Article Text

Original research
Initial experience using the Penumbra coil 400: comparison of aneurysm packing, cost effectiveness, and coil efficiency
1. James Milburn1,
2. Ankit Limba Pansara2,
3. Gabriel Vidal3,
4. Raisa Coral Martinez3
1. 1Department of Radiology, Ochsner Medical System, New Orleans, Louisiana, USA
2. 2Department of Radiology, Ochsner Clinic Foundation, New Orleans, Louisiana, USA
3. 3Department of Neurology, Ochsner Clinic Foundation, New Orleans, Louisiana, USA
1. Correspondence to Dr J Milburn, Department of Radiology, Ochsner Medical System, 1514 Jefferson Highway, New Orleans, LA 70121, USA; jmilburn{at}ochsner.org

## Abstract

Background/objective This was a retrospective single operator experience of aneurysm coiling at the Ochsner Clinic, comparing the Penumbra coil 400 with the other most commonly used coil types, Orbit and Galaxy.

Methods 40 aneurysms treated exclusively with Orbit or Galaxy coils and 18 aneurysms treated only with Penumbra coils from 2010 to February 2012 were compared. Measurements included packing density, number of coils per unit aneurysm volume, and total coil length per unit aneurysm volume.

Results Mean packing density of 33.7% using Penumbra coils was significantly greater than 24.4% for Orbit/Galaxy coils. The number of coils per aneurysm volume of 0.026 coils/mm3 for the Penumbra coil was significantly lower than 0.114 coils/mm3 for the Orbit/Galaxy coils. Aneurysm occlusion rates were similar in both groups.

Conclusions Compared with the Orbit/Galaxy coils, our analysis suggests that the Penumbra coil is more efficient and cost effective in the treatment of intracranial aneurysms.

• Aneurysm
• Angiography
• Brain
• Coil
• Subarachnoid

## 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.

## Introduction

Penumbra coil 400 gained Food and Drug Administration approval and was introduced to the market in April 2011, and early reports demonstrated safety and periprocedural efficacy.1 Due to its larger 0.020 inch diameter and inherent softness, there is potential to provide more efficient and cost effective coiling. More efficient coiling may result in shorter catheter in vessel time and less radiation exposure for the patient and healthcare team.

Physical properties unique to the Penumbra coil provide this potential for efficiency. The Penumbra coil provides between 400% and 178% more volume per unit length than traditional 0.010 inch and 0.015 inch coils, respectively. Because coil softness increases by the third power of the coil diameter, all other factors being equal,2 the Penumbra coil has the potential to be much softer than any of the traditional 10 (0.010 inch) or 18 (0.015 inch) system coils. The Penumbra coil’s final softness levels which they describe as ‘standard’, ‘soft’, and ‘extra soft’ are not determined by the filament diameter of the wire the coil is wound from, as in traditional coil systems, but rather by the thickness of the nitinol stretch resistance wire that runs the length of the coils. Finally, the Penumbra coil employs a ‘coil within a coil’ design that is intended to help it resist compaction.

Penumbra coils were initially marketed as an alternative device to treat larger aneurysms, but neurointerventionalists, including the authors, have also found them helpful in treating smaller aneurysms using fewer coils, quickly reaching high packing densities (figures 1 and 2). Because a larger diameter microcatheter is required for the Penumbra coils, there may be an increased risk when treating these smaller aneurysms.

Figure 1

Large aneurysm. (A) Pretreatment: a woman in her sixties with subarachnoid hemorrhage from a ruptured 15×12 mm right A-1 segment aneurysm was treated with balloon assisted coiling using a 3×10 Hyperglide balloon (Covidien) and 11 Penumbra coils achieving a 40% packing density. (B) Post-treatment shows complete occlusion of the aneurysm.

Figure 2

Small aneurysm. (A) Pretreatment: a woman in her forties with a 4×3 mm right internal carotid artery terminus aneurysm was treated with a single 3×5 mm Penumbra soft coil achieving 27.5% packing density. (B) Post-treatment images show complete occlusion of the aneurysm. There was no recanalization on an 11 month follow-up angiogram.

This was a retrospective single operator experience at the Ochsner Clinic comparing cerebral aneurysm coiling using the Penumbra coil 400 with Orbit and/or Galaxy coils. The Orbit and Galaxy coils were used as a comparison group because they statistically have been the most common coil types used at this institution for the past 3 years. Orbit and Galaxy coils are both manufactured by Codman, and they have a 0.012 inch diameter.

## Materials and methods

Aneurysm treatments performed exclusively with the Penumbra coil 400 were compared with those treated exclusively using the Orbit or Galaxy coils. There were 40 aneurysms in 40 patients treated using Orbit or Galaxy coils and 18 aneurysms in 17 patients treated only with Penumbra coils from January 2010 to February 2012. The first Penumbra treatments were in May of 2011. There were 36 other aneurysm treatments excluded during this time period because they were not initial treatments, other manufacturers’ coils were used, or they had mixed coil types. Measurements included packing density, number of coils per unit aneurysm volume, and total coil length per unit aneurysm volume. Packing density was measured using AngioCalc, with measurements from both three-dimensional and two-dimensional images. Other parameters including presenting aneurysm rupture status, frequency of stent/balloon use, aneurysm occlusion rates, and complications were also compared. Aneurysm occlusion rates and follow-up imaging were graded using the Raymond Scale.3 Statistical analyses were performed using the Wilcoxon rank sum test. Institutional review board approval was obtained and the study was fully Health Insurance Portability and Accountability Act compliant.

For this analysis, we chose to evaluate the number and total length of coils used per unit volume of aneurysm. Others have compared only size matched aneurysms4 ,5 but it is believed that using such methodology could have added selection bias to this analysis. The method used in this paper was to include all aneurysms that met the selection criteria rather than investigator selected cases. By analyzing the number of coils and length of coils used per unit volume in this more heterogeneous group, we believe the results are more easily generalized to a wide variety of aneurysm sizes and shapes.

## Results

The Orbit/Galaxy group had an age range of 31–84 years old; the Penumbra group consisted of an age range of 30–66 years old. The Orbit/Galaxy arm had 33 anterior circulation and seven posterior circulation aneurysms. The Penumbra arm had 15 anterior and three posterior circulation aneurysms. There were more ruptured aneurysms (27/40) treated with the Orbit/Galaxy coils than with the Penumbra coil (5/18). Aneurysm sizes ranged from 6.3 mm3 to 5564.0 mm3 in the Orbit/Galaxy group and from 31.0 mm3 to12 666.4 mm3 in the Penumbra group. The largest diameter of the aneurysm was 7 mm or less in 27/40 cases treated with the Orbit/Galaxy coils and in 8/18 cases treated with the Penumbra coil. The percentage of cases performed with balloon or stent assistance was 25% in the Orbit/Galaxy group and 28% in the Penumbra group.

Mean packing density of 33.7% using Penumbra coils was significantly greater than 24.4% for Orbit/Galaxy coils (p=0.0001). The number of coils per aneurysm volume of 0.026 coils/mm3 for the Penumbra was significantly lower than 0.114 coils/mm3 for the Orbit/Galaxy coils (p<0.001). Total coil length per aneurysm volume of 0.11 cm/mm3 for the Penumbra coil was significantly less than 0.32 cm/mm3 for the Orbit/Galaxy treatments (p<0.001) (table 1). Using the Raymond Scale, there were 22 aneurysms assessed as class 1, 10 as class 2, and eight as class 3 treated with the Orbit/Galaxy coils. For the Penumbra coil, there were 10 aneurysms in class 1, three in class 2, and four in class 3. There were no procedural complications in the Penumbra cases, and there were two small branch infarctions in the Orbit/Galaxy group that were not clinically apparent. Both of the patients with small infarcts presented as Hunt and Hess grade 3 subarachnoid hemorrhages.

Table 1

Comparisons by coil type

Follow-up imaging with angiography or MRI was available in 25/40 patients in the Orbit/Galaxy group, study times ranging from 1 month to 22 months with an average follow-up time of 11.4 months. There were 15 whose Raymond grade was unchanged, four who were worse, and six who were improved. There was one aneurysm retreatment performed with stent coiling in a previously ruptured 5 mm posterior communicating artery aneurysm. The Penumbra group had follow-up in 11/18 cases, ranging from 2 weeks to 17 months with an average follow-up time of 8.6 months. There were 10 unchanged and one recurrence that was retreated. This was performed in a previously ruptured 25 mm posterior communicating artery aneurysm with mural thrombus and a 14 mm lumen. The coil mass migrated into the thrombus and was retreated with three additional coils.

## Discussion

Significantly greater average packing density using less than 25% of the number of coils per volume was achieved with the Penumbra coil compared with the Orbit/Galaxy coils. Aneurysm occlusion rates and complication rates were similar and acceptable in both groups. The safety of coiling in each group was similar, with no clinically identifiable complications. There were two patients in the Orbit/Galaxy arm who developed small infarcts in the inferior cerebellum. Both patients were vertebral artery posterior inferior cerebellar artery origin aneurysm treatments, and the lesions on CT could be attributed to slow filling posterior inferior cerebellar artery branches on angiography at the end of the procedures. One of these was performed with balloon assistance, and both were considered to have wide necks. Both of the patients with small infarcts presented as Hunt and Hess grade 3, Fisher grade 3 subarachnoid hemorrhages. The Orbit/Galaxy group had more ruptured aneurysms than the Penumbra group, which may have given that group a higher risk of thromboembolic complications.6 There were no complications in the 18 Penumbra cases.

Investigators have shown that the softness combined with the large volume of the Penumbra coil makes it possible to rapidly achieve very high packing densities in some aneurysms.7 In the opinion of the authors, more aggressive packing beyond 35–40% has little or no benefit. Such high density packing would seem to only increase the complication risk, and it certainly increases material costs. By attempting to pack beyond this level, the benefits of the Penumbra coil's increased efficiency are reduced. In this series of patients, the operator (JM) had the goal of safely occluding the aneurysms without over packing.

For a given volume of aneurysm in this study, coiling was performed with 0.23 Penumbra coil for every Galaxy/Orbit coil. This was highly dependent on the length of coils selected and the stopping points in each case. The Penumbra aneurysms were generally larger, and the Penumbra coils were available in longer lengths, which may have helped reduce the number of coils required to complete the cases, adding to efficiency. The fact that the Penumbra cases used 34% of the coil length while only 23% of the coil number per unit volume indicates that average coil lengths were longer in the Penumbra group. The stopping point chosen by the operator in each case also directly affects coil efficiency. The fact that the same physician operator performed all of the treatments in both arms should reduce bias, because aggressiveness of treatment is more consistent. The coiling endpoints for Penumbra resulted in significantly improved packing density with equivalent durability on follow-up.

The results imply significant coil cost savings using the Penumbra coil compared with the Orbit/Galaxy coils. Retail prices for the Penumbra coils range from $1715 for a 2 mm×1 cm coil up to$2900 for a 32 mm×60 cm coil. The Orbit family of coils range from a list price of $1295 for the 2 mm×4 cm helical coil up to$1695 for most Galaxy coils. The largest Orbit coil is 20 mm×30 cm. For the purpose of comparison, if one assumes an average retail price of $2295 for Penumbra coils and$1595 for Orbit/Galaxy, this study predicts a 67% cost savings using the Penumbra coil. While there may be price variations for these coils at individual hospitals, cost equivalence would only be achieved if the Orbit/Galaxy coil were priced at \$523 each compared with the assumed price for the Penumbra coil. Such value should be important in our current health care environment which stresses reducing hospital costs. Furthermore, this model of evaluating material costs per unit volume of aneurysm could be an intriguing way to compare treatment efficiency and cost effectiveness for other types of coils, balloon remodeling, stent assisted coiling, and flow diversion techniques in the future.

The limitations of this study include that it was retrospective, single center, and non-adjudicated. In addition, the aneurysms in the Penumbra group were larger, so there may have been more ability to treat them with longer coils. This could partially explain the improved coil efficiency. While similar durability was shown in follow-up studies, the average length of follow-up was shorter with the Penumbra coil. The fact that there were fewer ruptured aneurysms in the Penumbra group may have also added to their durability.

Coiling aneurysms efficiently implies that there should be a reduction in both procedure time and fluoroscopic time. Longer fluoroscopy times have been considered a risk factor in cerebral angiography,8 and it is reasonable to assume that this is also the case for complex neurointerventional procedures. In this retrospective analysis, the fluoroscopy and procedure times were not reliably documented, so these were not analyzed. The authors believe that documentation of fluoroscopic and procedure time for both the diagnostic angiogram and intervention is important in the assessment of procedural efficiency. These are now rigorously documented in the authors’ laboratories as an area of future investigation.

## Footnotes

• Contributors JM performed the procedures, designed the study, interpreted/analyzed the data, organized the manuscript, drafted the manuscript, and made critical revisions to the paper. ALP assisted in some procedures, acquired the data, compiled the data, assisted in drafting the manuscript, made critical revisions on content, and organized the manuscript. GV assisted on all of the cases and made critical revisions to the manuscript. RCM assisted in acquiring the data and compiling the data. Statistical interpretation was done by the hospital statistician. Images were chosen by JM and implemented in the paper by ALP. The table in the manuscript was created by JM, edited by ALP, and put into the manuscript by ALP.

• Competing interests JM gives lectures for Penumbra.

• Ethics approval The study was approved by the institutional review board.

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