Article Text

Download PDFPDF

Original article
Flow diversion versus traditional aneurysm embolization strategies: analysis of fluoroscopy and procedure times
  1. Nohra Chalouhi1,
  2. Jeffrey F McMahon1,
  3. Lea A Moukarzel1,
  4. Robert M Starke1,
  5. Pascal Jabbour1,
  6. Aaron S Dumont1,
  7. Stavropoula Tjoumakaris1,
  8. Eric L Gingold2,
  9. Robert Rosenwasser1,
  10. L Fernando Gonzalez1
  1. 1Department of Neurosurgery, Division of Neurovascular Surgery and Endovascular Neurosurgery, Thomas Jefferson University and Jefferson Hospital for Neuroscience, Philadelphia, Pennsylvania, USA
  2. 2Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA
  1. Correspondence to Dr L F Gonzalez, Department of Neurological Surgery, Division of Neurovascular Surgery and Endovascular Neurosurgery, Thomas Jefferson University and Jefferson Hospital for Neuroscience, 901 Walnut Street 3rd Floor, Philadelphia, PA 19107, USA; fernando.gonzalez{at}


Background and objective Flow diverters are increasingly used for treatment of complex intracranial aneurysms. The purpose of this study was to compare the pipeline embolization device (PED) and traditional embolization strategies in terms of fluoroscopy and procedure time.

Material and methods Fluoroscopy and procedure times (in minutes) were retrospectively analyzed and compared between 127 patients treated with the PED, 86 patients treated with single stage stent assisted coiling (SAC), and 16 patients treated with Onyx HD 500 at our institution. A multivariate logistic regression analysis was performed to determine independent predictors of fluoroscopy and procedure time.

Results The three groups were comparable with respect to patient age, gender, and ruptured/unruptured aneurysm status. Aneurysms treated with the PED were significantly larger than stent coiled aneurysms, and aneurysm location distribution differed significantly between the three groups. Mean fluoroscopy time was significantly increased in the SAC (55±31 min, p<0.001) and Onyx HD 500 (91±36 min, p<0.001) groups relative to the PED group (34±23 min). Likewise, mean procedure time was significantly longer in SAC (155±50 min, p<0.001) and Onyx HD 500 (176±65 min, p<0.001) patients compared with PED patients (131±36 min). In multivariate analysis, SAC/Onyx HD 500 versus PED independently predicted longer procedure and fluoroscopy times.

Conclusions PED treatment requires significantly shorter fluoroscopy and procedure times compared with SAC and Onyx HD 500. The results of this study may be used by advocates of flow diverters as an additional argument for using this treatment modality to treat intracranial aneurysms.

  • Aneurysm
View Full Text

Statistics from


Flow diversion has emerged as an important tool for management of intracranial aneurysms.1–4 The pipeline embolization device (PED) received Food and Drug Administration approval in 2011 for the treatment of large or giant wide necked aneurysms from the petrous to the superior hypophyseal segments of the internal carotid artery. Several recently published series have shown that the PED had unparalleled efficacy in the treatment of large, giant, and complex aneurysms.5–14

Although the PED has been adopted as a firstline option at a number of institutions, some investigators remain wary of the high morbidity rate associated with flow diverters and continue to prefer traditional endovascular strategies.15 No clinical study has quantitatively compared fluoroscopy and procedure times between the PED and standard endovascular techniques. In this study, we compared the PED, stent assisted coiling (SAC), and Onyx HD 500 (eV3, Irvine, California, USA) embolization in terms of fluoroscopy and procedure time in a cohort of patients treated at our institution.


The university institutional review board approved the study protocol. Consecutive patients who underwent treatment with the PED at our institution between May 2011 and December 2012 were identified from a prospectively maintained database. We also identified consecutive patients who underwent single staged SAC or embolization with Onyx HD 500. Medical charts were retrospectively reviewed to determine patient demographics, aneurysm characteristics, and procedural specifics, including intraoperative complications as they can potentially affect procedure time. Fluoroscopy times (in minutes) were obtained from the report that the fluoroscopy biplane provides, then introduced into a spreadsheet (only as fluoroscopy time) by senior radiology technicians, and collected retrospectively. Procedure times were computed from the time of patient arrival to discharge from the angiography suite by the charge nurse.

SAC was generally indicated for wide necked aneurysms (>4 mm) or those with an unfavorable fundus to neck ratio (<1.5) and as a rescue when coils prolapsed into the parent vessel. Onyx HD 500 was indicated for wide necked aneurysms and for aneurysms recurring after previous embolization.

All patients were treated under general endotracheal anesthesia. The techniques for PED treatment, SAC, and Onyx HD 500 embolization have been detailed previously.16–18 Briefly, for PED procedures, femoral access was obtained with an 8 F femoral sheath. A 6 F shuttle sheath (Cook Medical, Bloomington, Indiana, USA) was placed in the distal cervical segment of the index vessel, followed by navigation of a 6 F 070 Neuron catheter (Penumbra, Alameda, California, USA). PEDs were then deployed through a Marksman microcatheter (ev3, Irvine, California, USA). The number of stents deployed was left to the operator's discretion. For SAC, we generally use the microcatheter ‘jailing’ technique in which the stent is deployed after the aneurysm is microcatheterized but before coil deployment. Two types of stents were used: the Neuroform (Stryker Neurovascular, Fremont, California, USA) and the Enterprise (Cordis Neurovascular, Miami, Florida, USA). Coiling was interrupted when the aneurysm was completely occluded or when no additional coils could be deployed. For treatment with Onyx HD 500, a DMSO compatible balloon was inflated at the neck of the aneurysm. Following a seal test, Onyx was injected under continuous fluoroscopic imaging at a rate of 0.2 ml/2 min, with the balloon continuously inflated contingent upon the stability of neuromonitoring. After the aneurysm had been adequately obliterated, a 3 min wait period with the balloon inflated was allowed for Onyx solidification. Finally, the balloon was deflated and a 10 min hold period was allowed for complete solidification of the cast.

Statistical analysis

Data are presented as mean and range for continuous variables and as frequency for categorical variables. Analysis was carried out using unpaired t test, χ2, and Fisher's exact tests. Assessment of means of more than two groups was carried out by analysis of variance with post hoc analysis, with Bonferoni correction as appropriate. Assessment of normality was assessed graphically, numerically, and statistically with the Shapiro–Wilk test. As fluoroscopy time was not normally distributed, the logarithmic of fluoroscopy time was transformed using the logarithmic function to satisfy normal distribution. Univariate linear regression analysis was used to test covariates predictive of the following dependent variables: procedure time and logarithmic fluoroscopy time. Interaction and confounding was assessed through stratification and relevant expansion covariates. Factors predictive in univariate analysis (p<0.20)19 were entered into a multivariate linear regression analysis. A p value ≤0.05 was considered statistically significant. Statistical analysis was carried out with Stata V.10.0 (College Station, Texas, USA).


The study population included 127 patients treated with the PED, 86 patients treated with SAC, and 16 patients treated with Onyx HD 500.

PED therapy

A total of 127 patients, including 102 women and 25 men ranging in age from 17 to 88 years (mean 56 years), were treated with the PED. Mean aneurysm size was 10 mm. Most treated aneurysms were located in the anterior circulation (table 1). Three patients (2.4%) were treated in the setting of subarachnoid hemorrhage. The number of PEDs used was 1.37 per patient. A single PED was used in 91 (71.6%) patients, two PEDs in 28 (22%), three PEDs in five (3.9%), four PEDs in two (1.7%), and five PEDs in one (0.8%). Eleven aneurysms (8.7%) were treated with adjunctive coils (mean 2.2 coils) in addition to the PED. Intraoperative events occurred in six (4.7%) patients (internal carotid artery dissection in two, stent deployment complications in four).

Table 1

Aneurysm locations

Stent assisted coiling

Of 86 patients treated with single stage SAC, 74 (86%) were women and 12 (14%) were men, with a mean age of 55.5 years (range 24–80 years). Aneurysm size was 7.5 mm on average. Aneurysms were predominantly located in the anterior circulation (table 1). The mean number of coils used was 5.6 per aneurysm. A single stent was used in 75 (87.2%) patients, two stents in 10 (11.6%) patients, and three stents in one (1.2%) patient. Six patients (7%) were treated in the acute setting of subarachnoid hemorrhage. Intraoperative complications occurred in two (2.3%) patients (stent deployment complication in one, intraoperative aneurysm rupture in one).

Onyx HD 500

Of 16 patients treated with Onyx HD 500, 13 (81%) were women and three (19%) were men, with a mean age of 57.4 years. Mean aneurysm size was 11.3 mm. All treated aneurysms were located in the anterior circulation (table 1). One patient (6.3%) had an acutely ruptured aneurysm. The total volume of Onyx used was 1.57 ml on average. There were no intraoperative complications in this group.

