Article Text
PDF
Abstract
Background Hyperperfusion syndrome after carotid interventions has a low incidence but it can lead to morbidity and mortality.
Objective To evaluate the usefulness of quantitative DSA for predicting hyperperfusion phenomenon (HPP) after carotid artery stenting and angioplasty.
Methods Thirty-three consecutive patients with carotid stenosis treated with carotid artery stenting or angioplasty between February 2014 and August 2016 were included. Color-coded digital subtraction angiograms showing the time to maximum contrast intensity of each image pixel were obtained from conventional DSA before and after procedures. The cerebral circulation time (CCT) was defined as the difference in the relative time to maximum intensity between arterial and venous regions of interest set on the angiograms. HPP was diagnosed straight after the procedure with qualitative 123I-IMP single-photon emission CT (SPECT). Cut-off points for detecting HPP for preprocedural CCT and periprocedural change of CCT were assessed by receiver operating characteristic analysis using 123I-IMP SPECT as reference standard.
Results 123I-IMP SPECT showed HPP in 4 of 33 patients. In these 4 patients, preprocedural prolongation of CCT (13.0±6.1 vs 7.2±1.3 s; p<0.001) was seen compared with patients without HPP as well as larger periprocedural changes of CCT (5.9±5.7 vs 0.5±1.3 s; p<0.001). The optimal cut-off points of preprocedural CCT and change of CCT for predicting HPP were 8.0 s (100% sensitivity, 69% specificity) and 3.2 s (75% sensitivity, 100% specificity), respectively.
Conclusions Preprocedural prolongation and greater periprocedural change of CCT are associated with the occurrence of HPP. Periprocedural evaluation of CCT may be useful for predicting HPP.
- angiography
- angioplasty
- stent
Statistics from Altmetric.com
Introduction
Carotid artery stenting (CAS) and carotid endarterectomy are established procedures for the prevention of further ischemic events of extracranial carotid stenosis.1 2 However, several complications, including cerebral embolism, vessel dissection, and hyperperfusion syndrome (HPS), may occur after CAS.3 Among them, despite the low incidence of HPS (1.1%) and HPS-related intracranial hemorrhage (0.7%) after CAS,4 HPS sometimes develops and leads to intracranial hemorrhage, resulting in morbidity and mortality.5–7 Treatment should thus be started when hyperperfusion phenomenon (HPP) is detected before HPS develops fully.
Color-coded digital subtraction angiography (DSA) provides visualization of a complete DSA run in a color-coded single image according to the relative time to maximum intensity and thereby provides measurement of cerebral circulation time (CCT). It allows real-time monitoring of periprocedural CCT without additional contrast medium and dosage of radiation in patients with steno-occlusive arterial disease.8 9
In this study, we evaluated preprocedural CCT and the change of CCT before and after the procedure by using quantitative DSA and investigated their usefulness for the prediction of HPP after CAS and carotid angioplasty.
Materials and methods
The clinical and radiological data of 33 patients with carotid artery stenosis treated with CAS or carotid angioplasty between February 2014 and August 2016 were reviewed. At our hospital, asymptomatic North American Symptomatic Carotid Enarterectomy Trial (NASCET) >80% stenosis lesions and symptomatic NASCET >60% stenosis lesions without vulnerable plaque are indications for CAS or carotid angioplasty. In patients with severe cerebral blood flow (CBF) impairment, carotid angioplasty was considered as the first session of staged CAS.10 This study was approved by the institutional review board of our hospital. Informed consent was obtained from all patients before the procedure.
Periprocedural management
All patients received dual antiplatelet therapy with two of the following three antiplatelet agents: 100 mg aspirin, 75 mg clopidogrel, and 200 mg cilostazol at least 2 weeks before the procedure. All procedures were conducted under local anesthesia using the transfemoral approach. During the procedure, the activated coagulation time was maintained for >250 s with intravenous bolus injection of heparin. All procedures were performed using embolic protection devices, including Mo.Ma Ultra (Medtronic, Minneapolis, Minnesota, USA) for proximal embolic protection and Carotid Guard Wire PS (Medtronic, Minneapolis, Minnesota, USA) for distal embolic protection. After the procedure, anticoagulation therapy was continued by intravenous administration of argatroban (Daiichi Sankyo, Tokyo, Japan) for 48 hours. The dual antiplatelet therapy was continued for at least 1 month, followed by single antiplatelet therapy.
Cerebral circulation time measurement
All DSA procedures were performed with a biplane flat-panel angiography system (Artis zee BA Twin; Siemens Healthcare GmbH, Forchheim, Germany). The frequencies of the sequences were four frames per second for the arterial to capillary phase and two frames per second for the venous phase. The contrast medium (Omnipaque 300, iodine 300 mg/mL; Daiichi Sankyo, Tokyo, Japan) was injected through a 9 Fr balloon catheter (Mo.Ma Ultra, Medtronic) or a 6 Fr guiding catheter (Flexor Shuttle Guiding Sheath; Cook Medical, Bloomington, Indiana, USA) placed at the common carotid artery at a speed of 6 mL per 5 s using a power injector (PRESS DUO; Nemoto Kyorindo Co, Ltd, Tokyo, Japan). The color-coded digital subtraction angiogram was obtained with postprocessing software (syngo iFlow; Siemens Healthcare GmbH, Forchheim, Germany). The arterial region of interest (ROI) was set at the petrous portion of the internal carotid artery (ICA), and the venous ROI was set at the venous sinus (superior sagittal sinus to transverse sigmoid sinus) using the lateral view of color-coded DSA. CCT was defined as the difference of the time to peak intensity between the arterial and venous ROIs. ΔCCT was calculated with the following formula: ΔCCT=preprocedural CCT – postprocedural CCT.
Diagnosis and definition of HPP
HPP was diagnosed with qualitative 123I-IMP SPECT performed after the procedures. Patients received 167 MBq of 123I-IMP via intravenous infusion. Fifteen minutes after injection, the first scan for CBF was started with 30 min of acquisition time. Irregular mirror-shaped ROIs were set at ipsilateral and contralateral middle cerebral artery (MCA) territory. HPP was defined as a CBF increase of >100%, as compared with the normal hemisphere.11
Statistical analysis
All data are expressed as mean and SD. Differences between patients with and without HPP were analyzed by Student’s t test for continuous variables and Fisher’s exact test for categorical variables. A p value of <0.05 was considered statistically significant. Receiver operating characteristic curve analysis of preprocedural CCT and ΔCCT was performed for the prediction of HPP, with 123I-IMP SPECT as standard of reference. All statistical analyses were performed with JMP version 11 software (SAS Institute Inc, Cary, North Carolina, USA).
Results
Between February 2014 and August 2016, 28 CAS and five carotid angioplasties were performed in 33 consecutive patients. (27 male patients (81.8%)). The mean±SD age of all patients was 70.7±6.4 years (range 56–82 years). The mean±SD degree of stenosis was 84.5±7.0% (range 70–95%) according to the NASCET criteria. Twenty-one (63.6%) lesions were symptomatic.
In four (12%) patients, HPP was observed in postprocedural qualitative 123I-IMP SPECT. There were no symptoms or intracranial hemorrhage related to HPP. The characteristics of patients with and without HPP are shown in table 1.
Characteristics of patients with and without hyperperfusion
The preprocedural CCT in patients with HPP was prolonged compared with that in patients without HPP (13.0±6.1 s vs 7.2±1.3 s; p<0.01). There was no significant difference in postprocedural CCT between patients with and without HPP (7.1±1.6 s vs 6.7±1.3 s; p=0.70; figure 1A). The ΔCCT was significantly greater in patients with HPP (5.9±5.7 s vs 0.5±1.3 s; p<0.01; figure 1B). Figure 2 shows the case of a patient with prolonged preprocedural CCT who developed HPP.
Periprocedural cerebral circulation time (CCT) (A) and periprocedural change of CCT (ΔCCT) (B) in patients with and without hyperperfusion phenomenon (HPP).
Images obtained from a 50 s patient with symptomatic right internal carotid artery stenosis (ICS). (A) Preprocedural carotid angiogram showing severe right ICS. (B) An arterial region of interest (ROI) (red arrow) and a venous ROI (blue arrow) set on a color-coded digital subtraction angiogram obtained from the lateral view of preprocedural intracranial DSA. (C) Time intensity curves of arterial (green) and venous (orange) ROIs. The preprocedural cerebral circulation time (CCT) was 8.8 s. (D) A carotid angiogram obtained after carotid artery stenting (CAS) with distal balloon protection. (E) Arterial (red arrow) and venous (blue arrow) ROIs set on a color-coded digital subtraction angiogram obtained from the lateral view of postprocedural intracranial DSA. (F) Time intensity curves of arterial (green) and venous (orange) ROIs. The postprocedural CCT was 5.1 s. (G) Qualitative 123I-IMP SPECT obtained immediately after CAS showing hyperperfusion phenomenon (HPP) in the right hemisphere. (H) Qualitative 123I-IMP SPECT obtained 6 days after CAS showing improvement of HPP.
