol. The National Institute of Health Stroke Scale (NIHSS) was recorded at admission and at 3 months. The modified Rankin scale (mRS) was recorded at 3 months; mRS ≤ 2 and > 2 were classified as good and poor outcomes, respectively. Imaging protocol CTP studies were performed with a dynamic first-pass bolus-tracking methodology on a 64-section CT scanner (Lightspeed VCT 64, GE Helthcare) according to a one-phase imaging protocol consisting of an acquisition of 50-seconds continuous (cine) scans reconstructed at 0.5 second intervals to produce a series of 99 sequential images for each of 8 sections which covered a total of 4 cm from the basal ganglia to the lateral ventricles. Cine CTP scanning (80 kVp; 100 mAs; matrix size = 512 x 512; FOV = 25-cm; scan type = cine full 1 sec.; rotation = 0.5 sec.; total scan time 50 sec) was initiated 5 seconds after the automatic injection of 40 ml of non-ionic contrast agent (Iomeron 300 mg/ml, Bracco Imaging SpA) at the rate of 4 ml/sec. Data processing All CTP scans were assessed with a deconvolution-based delay sensitive algorithm by using an imaging workstation (Advantage Windows; GEHelthcare). CBF, CBV and MTT CTP maps were generated for each patient. CBF, CBV and MTT values were expressed in ml/100g/min, ml/100g and seconds, respectively. After identification by visual inspection on MTT and CBV maps, total ipoperfusion, infarct core and ischemic penumbra volumes were calculated by manually drawing MTT, CBV and MTT-CBV (CTP mismatch) lesions, respectively. Clot Burden Score (CBS) and regional LeptoMeningeal Collateral (rLMC) score were calculated on CTA at onset. CBS ≤ 7 and > 7 (range 0-10) and rLMC ≤ 10 and > 10 (range 0-20) were classified as poor and good, respectively. Recanalization was assessed on 24 hour CTA according to TICI criteria. Reperfusion was defined as a percentage reduction > 80% in the baseline-24 hour MTT lesion. The final infarct volume was manually outlined on follow-up 7 days unenhanced CT (NECT). Statistical analysis The normality of each variable was checked by using the Kolmogorov-Smirnov test. When normality of data distribution was found in all variables, statistical analysis was performed by a parametric approach. Conversely, when normality of data distribution was rejected in several variables, a non-parametric analysis was applied. Accordingly, continuous variables were compared by t-test, whereas their correlation were assessed by linear regression or Spearman's rank correlation coefficient, respectively. A value of p < 0.05 was accepted as statistically significant. RESULTS CTP volumetric and absolute values detected in the three different ischemic ROI analyzed, showed that the absolute values of CBF and CBV measured in core and penumbra were in agreement with those reported in previous studies. Furthermore, a CTP mismatch indicative of salvageable penumbra was detected at onset in all patients. MTT total hypoperfusion, CBV infarct core and (MTT/CBF-CBV) penumbra volumes were smaller (P<0.0001) in good than in poor CBS. Total hypoperfusion and infarct core were smaller in in good than in poor rlMC (p < 0.05 e p < 0.001, respectively). No differences were detected between MTT-CBV penumbra volumes in patients with good and poor rLMC. Total hypoperfusion, infarct core and penumbra volumes were smaller (p < 0.001, p < 0.01 e p < 0.001, respectively) recanalized than occluded patients. Penumbra volumes were smaller (p < 0.02) in reperfused than not reperfused patients. No differences were detected between total hypoperfusion and infarct core volumes in reperfused and not reperfused. Total hypoperfusion, infarct core and penumbra volumes were smaller (p < 0.0001, p < 0.0001 e p < 0.01, respectively) in good than poor outcome patients. These results were confirmed by correlations observed between CTP volumes and physiopathological, clinical and radiological parameters. DISCUSSION Our results suggest that: -absolute values measured in infarct core and penumbra were in agreement with data previously reported, indicating that the identification of these two regions performed by visual inspection was reliable. -CTP represents an useful tool to predict prognosis in acute stroke patients; the extent of MTT (total ipoperfusion) and CBV (infarct core) volumes is indeed strongly related to physiopathological, clinical and radiological outcome measures; in particular, the CBV map seems to be the most important determinant of outcome. This is in agreement with some previous studies. Thus, “classical” CTP maps (MTT, CBF and CBV) analysis by visual inspection (and particularly the CBV map) still represent a valuable method to describe cerebral haemodynamics in acute ischaemic stroke. This is most important from a daily clinical practice point of view in the aim to implement the number of treated patients and to predict their outcome.
GRANIERI, Enrico Gavino Giuseppe
CAPITANI, Silvano
File in questo prodotto:
File Dimensione Formato  
1.Tesi PhD_C.Tamborino.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: PUBBLICO - Pubblico senza Copyright
Dimensione 7.09 MB
Formato Adobe PDF
7.09 MB Adobe PDF Visualizza/Apri

I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11392/2403468
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact