RESEARCH—HUMAN—CLINICAL STUDIES RESEARCH—HUMAN—CLINICAL STUDIES
Impact of Indocyanine Green Videoangiography on Rate of Clip Adjustments Following Intraoperative Angiography Justin M. Caplan, MD* Eric Sankey, BS* Wuyang Yang, MD* Martin G. Radvany, MD*‡§ Geoffrey P. Colby, MD, PhD*‡§ Alexander L. Coon, MD*‡§ Rafael J. Tamargo, MD*¶ Judy Huang, MD* *Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland; ‡Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland; §Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Maryland; ¶Department of Otolaryngology-Head and Neck Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland Correspondence: Judy Huang, MD, Associate Professor of Neurosurgery, Johns Hopkins University School of Medicine, Johns Hopkins Hospital, 1800 Orleans St, Zayed Tower 6115F, Baltimore, MD 21287. E-mail: [email protected] Received, January 31, 2014. Accepted, June 2, 2014. Published Online, June 30, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.
BACKGROUND: Intraoperative angiography (IA) is used to evaluate the adequacy of clip reconstruction of intracranial aneurysms. Alternative imaging such as indocyanine green videoangiography (ICG-VA) has been proposed. The additional benefit of ICG-VA when IA is routinely used has not been previously determined. OBJECTIVE: To report our experience with the use of ICG-VA in combination with IA vs IA alone. METHODS: We retrospectively reviewed cases of aneurysm clipping during a 21-month period by a single surgeon in which ICG-VA was performed after clip reconstruction prior to IA, or IA alone was performed to verify optimal clipping. Records were reviewed for age, sex, race, length of stay, rupture status, Hunt and Hess grade, aneurysm size, location, and temporary clipping. Intraoperative decision making was determined for each group. RESULTS: Ninety-four patients who underwent 97 craniotomies for 128 aneurysms met inclusion criteria for this study. ICG-VA1IA was performed in 37 craniotomies; IA alone was performed for 60 craniotomies. Baseline characteristics were similar with the exception that median aneurysm size was slightly larger in the ICG-VA group (5.6 mm vs 4.3 mm, P = .04). ICG-VA produced 4 false negatives, which required clip adjustments following IA (10.8%), vs 7 patients (11.7%) in the IA-alone group requiring clip adjustments (P = .897). CONCLUSION: When IA is routinely performed, the additional use of ICG-VA does not eliminate the need for post-IA clip adjustments owing to the possibility of false negatives. When ICG-VA suggests optimal clipping, but is followed by IA, the rate of post-IA modifications in this study did not differ significantly than if ICG-VA had not been performed. KEY WORDS: Aneurysm, Indocyanine green, Intraoperative angiography, Microsurgical clipping Neurosurgery 75:437–444, 2014
DOI: 10.1227/NEU.0000000000000468
I
ntraoperative angiography (IA) is used to evaluate the adequacy of microsurgical clipping of intracranial aneurysms. Its routine use has become the standard of care at several largevolume cerebrovascular centers.1-4 However, despite the improvements in technology, safety, and efficiency, IA increases operative time and is
ABBREVIATIONS: Acomm, anterior communicating artery; CI, confidence interval; IA, intraoperative angiography; ICG-VA, indocyanine green videoangiography; JHBMC, Johns Hopkins Bayview Medical Center; JHH, Johns Hopkins Hospital; Pcomm, posterior communicating artery
NEUROSURGERY
www.neurosurgery-online.com
not without risk. As a result, several adjunctive modalities have been introduced to provide realtime feedback on the acceptability of clip reconstruction. Indocyanine green videoangiography (ICG-VA) is one such tool.5,6 Several studies have assessed the diagnostic accuracy of ICG-VA in comparison with other imaging modalities, including IA, postoperative angiography, or other postoperative imaging.7-16 In comparison with intraoperative angiography, reported rates of ICG-VA/IA discordance have ranged from 2.5% to 14.3%.6,7,15 Given these rates, ICG-VA does not appear to provide a suitable alternative to serve as a replacement of IA, but rather it has been suggested to
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 437
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
augment IA. However, no study to date has clarified if there is added benefit to IA if ICG-VA is used. At our institution, IA is routinely used in craniotomies for aneurysm clipping. Capabilities for ICG-VA are available on 1 campus, and ICG-VA is routinely used by the senior author on all craniotomies for aneurysms at that campus. In this study, we present our findings of the routine use of ICG-VA in combination with IA vs the routine use of IA alone.
METHODS Patient Characteristics We retrospectively reviewed a prospectively maintained, institutional review board-approved, aneurysm database to identify patients operated on by the lead study author (J.H.) after ICG-VA capabilities became routinely available at our institution. Aneurysm surgery is performed at 2 campuses of the Johns Hopkins Medical Institutions: the Johns Hopkins Hospital (JHH) and the Johns Hopkins Bayview Medical Center (JHBMC). We identified all patients operated on by the senior author from October 2011 to July 2013. All patients in whom IA was performed for intradural aneurysms were included. Patients who underwent craniotomy for aneurysm during the 21-month period were excluded if IA was not performed (n = 3, aneurysms were wrapped only), the aneurysm was found to be extradural (n = 2), or ICG-VA, but not IA, was performed (n = 3). Medical, imaging and operative records were reviewed. Baseline characteristics were identified, including age, sex, race, length of hospital stay, rupture status, Hunt and Hess grade, and the use of temporary clipping. In addition, aneurysm size and location were recorded.
Choice of IA Alone or With ICG-VA At both JHH and JHBMC, IA is routinely performed on all patients undergoing aneurysm clipping. Because ICG-VA was only available at JHBMC, ICG-VA was routinely performed on patients undergoing clipping at JHBMC, but was not performed on patients undergoing clipping at JHH. After aneurysm clipping at JHH, IA was performed and reviewed for satisfactory clipping. At JHBMC, after clipping, ICG-VA was first obtained. If the results of this were satisfactory, then IA was obtained. We reviewed the operative records of all cases to determine what, if any, changes were made based on the findings detected by ICG-VA or IA.
Indocyanine Green Videoangiography ICG-VA was performed at JHBMC after aneurysm clipping when the surgeon believed optimal aneurysm obliteration and vessel reconstruction had been achieved. This technique has been previously described.5,16 In brief, the area and vessels of interest were exposed in the operative field. The ICG-VA module on the microscope was activated and an intravenous bolus of 5 mL of a 2.5 mg/mL solution of ICG (Akorn, Inc, Lake Forest, Illinois) was administered by the anesthesiologist. The operating surgeon maintained the field of view, and the ICG illumination, which was recorded, was viewed on the monitor by an assistant. Once the recording was complete, it was played back, and the operating surgeon had the opportunity to view the video as well. If ICGVA revealed satisfactory clip reconstruction, then IA was performed.
Intraoperative Angiography We have previously described our techniques for our routine use of IA, which we have modified slightly.1 In brief, after the induction of general
438 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
anesthesia, a 5Fr sheath is placed in the femoral artery during the setup of the case before positioning the head. Following optimal clipping of the aneurysm, IA is performed. The surgical wound is covered with sterile drapes and the instrument tables are retracted from the field. A hole is made in the drapes over the femoral sheath, which is reprepped. A selective angiogram is then performed on the vessels of interest by using a portable C-arm digital subtraction angiography unit (ARCADIS Varic, Siemens AG, Erlangen, Germany and OEC 9900 Elite, General Electric, Fairfield, Connecticut). Following IA, the femoral sheath is covered in sterile drapes and the instrument tables are returned to the field. If IA demonstrated satisfactory clipping, the surgeon proceeded with wound closure. If IA demonstrated the need for clip adjustments, the microscope was returned to the field and, after clip adjustments, IA was performed again.
Statistical Analysis Baseline patient and aneurysm characteristics were compared by using x2 (for categorical variables) or t test (for continuous variables). Clip adjustment rates were estimated by using binomial distribution along with 95% confidence intervals. Statistical significance of clip adjustment rates between groups was compared by using x2. A P value of ,.05 was considered statistically significant. R Statistical Software (R Development Core Team, Vienna, Austria) was used for statistical analysis.