Comparison between the three groups

The three groups were comparable with respect to patient age (p=0.09), gender (p=0.3), and ruptured/unruptured aneurysm status (p=0.2). Aneurysm location (p=0.007) and size (p<0.001) differed significantly between the three groups. Aneurysms treated with the PED were significantly larger than stent coiled aneurysms (p=0.01). There was no statistically significant difference in aneurysm size between the Onyx and SAC groups (p=0.1) or between the Onyx and PED groups (p=0.99). The rate of intraoperative complications did not differ significantly between the three groups (p=0.7).

Mean fluoroscopy time was significantly increased in the SAC (55±31 min, p<0.001) and Onyx HD 500 (91±36 min, p<0.001) groups relative to the PED group (34±23 min). The difference in fluoroscopy time between SAC and Onyx HD 500 was also statistically significant (p<0.001). Mean procedure time was significantly increased in SAC (155±50 min, p<0.001) and Onyx HD 500 (176±65 min, p<0.001) patients relative to PED (131±36 min) patients but was comparable between the SAC and Onyx HD 500 cohorts (p=0.3).

Of note, mean fluoroscopy time remained significantly longer in SAC (p=0.04) and Onyx HD 500 (p<0.001) patients compared with those who underwent placement of more than one PED (40±20 min). Likewise, mean procedure time remained significantly longer in the Onyx HD 500 group (p<0.001) relative to treatment with multiple PEDs (142±32 min) but the difference between SAC and treatment with multiple PEDs was no longer significant (p=0.1).

In multivariate analysis, predictors of longer fluoroscopy time were SAC/Onyx HD500 versus PED (OR=0.6; p<0.001), larger aneurysm size (OR=0.02; p=0.001), and intraoperative complications (OR=0.93; p<0.001). Likewise, treatment with SAC/Onyx HD 500 versus PED (OR=60; p<0.001) and larger aneurysm size (OR=1.6; p=0.001) predicted longer procedure time (table 2). The operator was not predictive of fluoroscopy or procedure time.

Table 2

Factors affecting procedure time and fluoroscopy time


The results of the International Subarachnoid Aneurysm Trial20 have established coiling as an important treatment modality for ruptured intracranial aneurysms. However, large, wide necked, and fusiform aneurysms pose significant challenges to conventional coiling and are associated with high rates of recanalization.21 This deficiency in the endovascular treatment of aneurysms has led to the development of newer techniques and materials to completely occlude these difficult to treat cerebral aneurysms. Specifically, liquid embolic agents and self-expanding stents, especially flow diverters, have placed even the most complex aneurysms within the reach of endovascular therapy.18 22 23 The PED is a dedicated flow diverter that has received considerable attention in the recent literature. Some investigators, however, remain skeptical of flow diverters and have been reluctant to embrace this new technology.

This is the first study to compare fluoroscopy and procedure times in patients treated with PED, SAC, and Onyx HD 500. We found that PED treatment required significantly less fluoroscopy and procedure time than SAC or Onyx HD 500. In multivariate analysis, after accounting for potentials confounders such as aneurysm size and intraoperative events, treatment with the PED remained a strong independent predictor of shorter fluoroscopy and procedure time. These findings are important to both the patient and operator because longer fluoroscopy and procedure times expose both the patient and operator to higher radiation doses. Among the side effects of radiation exposure are skin changes, including erythema, desquamation, and hair loss, which are commonly referred to as deterministic effects (ie, severity of effects increases with dose) of radiation and may occur with exposure of 2 Gy.24 ,25 The risk of carcinogenesis (benign and malignant brain tumors) is the primary stochastic effect (ie, probability of effect but not severity increases with dose) of radiation exposure. In a recent well designed study, Peterson et al26 examined the rates of skin and hair effects from radiation exposure following 702 neuroendovascular procedures and found that nearly 40% of patients receiving >2 Gy developed subacute skin or hair changes, of which 30% were permanent. Importantly, up to 72% of interventional procedures involved an entrance skin dose >2 Gy. Thus the issue of radiation exposure is an important one and should be taken into consideration when evaluating different treatment options for an aneurysm. With regard to treatment cost effectiveness, shorter procedure times, improvement in treatment efficacy, and better resource utilization may counterbalance the elevated cost of the PED.