The receiver operating characteristic curves of preprocedural CCT and ΔCCT for the prediction of HPP with 123I-IMP SPECT as standard of reference are shown in figure 3. The optimal cut-off point of preprocedural CCT (8.0 s) predicted HPP with 100% sensitivity and 69% specificity. The optimal cut-off point of ΔCCT (3.2 s) predicted HPP with 75% sensitivity and 100% specificity.
Receiver operating characteristic curves of preprocedural cerebral circulation time (CCT) (A) and periprocedural change of CCT (ΔCCT) (B) for prediction of hyperperfusion phenomenon. AUC, area under the curve.
Discussion
HPS is one of the main complications after CAS and carotid angioplasty. Despite its low incidence, HPS sometimes leads to intracranial hemorrhage, resulting in morbidity and mortality.5–7
In this study, we evaluated the usefulness of preprocedural and periprocedural measurements of CCT using qualitative DSA for prediction of HPP after CAS or carotid angioplasty.
In patients with HPP, preprocedural CCT was significantly prolonged and ΔCCT was significantly greater than those in patients without HPP.
Several authors have investigated the correlation between CCTs measured by angiography and by other modalities. Gado et al demonstrated the correlation between angiographic CCT and mean transient time using 15C-labeled hemoglobin.12 Yamamoto et al demonstrated the correlation between cerebral vasoreactivity measured with SPECT with acetazolamide challenge and arterio-capillary CCT measured through visual observation in patients with unilateral occlusive lesions in the ICA or MCA.13 Aikawa et al investigated the periprocedural change of CCT in patients treated with CAS and reported the correlation between ΔCCT and change in %CBF measured with SPECT.14 These results are in line with our findings that prolonged preprocedural CCT may reflect the impairment of both CBF and cerebral vasoreactivity and indicate severe impairment of cerebral circulation.
Lin et al investigated the CCT in 34 normal subjects using quantitative DSA. According to their report, the relative time to maximum intensity of the ICA and sigmoid sinus were 0.10±0.11 s and 6.51±1.10 s, respectively.8 Compared with these results, the preprocedural CCT in patients of our study with HPP was prolonged, and the postprocedural CCT in both groups was normal. Our results also suggest that prolonged preprocedural CCT potentially leads to greater ΔCCT, which results in HPP. This may reflect the pathophysiology of HPP and HPS, which includes hemodynamic changes in cerebral vessels with impairment of autoregulation and cerebral vessel disruption.
There are few reports of using quantitative DSA to predict HPS. Lin et al reported that a stenotic transverse sinus leads to the prolongation of preprocedural CCT and shortening of CCT after CAS, and is associated with HPS.15
Our study suggested that the sensitivity and specificity of preprocedural CCT and periprocedural change of CCT may be high; however, the cut-off points need to be validated in larger studies.
Quantitative DSA enables real-time CCT monitoring without additional invasion, contrast medium, and irradiation. This real-time evaluation allows immediate decisions to be made about the treatment strategy for preventing hyperperfusion, such as severe management of blood pressure or staged CAS.10
The limitations of this study include the small number of cases and that it was a non-randomized, single-center retrospective study. Further investigations with a larger number of cases are needed to validate our results.
Conclusions
In patients with HPP, preprocedural CCT was prolonged and ΔCCT was greater than in patients without HPP. Periprocedural evaluation of CCT using quantitative DSA may be useful for the prediction of HPP.
Acknowledgments
The authors gratefully acknowledge Siemens Healthcare K.K., Japan for encouraging this work.
References
Footnotes
Contributors KY: drafting of manuscript, study design, data analysis. YEn: study design, data collection, revision of the drafted manuscript, KO: revision of the drafted manuscript. YEg: data collection. TI: study design, revision of the drafted manuscript.
Competing interests None declared.
Patient consent Obtained.
Ethics approval Ethics approval was obtained from the institutional review board of Gifu University Graduate School of Medicine.
Provenance and peer review Not commissioned; externally peer reviewed.