RESULTS After the application of exclusion criteria, a total of 97 craniotomies were performed on 94 patients to treat 128 aneurysms. The median age of patients at the time of surgery was 56 years. Seventy-six percent were women. The median length of hospital stay was 4 days. Unruptured aneurysms comprised 74% of the cases. The median aneurysm size was 4.7 mm. The most common aneurysm locations were middle cerebral artery (38%), anterior communicating artery (Acomm, 28%), posterior communicating artery (Pcomm, 22%) and paraophthalmic artery (18%) (Table 1). Of the 97 craniotomies performed, 38% (n = 37) patients underwent both ICG-VA and IA, whereas 62% (n = 60) patients underwent IA alone (Figure). Baseline patient demographics were compared between the 2 groups. There were no statistically significant differences between groups for age, sex, race, length of hospital stay, rupture status, Hunt and Hess grade, or the use of temporary clipping (see Table 1). Furthermore, there was no difference in aneurysm location between groups. There was a small, but statistically significant difference in median aneurysm size between groups, with the ICG-VA1IA group having slightly larger aneurysms (5.6 mm) than the IA-only group (4.3 mm) (P = .040). ICG-VA and IA There were 37 craniotomies performed that included both ICG-VA and IA, of which there were 3 instances when ICG-VA indicated the need for clip modification. All 3 cases requiring adjustments were unruptured aneurysms. In 2 cases there was persistent aneurysm filling. Additional clips were placed before the confirmation of final clip reconstruction with IA. In the third case,
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
INTRAOPERATIVE ANGIOGRAPHY AND INDOCYANINE GREEN
TABLE 1. Baseline Characteristicsa Patient Characteristics Age, y, median (range) Sex, male, n (%) Race White, n (%) Black, n (%) Asian, n (%) Hispanic, n (%) Other, n (%) Length of stay, days, median (range) Subarachnoid hemorrhage, n (%) Hunt and Hess grade (n = 25), n (%) I II III IV V Temporary clipping, n (%) Aneurysm Characteristics Median size, mm, (range) Location MCA, n (%) AComm, n (%) PComm n (%) Paraophthalmic, n (%) AChA, n (%) ICA terminus, n (%) PICA, n (%) Pericallosal, n (%) Dorsal ICA, n (%) A1, n (%)
All Craniotomies (n = 97)
ICG-VA1IA (n = 37)
IA (n = 60)
P Value
56 (18-87) 23 (23.7)
52 (18-68) 7 (18.9)
58 (25-87) 16 (26.7)
.053 .384 .932
51 34 5 5 2 4 25
(52.6) (35.0) (5.2) (5.2) (2.1) (1-67) (25.8)
20 13 1 2 1 4 11
(54.1) (35.0) (2.7) (5.4) (2.7) (1-65) (29.7)
31 21 4 3 1 4.5 14
7 2 10 1 5 48
(28.0) (8.0) (40.0) (4.0) (20.0) (49.5)
2 1 4 1 3 19
(18.2) (9.1) (36.4) (9.1) (27.3) (51.4)
5 (35.7) 1 (7.1) 6 (42.9) 0 (0.0) 2 (14.3) 29 (48.3)
(51.7) (35.1) (6.7) (5.0) (1.7) (2-67) (23.3)
.395 .484 .634
.773
All Aneurysms (n = 128)
ICG-VA1IA (n = 47)
IA (n = 81)
P Value
4.7 (1.0-16.0)
5.6 (1.0-14.0)
4.3 (1.0-16.0)
.040b .088
38 28 22 18 6 6 4 4 1 1
22 (46.8) 6 (12.8) 8 (17.0) 3 (6.4) 2 (4.3) 2 (4.3) 2 (4.3) 2 (4.3) 0 (0.0) 0 (0.0)
16 22 14 15 4 4 2 2 1 1
(29.7) (21.9) (17.2) (14.1) (4.7) (4.7) (3.1) (3.1) (0.8) (0.8)
(19.8) (27.2) (17.3) (18.5) (4.9) (4.9) (2.5) (2.5) (1.2) (1.2)
a
ICG-VA, indocyanine green video angiography; IA, intraoperative angiography; MCA, middle cerebral artery; Acomm, anterior communicating artery; Pcomm, posterior communicating artery; AChA, anterior choroidal artery; ICA, internal carotid artery; PICA, posterior inferior cerebellar artery. b Statistical significance.
ICG-VA was used following temporary clip trapping of a fusiform aneurysm to ensure distal vessel backfilling prior to the placement of permanent clips. In all 3 cases, no further changes were needed following intraoperative angiography (Table 2). In the remaining 34 cases in which ICG-VA indicated satisfactory aneurysm clipping, IA subsequently identified suboptimal reconstruction requiring clip modifications in 4 cases (Table 3). Of the cases requiring adjustments, 3 cases were unruptured, and 1 was a subarachnoid hemorrhage. In 2 cases, IA identified poor filling of adjacent vessels. One case involved residual aneurysm filling. In the fourth case, the patient underwent clipping of both Pcomm and Acomm aneurysms via the same craniotomy. In this case, the ICG-VA used to focus on the Acomm aneurysm (primary focus) failed to reveal persistent neck filling of the initially clipped Pcomm aneurysm that was believed to be straightforward. Overall, there was a postIA clip adjustment rate of 10.8% (Table 3).
NEUROSURGERY
IA Only There were 60 craniotomies performed that included IA alone. Following IA, there were 7 instances in which changes were made based on the IA findings (11.7%). Of the cases requiring adjustments, 5 cases were unruptured, and 2 had subarachnoid hemorrhages. This included persistent aneurysm filling in 4 cases, 1 case of persistent aneurysm neck filling, 1 case of a previously unidentified aneurysm lobule, and 1 case of poor filling of a fetal posterior cerebral artery (Table 4). ICG-VA1IA vs IA Alone When comparing the post-IA clip adjustment rates of all patients, there was no statistically significant difference between ICG-VA1IA (10.8% [95% confidence interval (CI), 3%-25%]) vs IA alone (11.7% [95% CI, 5%-23%]) (P = .897). If the patients who had immediate post-ICG-VA adjustments are
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 439
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
FIGURE. Distribution of aneurysm cases by intraoperative imaging. IA, intraoperative angiography; ICGVA, indocyanine green videoangiography.
excluded, the post-IA adjustment rate increases to 11.8% (95% CI, 3%-28%) (4 patients of 34), which did not reach statistical significance compared with IA alone (11.7% [95% CI, 5%23%]) (P = .999). Finally, if one assumes that the 3 patients who required adjustments after ICG-VA would have otherwise required changes after IA, and thus include them in that group (ie, rate of cases which need any postimaging adjustment), this would increase the postclip adjustment rate after IA to 18.9% (95% CI, 8%-35%) (7 of 37 patients). There is no statistical
difference between this rate and the rate of adjustments in the IAalone group (P = .324).
DISCUSSION Intraoperative Angiography Intraoperative angiography is the gold standard for the immediate assessment of microsurgical clipping of intracranial aneurysms, although it remains to be performed routinely in all
TABLE 2. ICG-VA Positive, Change Made, IA Negative (Presumed True Positives)a
TABLE 3. ICG-VA Negative, IA Positive (False Negatives)a
Patient No.
Aneurysm Aneurysm Reason for Clip Location Size, mm Adjustment
Patient No.
MCA
1
Unruptured
Acomm
2
Unruptured
PICA
3
Unruptured
MCA
4
SAH
1
Rupture Status Unruptured
2
Unruptured
3
Unruptured
ICA terminus MCA MCA MCA
a
9.4
5
2 13
Persistent aneurysm filling Persistent aneurysm filling Fusiform aneurysm, ICG-VA used to confirm adequacy of collaterals before permanent clipping
ICG-VA, indocyanine green video angiography; IA, intraoperative angiography; ICA, internal carotid artery; MCA, middle cerebral artery.
440 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
Rupture Status
Aneurysm Aneurysm Reason for Clip Location Size, mm Adjustment 12 4.2 14
Acomm Acomm
2.2 10
Pcomm
8.3
Poor A1-A2 inflow Poor vertebral artery filling Residual filling at base of Acomm aneurysm Residual neck of Pcomm aneurysmb
a
ICG-VA, indocyanine green video angiography; IA, intraoperative angiography; MCA, middle cerebral artery. Acomm, anterior communicating artery; Pcomm, posterior communicating artery; PICA, posterior inferior cerebellar artery; SAH, subarachnoid hemorrhage. b PComm was not visualized with the ICG as field of view was on the AComm aneurysm.
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
INTRAOPERATIVE ANGIOGRAPHY AND INDOCYANINE GREEN
TABLE 4. IA Only, Positive Angiograma Patient No.
Rupture Status
Aneurysm Location
Aneurysm Reason for Clip Size, mm Adjustment
1
Unruptured Paraophthalmic
6
Persistent aneurysm neck
2
Unruptured
MCA
8
3
SAH
Pcomm
6.8
Persistent aneurysm filling Persistent aneurysm filling
4
Unruptured
Acomm
7.4
5
SAH
Pcomm
6
Unruptured
PICA
5.6
7
Unruptured
Acomm
4.5
ICA blister
15
Persistent aneurysm filling Poor filling of fetal PCA Persistent aneurysm filling Residual filling of previously unidentified aneurysm lobule
1
a
IA, intraoperative angiography; ICA, internal carotid artery; MCA, middle cerebral artery. Acomm, anterior communicating artery; PCA, posterior cerebral artery; Pcomm, posterior communicating artery; PICA, posterior inferior cerebellar artery; SAH, subarachnoid hemorrhage.
centers. The technique, safety profile, and efficacy have been well described in the literature with clip readjustment rates reported in the range of 7% to 12.4% in large series.1-4,17 Intraoperative angiography is safe in the setting of its routine use at high-volume centers.1,3,4 Although the routine use of intraoperative angiography is not universally accepted, some have argued that, to make it safe and efficient in cases where it is most needed, it should be used routinely in all cases.18 In this study, the rate of clip adjustments in IA alone was 11.7%, consistent with previous reports. Despite the speed at which IA can be performed when done routinely, we continue to use intraoperative somatosensory and electroencephalographic monitoring as additional adjuncts to detect for signs of ischemia necessitating clip adjustment prior to IA.19 Intraoperative ICG-VA Compared With Postoperative Imaging In contrast to the relatively few studies comparing ICG-VA to IA, there have been a number of studies that compare ICG-VA to postoperative imaging modalities. The major limitation of assessing the adequacy of ICG-VA with postoperative imaging is the inability to implement adjustments as soon as they are needed. The ICG-VA to postoperative imaging discordance rates vary from 0% to 12%.8-14,20
NEUROSURGERY
Intraoperative ICG-VA in Combination With IA The use of ICG-VA was first reported for use in intracranial aneurysm surgery in 2003 with the use of a digital camcorder and was subsequently modified for integration into the operative microscope.5,16 Subsequently, a few studies have assessed the accuracy of ICG-VA compared with IA. Raabe et al6 report a 10% disagreement rate between ICG-VA and IA in a series of 60 aneurysms. Gruber et al15 report a 2.5% ICG-VA and IA rate of disagreement in 123 aneurysms. However, this study also utilized other adjunctive modalities such as microvascular Doppler and intraoperative neuroendoscopy, which may have helped to decrease the discordance rate. More recently, Washington et al7 reported a discordance rate of 14.3% in 49 patients undergoing aneurysm surgery. In our study, the rate of disagreement between IA and ICG-VA was 10.8%, similar to the above studies. The rate of post-IA clip adjustment in cases using ICG-VA and IA is likely lower than it would have been if ICG-VA had not been used. In our series, there were 3 cases in which ICG-VA led to a clip adjustment that was then confirmed with a satisfactory intraoperative angiogram. Assuming that the ICG-VA finding was not a false positive (ie, ICG-VA suggests residual aneurysm filling, when IA would not have), these findings would have presumably been discovered on IA and represent 3 more aneurysms that would have required post-IA adjustments, which would have raised the rate from 10.8% to 18.9%. However, in our study, this increased rate did not reach statistical significance in comparison with the adjustment rate of 11.7% in the IA-alone group. Suboptimal Clipping Theory One difficulty in understanding the benefit of ICG-VA is that it can be difficult to know the true value of ICG-VA, because knowledge that IA will follow ICG-VA may impact how a surgeon performs the initial clip reconstruction.21 In such circumstances, the surgeon may, consciously or subconsciously, pursue a less aggressive clipping strategy with the anticipation of imaging feedback from the relatively quick and low-risk ICG-VA. The surgeon would then have the opportunity to make adjustments based on the ICG-VA before performing the more timeconsuming and invasive IA. If this “suboptimal clipping theory” is applied, an elevated rate of inadequate clippings detected with ICG-VA that could then be corrected prior to IA would be anticipated. This would lead to a higher rate of ICG-VA to IA disagreement, and thus suggest that ICG-VA may be less valuable when compared with IA.21 A similar argument has been put forth when assessing the utility of IA alone.18 It is true that, when determining ICG-VA/IA discordance rates, the value of ICG-VA may in fact be undervalued because of this principle. In our study, however, rather than examining ICG-VA/IA discordance rates, we focus on the rate of clip adjustments following IA (whether or not ICG-VA was performed) in order to determine the added benefit of ICG-VA in addition to IA. Here, knowledge that IA will follow ICG-VA may overvalue the role of ICG-VA as follows. If the surgeon initially performs a suboptimal clipping knowing
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 441
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
that ICG-VA is available, then the mistake can be recognized following ICG-VA and corrected prior to IA. As such, the number of post-IA adjustments will decrease. This is highlighted in our study by the 3 patients who had “true positive” ICG-VA results leading to clip adjustments following ICG-VA and prior to IA. If ICG-VA had not been performed for some reason, but the surgeon had performed the same “suboptimal” clipping techniques, then presumably these suboptimal clippings would have been seen on subsequent IA, and thus these 3 patients, in addition to the 4 others who required post-IA clip adjustments, would have resulted in a post-IA clip adjustment rate of 18.9% (7 of 37 patients). In this cohort, the actual post-IA clip adjustment rate was just 10.8%. Therefore, it can be hypothesized that the use of ICGVA decreased the need for clip adjustments by 8.1%. However, under the “suboptimal clipping theory,” if ICG-VA were truly not available, and thus the surgeon was forced to give his/her “best effort,” it could be argued that the surgeon might have been more selective in the initial clip placement, and thus would have obtained an overall decreased rate of clip adjustments of less than 18.9%. Thus, under the “suboptimal clipping theory” when comparing rates of post-IA clip adjustments (rather than ICG-VA/ IA discordance rates), the knowledge that IA will follow ICG-VA likely leads to overvaluing the role of ICG-VA. Interestingly, in our study, even if the “suboptimal clipping theory” were true and the value of ICG-VA is overestimated, it still did not lead to a statistically significant difference in rates of post-IA clip adjustments between cohorts (18.9% vs 11.7%, P = .324). Limitations and Benefits of ICG-VA Many have suggested that ICG-VA and IA be used as complementary modalities rather than mutually exclusive methods. We agree that ICG-VA should not replace IA for the intraoperative assessment of aneurysm clipping. However, our data suggest when IA is routinely performed, there is little benefit in the use of ICG-VA to decrease post-IA clip adjustment rates. Our study did not address several situations where the use of ICGVA may be beneficial, including bypasses, arteriovenous malformation resections, and dural arteriovenous fistula obliterations. Furthermore, ICG-VA may be a useful adjunct in aneurysm surgery when operating on multiple aneurysms via a single craniotomy and the surgeon wishes to have a “first pass” look at the initial vessel reconstruction following clipping of the first aneurysm, but to defer IA until all aneurysms are clipped to minimize patient risk. In addition, ICG-VA is significantly faster than IA, and may be particularly useful in instances where rapid assessment and reassurance of relevant vessel patency is desired before formal IA, with less risk of prolonged ischemic time. Furthermore, ICG-VA may be a useful tool in patients for whom IA may be associated with increased risk (ie, contrast allergy or history of previous stroke after angiography). In our study, we excluded 3 patients in whom ICG-VA was performed, but not IA. One patient had a previous stroke from an angiogram. Two patients were deemed too critically ill to tolerate the time needed for an IA.
442 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
Our study assessed the overall rate of clip adjustments required when ICG-VA is added to the routine use of IA. However, there may be specific examples when ICG-VA is useful in aneurysm surgery. For example, de Oliveira et al22 examined the use of ICGVA for assessing perforating arteries not visible on preoperative DSA during intracranial aneurysm surgery, although this too had a high false-negative rate. ICG-VA is superior to IA in evaluating the patency of the perforating arteries in the operative field and is likely superior in identifying the critical vessels of small caliber such as the anterior choroidal artery. Regardless of the method used to evaluate adequate clipping, judicious identification of the perforating arteries remains critical as a central tenet of aneurysm surgery, because compromise of these vessels in the operative field will only be appreciated with ICG-VA if they were first recognized during the microsurgical dissection. Other examples of advances in the use of ICG-VA include combining it with Doppler and endoscopy.23-25 This exciting new application of ICG-VA allows direct visualization of vasculature not otherwise readily seen on microscope-integrated ICG-VA. Another elegant method to visualize hidden arteries during ICGVA is the use of a Yasargil mirror.26 In a small series of patients, Schnell et al27 examined the feasibility of ICG-VA combined with intraoperative computed tomography angiography/perfusion as complementary methods. Furthermore, ICG-VA uniquely allows the surgeon to manipulate the clipped aneurysm while simultaneously assessing vessel and aneurysm patency in a manner that is not possible with IA. Intraoperative angiography is limited to the assessment of the vessels injected, whereas systemically injected ICG-VA will provide for fluorescence of the entire cerebral vasculature. This may become particularly advantageous when operating on midline aneurysms associated with the Acomm. A single ICG injection may allow for complete visualization of the communicating complex, whereas complete assessment by using IA could require bilateral carotid injections, depending on the patency of the Acomm. We did not perform preclipping ICG-VA in this study, although used by others, it is possible preclipping ICG-VA could improve its utility by providing a “control” image prior to aneurysm clipping. Given the overall tendency in our study for larger aneurysms to require clip adjustments regardless of imaging modality, ICG-VA may be useful in larger aneurysms that are typically more difficult to reconstruct and require a more extensive dissection. Another potential improvement in ICGVA is the quantitative rather than qualitative analysis of ICG-VA, which is reported in multiple studies.28-30 This approach offers a numerical analysis of the ICG-VA imaging rather than relying on the surgeon’s interpretation of the video images. However, 1 major limitation continues to be that only the microscope images directly included in the operative view can be analyzed. As such, it is difficult to view aneurysms in multiple locations. Intraoperative angiography at present is superior to ICG-VA for this indication. In our study, residual filling of 1 Pcomm aneurysm neck (Table 3, patient 4) was not identified by ICG-VA, because the field of view was focused on a different location (Acomm region for a second
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
INTRAOPERATIVE ANGIOGRAPHY AND INDOCYANINE GREEN
aneurysm). Although ICG-VA would likely have discovered this filling if the field of view were focused on this region, the inability to view multiple fields of view simultaneously presents a limitation of ICG-VA that is not encountered with IA.
CONCLUSION Intraoperative angiography is the gold standard for the intraoperative assessment of clipping of intracranial aneurysms. The use of ICG-VA has not replaced the need for IA. Despite the expectation that ICG-VA might decrease the need for clip adjustments following IA, our study fails to demonstrate a statistically significant decrease in clip adjustment rates with ICG-VA when IA is routinely performed. Given the complexity of interpreting the results in which the presence of 1 diagnostic test may influence the outcome of a subsequent test, a randomized study should be performed to determine the true utility of this exciting emerging imaging modality in comparison with intraoperative angiography alone. ICG-VA continues to have a valuable role in cerebrovascular microsurgery, but does not appear, in our study, to provide any significant decrease in post-IA clip adjustment rate when IA is routinely performed for all aneurysm surgery. As with any tool, it is essential to recognize the strengths and limitations of each in order to tailor its application to the individual and optimize patient outcomes. Further studies are needed to identify the specific instances when ICG-VA is most useful for aneurysm surgery. Disclosure Dr Radvany has received honoraria from Siemens Medical; Dr Colby, from Covidien, Microvention; and Dr Coon, from Covidien. The other authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES 1. Chiang VL, Gailloud P, Murphy KJ, Rigamonti D, Tamargo RJ. Routine intraoperative angiography during aneurysm surgery. J Neurosurg. 2002;96(6):988-992. 2. Klopfenstein JD, Spetzler RF, Kim LJ, et al. Comparison of routine and selective use of intraoperative angiography during aneurysm surgery: a prospective assessment. J Neurosurg. 2004;100(2):230-235. 3. Chalouhi N, Theofanis T, Jabbour P, et al. Safety and efficacy of intraoperative angiography in craniotomies for cerebral aneurysms and arteriovenous malformations: a review of 1093 consecutive cases. Neurosurgery. 2012;71(6):1162-1169. 4. Tang G, Cawley CM, Dion JE, Barrow DL. Intraoperative angiography during aneurysm surgery: a prospective evaluation of efficacy. J Neurosurg. 2002;96(6):993-999. 5. Raabe A, Beck J, Gerlach R, Zimmermann M, Seifert V. Near-infrared indocyanine green video angiography: a new method for intraoperative assessment of vascular flow. Neurosurgery. 2003;52(1):132-139; discussion 139. 6. Raabe A, Nakaji P, Beck J, et al. Prospective evaluation of surgical microscopeintegrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. J Neurosurg. 2005;103(6):982-989. 7. Washington CW, Zipfel GJ, Chicoine MR, et al. Comparing indocyanine green videoangiography to the gold standard of intraoperative digital subtraction angiography used in aneurysm surgery. J Neurosurg. 2013;118(2):420-427. 8. Jing Z, Ou S, Ban Y, Tong Z, Wang Y. Intraoperative assessment of anterior circulation aneurysms using the indocyanine green video angiography technique. J Clin Neurosci. 2010;17(1):26-28. 9. Ma CY, Shi JX, Wang HD, Hang CH, Cheng HL, Wu W. Intraoperative indocyanine green angiography in intracranial aneurysm surgery: microsurgical clipping and revascularization. Clin Neurol Neurosurg. 2009;111(10):840-846.
NEUROSURGERY
10. Li J, Lan Z, He M, You C. Assessment of microscope-integrated indocyanine green angiography during intracranial aneurysm surgery: a retrospective study of 120 patients. Neurol India. 2009;57(4):453-459. 11. Lin J, Zhao J, Zhao Y, et al. Multiple intraoperative monitoring-assisted microneurosurgical treatment for anterior circulation cerebral aneurysm. J Int Med Res. 2011;39(3):891-903. 12. Khurana VG, Seow K, Duke D. Intuitiveness, quality and utility of intraoperative fluorescence videoangiography: Australian Neurosurgical Experience. Br J Neurosurg. 2010;24(2):163-172. 13. Wang S, Liu L, Zhao Y, Zhang D, Yang M, Zhao J. Evaluation of surgical microscope-integrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. Neurosurg Rev. 2011;34(2):209-215. 14. Moon HS, Joo SP, Seo BR, Jang JW, Kim JH, Kim TS. Value of indocyanine green videoangiography in deciding the completeness of cerebrovascular surgery. J Korean Neurosurg Soc. 2013;53(6):349-355. 15. Gruber A, Dorfer C, Standhardt H, Bavinzski G, Knosp E. Prospective comparison of intraoperative vascular monitoring technologies during cerebral aneurysm surgery. Neurosurgery. 2011;68(3):657-673; discussion 673. 16. Raabe A, Beck J, Seifert V. Technique and image quality of intraoperative indocyanine green angiography during aneurysm surgery using surgical microscope integrated nearinfrared video technology. Zentralbl Neurochir. 2005;66(1):1-6; discussion 7-8. 17. Martin NA, Bentson J, Viñuela F, et al. Intraoperative digital subtraction angiography and the surgical treatment of intracranial aneurysms and vascular malformations. J Neurosurg. 1990;73(4):526-533. 18. Heros RC. Intraoperative angiography. J Neurosurg. 2002;96(6):979-980; discussion 980. 19. Wicks RT, Pradilla G, Raza SM, et al. Impact of changes in intraoperative somatosensory evoked potentials on stroke rates after clipping of intracranial aneurysms. Neurosurgery. 2012;70(5):1114-1124. 20. Dashti R, Laakso A, Niemelä M, Porras M, Hernesniemi J. Microscope-integrated near-infrared indocyanine green videoangiography during surgery of intracranial aneurysms: the Helsinki experience. Surg Neurol. 2009;71(5):543-550; discussion 550. 21. Morcos JJ. Editorial: indocyanine green videoangiography or intraoperative angiography? J Neurosurg. 2013;118(2):417-418; discussion 418-419. 22. de Oliveira JG, Beck J, Seifert V, Teixeira MJ, Raabe A. Assessment of flow in perforating arteries during intracranial aneurysm surgery using intraoperative nearinfrared indocyanine green videoangiography. Neurosurgery. 2007;61(3 suppl): 63-72; discussion 72-73. 23. Imizu S, Kato Y, Sangli A, Oguri D, Sano H. Assessment of incomplete clipping of aneurysms intraoperatively by a near-infrared indocyanine green-video angiography (Niicg-Va) integrated microscope. Minim Invasive Neurosurg. 2008;51:199-203. 24. Bruneau M, Appelboom G, Rynkowski M, Van Cutsem N, Mine B, De Witte O. Endoscope-integrated ICG technology: first application during intracranial aneurysm surgery. Neurosurg Rev. 2013;36(1):77-84; discussion 84-85. 25. Nishiyama Y, Kinouchi H, Senbokuya N, et al. Endoscopic indocyanine green video angiography in aneurysm surgery: an innovative method for intraoperative assessment of blood flow in vasculature hidden from microscopic view. J Neurosurg. 2012;117(2):302-308. 26. Wilson J, Screven R, Volk J, Payner T. Use of a Yas¸argil mirror as an adjunct to indocyanine green angiography to evaluate the patency of elusive posterior communicating arteries during aneurysm clipping: case report. Neurosurgery. 2012; 71(1 suppl operative):195-197. 27. Schnell O, Morhard D, Holtmannspötter M, Reiser M, Tonn JC, Schichor C. Nearinfrared indocyanine green videoangiography (ICGVA) and intraoperative computed tomography (iCT): are they complementary or competitive imaging techniques in aneurysm surgery? Acta Neurochir (Wien). 2012;154(10):1861-1868. 28. Son YJ, Kim JE, Park SB, Lee SH, Chung YS, Yang HJ. Quantitative analysis of intraoperative indocyanine green video angiography in aneurysm surgery. J Cerebrovasc Endovasc Neurosurg. 2013;15(2):76-84. 29. Oda J, Kato Y, Chen SF, et al. Intraoperative near-infrared indocyanine greenvideoangiography (ICG-VA) and graphic analysis of fluorescence intensity in cerebral aneurysm surgery. J Clin Neurosci. 2011;18(8):1097-1100. 30. Chen SF, Kato Y, Oda J, et al. The application of intraoperative near-infrared indocyanine green videoangiography and analysis of fluorescence intensity in cerebrovascular surgery. Surg Neurol Int. 2011;2(1):42.
Acknowledgment The authors thank Xiaobu Ye, MD, for her assistance with the statistical analysis.
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 443
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
COMMENTS
T
he authors report a retrospective single-surgeon series of 94 patients who underwent 97 craniotomies for 128 aneurysms at 2 campuses of the same center over a 21-month period. In 37 of the 97 craniotomies considered, clip placement was assessed by using both indocyanine green angiography (ICGA) and conventional intraoperative digital subtraction angiography (DSA), whereas in the remaining 60 craniotomies DSA was the only technique available. The authors compare clip repositioning rates between the 2 groups and find an insignificant difference of 10.8% (4/37) for the ICGA1DSA group vs 11.7% (7/60) for the DSA-only group (P = .89). In the first group, ICGA was performed after aneurysm clipping and indicated the need for clip repositioning in 3/37 cases (8.1%). Thereafter, DSA was performed and disclosed the need for clip modification in another 4/37 cases (10.8%). Therefore, in a total of 7/37 cases (18.9%) aneurysm misclipping was discovered by using ICGA and DSA in conjunction. In the second group, DSA was used alone and disclosed a misclipping rate of 11.7% (7/60). The key information of this article, however, is not that ICGA does not provide a significant benefit when used in conjunction with conventional DSA in decreasing clip repositioning rates compared with DSA alone (10.8% vs 11.7%), but that within the group of patients receiving both ICGA and DSA, ICGA resulted in a 43% relative risk reduction for aneurysm misclipping (18.9% vs 10.8%). Engelbert Knosp Andreas Gruber Vienna, Austria The study compared the rate of false-negative results after aneurysm clipping of 97 patients. In 37 craniotomies ICG-VA and IA was carried out and in other 60 craniotomies only IA was performed. In 8% of the craniotomies the ICG-VA detected a residual aneurysm filling or vessel occlusion, which could be solved immediately. A subsequently
444 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
performed IA showed a vessel stenosis in two cases, a residual aneurysm filling in one case and also a residual aneurysm filling of an aneurysm, which was not focused in ICG-VA. This leads to a rate of 8%-11%, depending on whether 3 or 4 cases were calculated as false-negative. In the “IA-only” group the clip position was adjusted in 12% of the cases. These findings are similar to our own prospective results.1 Unfortunately, complications were not presented in detail. Considering the self-cited publication about IA at JHBMC,2 2.6% of the IA had complications (including stroke; vs 0.1% complication rate for ICG-VA1). It might be possible that some of these complications can be reduced using ICG-VA routinely (time required for ICG-VA 1-2 min vs 15-35 min for IA). This publication also detected a false-negative rate of 8% for IA proven by postoperative angiography; and these false-negative patients subsequently rebled. Altogether, nowadays every clipped aneurysm should be controlled intraoperatively, either by (several) ICG-VA or by IA. Also a postoperative 3D angiography should be performed in any patient to identify falsenegative cases (with ICG-VA and IA approx. 10%) and if necessary to treat them. Despite the advantages and disadvantages of ICG-VA or IA, a local standard and setting should be created, ensuring the safest method and offering the most comfortable possibility to the surgeon to check the complete occlusion of the aneurysm during the operation. Volker Seifert Frankfurt, Germany
1. Raabe A, Nakaji P, Beck J, et al. Prospective evaluation of surgical microscopeintegrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. J Neurosurg. 2005;103(6):982-989. 2. Chiang VL, Gailloud P, Murphy KJ, Rigamonti D, Tamargo RJ. Routine intraoperative angiography during aneurysm surgery. J Neurosurg. 2009;96(6): 988-992.
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
Impact of Indocyanine Green Videoangiography on Rate of Clip Adjustments Following Intraoperative Angiography Justin M. Caplan, MD* Eric Sankey, BS* Wuyang Yang, MD* Martin G. Radvany, MD*‡§ Geoffrey P. Colby, MD, PhD*‡§ Alexander L. Coon, MD*‡§ Rafael J. Tamargo, MD*¶ Judy Huang, MD* *Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland; ‡Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland; §Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Maryland; ¶Department of Otolaryngology-Head and Neck Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland Correspondence: Judy Huang, MD, Associate Professor of Neurosurgery, Johns Hopkins University School of Medicine, Johns Hopkins Hospital, 1800 Orleans St, Zayed Tower 6115F, Baltimore, MD 21287. E-mail: [email protected] Received, January 31, 2014. Accepted, June 2, 2014. Published Online, June 30, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.
BACKGROUND: Intraoperative angiography (IA) is used to evaluate the adequacy of clip reconstruction of intracranial aneurysms. Alternative imaging such as indocyanine green videoangiography (ICG-VA) has been proposed. The additional benefit of ICG-VA when IA is routinely used has not been previously determined. OBJECTIVE: To report our experience with the use of ICG-VA in combination with IA vs IA alone. METHODS: We retrospectively reviewed cases of aneurysm clipping during a 21-month period by a single surgeon in which ICG-VA was performed after clip reconstruction prior to IA, or IA alone was performed to verify optimal clipping. Records were reviewed for age, sex, race, length of stay, rupture status, Hunt and Hess grade, aneurysm size, location, and temporary clipping. Intraoperative decision making was determined for each group. RESULTS: Ninety-four patients who underwent 97 craniotomies for 128 aneurysms met inclusion criteria for this study. ICG-VA1IA was performed in 37 craniotomies; IA alone was performed for 60 craniotomies. Baseline characteristics were similar with the exception that median aneurysm size was slightly larger in the ICG-VA group (5.6 mm vs 4.3 mm, P = .04). ICG-VA produced 4 false negatives, which required clip adjustments following IA (10.8%), vs 7 patients (11.7%) in the IA-alone group requiring clip adjustments (P = .897). CONCLUSION: When IA is routinely performed, the additional use of ICG-VA does not eliminate the need for post-IA clip adjustments owing to the possibility of false negatives. When ICG-VA suggests optimal clipping, but is followed by IA, the rate of post-IA modifications in this study did not differ significantly than if ICG-VA had not been performed. KEY WORDS: Aneurysm, Indocyanine green, Intraoperative angiography, Microsurgical clipping Neurosurgery 75:437–444, 2014
DOI: 10.1227/NEU.0000000000000468
I
ntraoperative angiography (IA) is used to evaluate the adequacy of microsurgical clipping of intracranial aneurysms. Its routine use has become the standard of care at several largevolume cerebrovascular centers.1-4 However, despite the improvements in technology, safety, and efficiency, IA increases operative time and is
ABBREVIATIONS: Acomm, anterior communicating artery; CI, confidence interval; IA, intraoperative angiography; ICG-VA, indocyanine green videoangiography; JHBMC, Johns Hopkins Bayview Medical Center; JHH, Johns Hopkins Hospital; Pcomm, posterior communicating artery
NEUROSURGERY
www.neurosurgery-online.com
not without risk. As a result, several adjunctive modalities have been introduced to provide realtime feedback on the acceptability of clip reconstruction. Indocyanine green videoangiography (ICG-VA) is one such tool.5,6 Several studies have assessed the diagnostic accuracy of ICG-VA in comparison with other imaging modalities, including IA, postoperative angiography, or other postoperative imaging.7-16 In comparison with intraoperative angiography, reported rates of ICG-VA/IA discordance have ranged from 2.5% to 14.3%.6,7,15 Given these rates, ICG-VA does not appear to provide a suitable alternative to serve as a replacement of IA, but rather it has been suggested to
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 437
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
augment IA. However, no study to date has clarified if there is added benefit to IA if ICG-VA is used. At our institution, IA is routinely used in craniotomies for aneurysm clipping. Capabilities for ICG-VA are available on 1 campus, and ICG-VA is routinely used by the senior author on all craniotomies for aneurysms at that campus. In this study, we present our findings of the routine use of ICG-VA in combination with IA vs the routine use of IA alone.
METHODS Patient Characteristics We retrospectively reviewed a prospectively maintained, institutional review board-approved, aneurysm database to identify patients operated on by the lead study author (J.H.) after ICG-VA capabilities became routinely available at our institution. Aneurysm surgery is performed at 2 campuses of the Johns Hopkins Medical Institutions: the Johns Hopkins Hospital (JHH) and the Johns Hopkins Bayview Medical Center (JHBMC). We identified all patients operated on by the senior author from October 2011 to July 2013. All patients in whom IA was performed for intradural aneurysms were included. Patients who underwent craniotomy for aneurysm during the 21-month period were excluded if IA was not performed (n = 3, aneurysms were wrapped only), the aneurysm was found to be extradural (n = 2), or ICG-VA, but not IA, was performed (n = 3). Medical, imaging and operative records were reviewed. Baseline characteristics were identified, including age, sex, race, length of hospital stay, rupture status, Hunt and Hess grade, and the use of temporary clipping. In addition, aneurysm size and location were recorded.
Choice of IA Alone or With ICG-VA At both JHH and JHBMC, IA is routinely performed on all patients undergoing aneurysm clipping. Because ICG-VA was only available at JHBMC, ICG-VA was routinely performed on patients undergoing clipping at JHBMC, but was not performed on patients undergoing clipping at JHH. After aneurysm clipping at JHH, IA was performed and reviewed for satisfactory clipping. At JHBMC, after clipping, ICG-VA was first obtained. If the results of this were satisfactory, then IA was obtained. We reviewed the operative records of all cases to determine what, if any, changes were made based on the findings detected by ICG-VA or IA.
Indocyanine Green Videoangiography ICG-VA was performed at JHBMC after aneurysm clipping when the surgeon believed optimal aneurysm obliteration and vessel reconstruction had been achieved. This technique has been previously described.5,16 In brief, the area and vessels of interest were exposed in the operative field. The ICG-VA module on the microscope was activated and an intravenous bolus of 5 mL of a 2.5 mg/mL solution of ICG (Akorn, Inc, Lake Forest, Illinois) was administered by the anesthesiologist. The operating surgeon maintained the field of view, and the ICG illumination, which was recorded, was viewed on the monitor by an assistant. Once the recording was complete, it was played back, and the operating surgeon had the opportunity to view the video as well. If ICGVA revealed satisfactory clip reconstruction, then IA was performed.
Intraoperative Angiography We have previously described our techniques for our routine use of IA, which we have modified slightly.1 In brief, after the induction of general
438 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
anesthesia, a 5Fr sheath is placed in the femoral artery during the setup of the case before positioning the head. Following optimal clipping of the aneurysm, IA is performed. The surgical wound is covered with sterile drapes and the instrument tables are retracted from the field. A hole is made in the drapes over the femoral sheath, which is reprepped. A selective angiogram is then performed on the vessels of interest by using a portable C-arm digital subtraction angiography unit (ARCADIS Varic, Siemens AG, Erlangen, Germany and OEC 9900 Elite, General Electric, Fairfield, Connecticut). Following IA, the femoral sheath is covered in sterile drapes and the instrument tables are returned to the field. If IA demonstrated satisfactory clipping, the surgeon proceeded with wound closure. If IA demonstrated the need for clip adjustments, the microscope was returned to the field and, after clip adjustments, IA was performed again.
Statistical Analysis Baseline patient and aneurysm characteristics were compared by using x2 (for categorical variables) or t test (for continuous variables). Clip adjustment rates were estimated by using binomial distribution along with 95% confidence intervals. Statistical significance of clip adjustment rates between groups was compared by using x2. A P value of ,.05 was considered statistically significant. R Statistical Software (R Development Core Team, Vienna, Austria) was used for statistical analysis.