Several reasons account for the shorter fluoroscopy and procedure times with the PED compared with traditional modalities. In fact, many of the basic techniques that are applicable to the endosaccular occlusion of aneurysms are completely different for PED. First, the pipeline procedure is relatively straightforward and simple, once the access has been established and the aneurysm neck is bridged and does not require multiple devices as with SAC (jailed microcatheter and stent delivery system) or Onyx embolization (microcatheter and balloon catheter). Also, selective aneurysm catheterization and endosaccular coil deployment are usually not necessary with the PED. This substantially affects procedure time because the mean number of coils necessary to occlude large and giant aneurysms is reportedly to be as high as 16.1 and 28.7, respectively.27 It should be noted, however, that the deployment process of the PED is much slower than with other self-expanding stents. Lastly, the need for a seal test, slow injection of Onyx HD 500, repeated cycles of balloon inflation–deflations, and wait time for solidification of Onyx explain why Onyx HD 500 is a lengthier procedure than PED or even SAC. This study did not analyze the cumulative radiation dose for repeat embolization; considering that flow diversion provides more definitive aneurysm occlusion than SAC or Onyx HD 500,6 ,9–,10 ,28–30 there is theoretically less need for further fluoroscopic imaging and radiation exposure with the PED.

Factors that can potentially influence the procedure and fluoroscopy time with the PED are the diameter of the delivery system, number of devices used, and adjunctive use of coils. The difficulty of the PED procedure is primarily related to the use of larger diameter delivery systems (currently a 0.027 inch catheter is required) which could create access problems, especially in the setting of aortic arch and carotid tortuosity. PED deployment would have been an easier and faster procedure if smaller diameter delivery systems were available. For distal aneurysms, greater support is needed from the system, and larger delivery system diameters are necessary. With regard to the number of PEDs used per aneurysm, remarkably high rates of aneurysm occlusion were achieved in recent series despite the use of only a single device in most cases.9 The use of a single device in more than 70% of aneurysms in this series is certainly one of the factors that contributed to the relatively low fluoroscopy and procedure times with the PED. Finally, some operators have used intraluminal coils to attenuate the effects of intra-aneurysmal thrombosis and prevent delayed aneurysm rupture but this strategy remains controversial. We rarely use any coils in addition to the PED unless the patient has sustained a subarachnoid hemorrhage.

Very few studies have directly compared flow diversion with standard endovascular techniques. In a small series of 42 patients, Lanzino et al10 reported similar rates of morbidity and a higher rate of complete angiographic obliteration with the PED compared with standard endovascular techniques, and concluded that careful long term follow-up was important to definitively validate flow diversion as a superior treatment for proximal internal carotid artery aneurysms. In a prospective, non-randomized, multicenter study, Yu et al29 reported a 3.5% risk of periprocedural death or major stroke with the PED with complete aneurysm occlusion in 84% at 18 months and concluded that the PED should be considered a first choice for treating unruptured aneurysms and recurrent aneurysms. On the other hand, a recent systematic review on unruptured aneurysms treated with endovascular techniques found that flow diversion and liquid embolic agents were associated with significantly higher risks compared with other endovascular techniques.31 While Saatci et al9 treated 191 patients using the PED with a permanent morbidity rate as low as 1%, O'Kelly et al32 reported a combined morbidity and mortality rate of 10.7% (6.3% mortality, 4.4% morbidity). Likewise, in a multicenter study that included 56 patients treated with the PED, Kan et al33 reported six periprocedural thromboembolic events and four fatal postprocedural hemorrhages for a major complication rate of 8.5%. These data highlight the need for randomized controlled trials comparing flow diversion with conventional endovascular strategies and surgical clipping.


This was a non-randomized, retrospective, single center study. The groups were well matched except for aneurysm location and size. The significant size difference between the PED and SAC groups provides additional support for the results of this study. If the mean aneurysm size of the SAC group was larger, a higher number of coils would be needed, and the fluoroscopy and procedure times would have been even longer. Lastly, the absolute doses of radiation (kerma) were not available for comparison, but fluoroscopy time has shown a correlation with radiation dose.34–36 A previous well designed study comparing radiation exposure with different embolic agents in the treatment of intracranial arteriovenous malformations used fluoroscopy time as a surrogate marker of radiation exposure.37


In this study, we compared flow diversion and traditional embolization strategies in terms of fluoroscopy and procedure time. We found that PED treatment requires significantly shorter fluoroscopy and procedure times compared with SAC and Onyx HD 500. The results of this study may be used by advocates of flow diverters as an additional argument for using this treatment modality to treat intracranial aneurysms. Randomized, controlled trials comparing patient outcomes between flow diverters, coiling, and surgical clipping are needed to determine the best treatment modality.


View Abstract


  • Contributors All authors made direct and significant contribution to the manuscript. LFG is the guarantor of the study. All authors contributed to the study design, data analysis, manuscript drafting, manuscript reviewing, and approval of the final version. RMS performed the statistical analysis. NC, RMS, PJ, LAM, and JFM collected the data. Literature review was done by NC and PJ.

  • Competing interests None.

  • Ethics approval The institutional review board of Thomas Jefferson University approved the study protocol.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.