RESULTS After the application of exclusion criteria, a total of 97 craniotomies were performed on 94 patients to treat 128 aneurysms. The median age of patients at the time of surgery was 56 years. Seventy-six percent were women. The median length of hospital stay was 4 days. Unruptured aneurysms comprised 74% of the cases. The median aneurysm size was 4.7 mm. The most common aneurysm locations were middle cerebral artery (38%), anterior communicating artery (Acomm, 28%), posterior communicating artery (Pcomm, 22%) and paraophthalmic artery (18%) (Table 1). Of the 97 craniotomies performed, 38% (n = 37) patients underwent both ICG-VA and IA, whereas 62% (n = 60) patients underwent IA alone (Figure). Baseline patient demographics were compared between the 2 groups. There were no statistically significant differences between groups for age, sex, race, length of hospital stay, rupture status, Hunt and Hess grade, or the use of temporary clipping (see Table 1). Furthermore, there was no difference in aneurysm location between groups. There was a small, but statistically significant difference in median aneurysm size between groups, with the ICG-VA1IA group having slightly larger aneurysms (5.6 mm) than the IA-only group (4.3 mm) (P = .040). ICG-VA and IA There were 37 craniotomies performed that included both ICG-VA and IA, of which there were 3 instances when ICG-VA indicated the need for clip modification. All 3 cases requiring adjustments were unruptured aneurysms. In 2 cases there was persistent aneurysm filling. Additional clips were placed before the confirmation of final clip reconstruction with IA. In the third case,
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
INTRAOPERATIVE ANGIOGRAPHY AND INDOCYANINE GREEN
TABLE 1. Baseline Characteristicsa Patient Characteristics Age, y, median (range) Sex, male, n (%) Race White, n (%) Black, n (%) Asian, n (%) Hispanic, n (%) Other, n (%) Length of stay, days, median (range) Subarachnoid hemorrhage, n (%) Hunt and Hess grade (n = 25), n (%) I II III IV V Temporary clipping, n (%) Aneurysm Characteristics Median size, mm, (range) Location MCA, n (%) AComm, n (%) PComm n (%) Paraophthalmic, n (%) AChA, n (%) ICA terminus, n (%) PICA, n (%) Pericallosal, n (%) Dorsal ICA, n (%) A1, n (%)
All Craniotomies (n = 97)
ICG-VA1IA (n = 37)
IA (n = 60)
P Value
56 (18-87) 23 (23.7)
52 (18-68) 7 (18.9)
58 (25-87) 16 (26.7)
.053 .384 .932
51 34 5 5 2 4 25
(52.6) (35.0) (5.2) (5.2) (2.1) (1-67) (25.8)
20 13 1 2 1 4 11
(54.1) (35.0) (2.7) (5.4) (2.7) (1-65) (29.7)
31 21 4 3 1 4.5 14
7 2 10 1 5 48
(28.0) (8.0) (40.0) (4.0) (20.0) (49.5)
2 1 4 1 3 19
(18.2) (9.1) (36.4) (9.1) (27.3) (51.4)
5 (35.7) 1 (7.1) 6 (42.9) 0 (0.0) 2 (14.3) 29 (48.3)
(51.7) (35.1) (6.7) (5.0) (1.7) (2-67) (23.3)
.395 .484 .634
.773
All Aneurysms (n = 128)
ICG-VA1IA (n = 47)
IA (n = 81)
P Value
4.7 (1.0-16.0)
5.6 (1.0-14.0)
4.3 (1.0-16.0)
.040b .088
38 28 22 18 6 6 4 4 1 1
22 (46.8) 6 (12.8) 8 (17.0) 3 (6.4) 2 (4.3) 2 (4.3) 2 (4.3) 2 (4.3) 0 (0.0) 0 (0.0)
16 22 14 15 4 4 2 2 1 1
(29.7) (21.9) (17.2) (14.1) (4.7) (4.7) (3.1) (3.1) (0.8) (0.8)
(19.8) (27.2) (17.3) (18.5) (4.9) (4.9) (2.5) (2.5) (1.2) (1.2)
a
ICG-VA, indocyanine green video angiography; IA, intraoperative angiography; MCA, middle cerebral artery; Acomm, anterior communicating artery; Pcomm, posterior communicating artery; AChA, anterior choroidal artery; ICA, internal carotid artery; PICA, posterior inferior cerebellar artery. b Statistical significance.
ICG-VA was used following temporary clip trapping of a fusiform aneurysm to ensure distal vessel backfilling prior to the placement of permanent clips. In all 3 cases, no further changes were needed following intraoperative angiography (Table 2). In the remaining 34 cases in which ICG-VA indicated satisfactory aneurysm clipping, IA subsequently identified suboptimal reconstruction requiring clip modifications in 4 cases (Table 3). Of the cases requiring adjustments, 3 cases were unruptured, and 1 was a subarachnoid hemorrhage. In 2 cases, IA identified poor filling of adjacent vessels. One case involved residual aneurysm filling. In the fourth case, the patient underwent clipping of both Pcomm and Acomm aneurysms via the same craniotomy. In this case, the ICG-VA used to focus on the Acomm aneurysm (primary focus) failed to reveal persistent neck filling of the initially clipped Pcomm aneurysm that was believed to be straightforward. Overall, there was a postIA clip adjustment rate of 10.8% (Table 3).
NEUROSURGERY
IA Only There were 60 craniotomies performed that included IA alone. Following IA, there were 7 instances in which changes were made based on the IA findings (11.7%). Of the cases requiring adjustments, 5 cases were unruptured, and 2 had subarachnoid hemorrhages. This included persistent aneurysm filling in 4 cases, 1 case of persistent aneurysm neck filling, 1 case of a previously unidentified aneurysm lobule, and 1 case of poor filling of a fetal posterior cerebral artery (Table 4). ICG-VA1IA vs IA Alone When comparing the post-IA clip adjustment rates of all patients, there was no statistically significant difference between ICG-VA1IA (10.8% [95% confidence interval (CI), 3%-25%]) vs IA alone (11.7% [95% CI, 5%-23%]) (P = .897). If the patients who had immediate post-ICG-VA adjustments are
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 439
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
FIGURE. Distribution of aneurysm cases by intraoperative imaging. IA, intraoperative angiography; ICGVA, indocyanine green videoangiography.
excluded, the post-IA adjustment rate increases to 11.8% (95% CI, 3%-28%) (4 patients of 34), which did not reach statistical significance compared with IA alone (11.7% [95% CI, 5%23%]) (P = .999). Finally, if one assumes that the 3 patients who required adjustments after ICG-VA would have otherwise required changes after IA, and thus include them in that group (ie, rate of cases which need any postimaging adjustment), this would increase the postclip adjustment rate after IA to 18.9% (95% CI, 8%-35%) (7 of 37 patients). There is no statistical
difference between this rate and the rate of adjustments in the IAalone group (P = .324).
DISCUSSION Intraoperative Angiography Intraoperative angiography is the gold standard for the immediate assessment of microsurgical clipping of intracranial aneurysms, although it remains to be performed routinely in all
TABLE 2. ICG-VA Positive, Change Made, IA Negative (Presumed True Positives)a
TABLE 3. ICG-VA Negative, IA Positive (False Negatives)a
Patient No.
Aneurysm Aneurysm Reason for Clip Location Size, mm Adjustment
Patient No.
MCA
1
Unruptured
Acomm
2
Unruptured
PICA
3
Unruptured
MCA
4
SAH
1
Rupture Status Unruptured
2
Unruptured
3
Unruptured
ICA terminus MCA MCA MCA
a
9.4
5
2 13
Persistent aneurysm filling Persistent aneurysm filling Fusiform aneurysm, ICG-VA used to confirm adequacy of collaterals before permanent clipping
ICG-VA, indocyanine green video angiography; IA, intraoperative angiography; ICA, internal carotid artery; MCA, middle cerebral artery.
440 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
Rupture Status
Aneurysm Aneurysm Reason for Clip Location Size, mm Adjustment 12 4.2 14
Acomm Acomm
2.2 10
Pcomm
8.3
Poor A1-A2 inflow Poor vertebral artery filling Residual filling at base of Acomm aneurysm Residual neck of Pcomm aneurysmb
a
ICG-VA, indocyanine green video angiography; IA, intraoperative angiography; MCA, middle cerebral artery. Acomm, anterior communicating artery; Pcomm, posterior communicating artery; PICA, posterior inferior cerebellar artery; SAH, subarachnoid hemorrhage. b PComm was not visualized with the ICG as field of view was on the AComm aneurysm.
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
INTRAOPERATIVE ANGIOGRAPHY AND INDOCYANINE GREEN
TABLE 4. IA Only, Positive Angiograma Patient No.
Rupture Status
Aneurysm Location
Aneurysm Reason for Clip Size, mm Adjustment
1
Unruptured Paraophthalmic
6
Persistent aneurysm neck
2
Unruptured
MCA
8
3
SAH
Pcomm
6.8
Persistent aneurysm filling Persistent aneurysm filling
4
Unruptured
Acomm
7.4
5
SAH
Pcomm
6
Unruptured
PICA
5.6
7
Unruptured
Acomm
4.5
ICA blister
15
Persistent aneurysm filling Poor filling of fetal PCA Persistent aneurysm filling Residual filling of previously unidentified aneurysm lobule
1
a
IA, intraoperative angiography; ICA, internal carotid artery; MCA, middle cerebral artery. Acomm, anterior communicating artery; PCA, posterior cerebral artery; Pcomm, posterior communicating artery; PICA, posterior inferior cerebellar artery; SAH, subarachnoid hemorrhage.
centers. The technique, safety profile, and efficacy have been well described in the literature with clip readjustment rates reported in the range of 7% to 12.4% in large series.1-4,17 Intraoperative angiography is safe in the setting of its routine use at high-volume centers.1,3,4 Although the routine use of intraoperative angiography is not universally accepted, some have argued that, to make it safe and efficient in cases where it is most needed, it should be used routinely in all cases.18 In this study, the rate of clip adjustments in IA alone was 11.7%, consistent with previous reports. Despite the speed at which IA can be performed when done routinely, we continue to use intraoperative somatosensory and electroencephalographic monitoring as additional adjuncts to detect for signs of ischemia necessitating clip adjustment prior to IA.19 Intraoperative ICG-VA Compared With Postoperative Imaging In contrast to the relatively few studies comparing ICG-VA to IA, there have been a number of studies that compare ICG-VA to postoperative imaging modalities. The major limitation of assessing the adequacy of ICG-VA with postoperative imaging is the inability to implement adjustments as soon as they are needed. The ICG-VA to postoperative imaging discordance rates vary from 0% to 12%.8-14,20
NEUROSURGERY
Intraoperative ICG-VA in Combination With IA The use of ICG-VA was first reported for use in intracranial aneurysm surgery in 2003 with the use of a digital camcorder and was subsequently modified for integration into the operative microscope.5,16 Subsequently, a few studies have assessed the accuracy of ICG-VA compared with IA. Raabe et al6 report a 10% disagreement rate between ICG-VA and IA in a series of 60 aneurysms. Gruber et al15 report a 2.5% ICG-VA and IA rate of disagreement in 123 aneurysms. However, this study also utilized other adjunctive modalities such as microvascular Doppler and intraoperative neuroendoscopy, which may have helped to decrease the discordance rate. More recently, Washington et al7 reported a discordance rate of 14.3% in 49 patients undergoing aneurysm surgery. In our study, the rate of disagreement between IA and ICG-VA was 10.8%, similar to the above studies. The rate of post-IA clip adjustment in cases using ICG-VA and IA is likely lower than it would have been if ICG-VA had not been used. In our series, there were 3 cases in which ICG-VA led to a clip adjustment that was then confirmed with a satisfactory intraoperative angiogram. Assuming that the ICG-VA finding was not a false positive (ie, ICG-VA suggests residual aneurysm filling, when IA would not have), these findings would have presumably been discovered on IA and represent 3 more aneurysms that would have required post-IA adjustments, which would have raised the rate from 10.8% to 18.9%. However, in our study, this increased rate did not reach statistical significance in comparison with the adjustment rate of 11.7% in the IA-alone group. Suboptimal Clipping Theory One difficulty in understanding the benefit of ICG-VA is that it can be difficult to know the true value of ICG-VA, because knowledge that IA will follow ICG-VA may impact how a surgeon performs the initial clip reconstruction.21 In such circumstances, the surgeon may, consciously or subconsciously, pursue a less aggressive clipping strategy with the anticipation of imaging feedback from the relatively quick and low-risk ICG-VA. The surgeon would then have the opportunity to make adjustments based on the ICG-VA before performing the more timeconsuming and invasive IA. If this “suboptimal clipping theory” is applied, an elevated rate of inadequate clippings detected with ICG-VA that could then be corrected prior to IA would be anticipated. This would lead to a higher rate of ICG-VA to IA disagreement, and thus suggest that ICG-VA may be less valuable when compared with IA.21 A similar argument has been put forth when assessing the utility of IA alone.18 It is true that, when determining ICG-VA/IA discordance rates, the value of ICG-VA may in fact be undervalued because of this principle. In our study, however, rather than examining ICG-VA/IA discordance rates, we focus on the rate of clip adjustments following IA (whether or not ICG-VA was performed) in order to determine the added benefit of ICG-VA in addition to IA. Here, knowledge that IA will follow ICG-VA may overvalue the role of ICG-VA as follows. If the surgeon initially performs a suboptimal clipping knowing
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 441
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
that ICG-VA is available, then the mistake can be recognized following ICG-VA and corrected prior to IA. As such, the number of post-IA adjustments will decrease. This is highlighted in our study by the 3 patients who had “true positive” ICG-VA results leading to clip adjustments following ICG-VA and prior to IA. If ICG-VA had not been performed for some reason, but the surgeon had performed the same “suboptimal” clipping techniques, then presumably these suboptimal clippings would have been seen on subsequent IA, and thus these 3 patients, in addition to the 4 others who required post-IA clip adjustments, would have resulted in a post-IA clip adjustment rate of 18.9% (7 of 37 patients). In this cohort, the actual post-IA clip adjustment rate was just 10.8%. Therefore, it can be hypothesized that the use of ICGVA decreased the need for clip adjustments by 8.1%. However, under the “suboptimal clipping theory,” if ICG-VA were truly not available, and thus the surgeon was forced to give his/her “best effort,” it could be argued that the surgeon might have been more selective in the initial clip placement, and thus would have obtained an overall decreased rate of clip adjustments of less than 18.9%. Thus, under the “suboptimal clipping theory” when comparing rates of post-IA clip adjustments (rather than ICG-VA/ IA discordance rates), the knowledge that IA will follow ICG-VA likely leads to overvaluing the role of ICG-VA. Interestingly, in our study, even if the “suboptimal clipping theory” were true and the value of ICG-VA is overestimated, it still did not lead to a statistically significant difference in rates of post-IA clip adjustments between cohorts (18.9% vs 11.7%, P = .324). Limitations and Benefits of ICG-VA Many have suggested that ICG-VA and IA be used as complementary modalities rather than mutually exclusive methods. We agree that ICG-VA should not replace IA for the intraoperative assessment of aneurysm clipping. However, our data suggest when IA is routinely performed, there is little benefit in the use of ICG-VA to decrease post-IA clip adjustment rates. Our study did not address several situations where the use of ICGVA may be beneficial, including bypasses, arteriovenous malformation resections, and dural arteriovenous fistula obliterations. Furthermore, ICG-VA may be a useful adjunct in aneurysm surgery when operating on multiple aneurysms via a single craniotomy and the surgeon wishes to have a “first pass” look at the initial vessel reconstruction following clipping of the first aneurysm, but to defer IA until all aneurysms are clipped to minimize patient risk. In addition, ICG-VA is significantly faster than IA, and may be particularly useful in instances where rapid assessment and reassurance of relevant vessel patency is desired before formal IA, with less risk of prolonged ischemic time. Furthermore, ICG-VA may be a useful tool in patients for whom IA may be associated with increased risk (ie, contrast allergy or history of previous stroke after angiography). In our study, we excluded 3 patients in whom ICG-VA was performed, but not IA. One patient had a previous stroke from an angiogram. Two patients were deemed too critically ill to tolerate the time needed for an IA.
442 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
Our study assessed the overall rate of clip adjustments required when ICG-VA is added to the routine use of IA. However, there may be specific examples when ICG-VA is useful in aneurysm surgery. For example, de Oliveira et al22 examined the use of ICGVA for assessing perforating arteries not visible on preoperative DSA during intracranial aneurysm surgery, although this too had a high false-negative rate. ICG-VA is superior to IA in evaluating the patency of the perforating arteries in the operative field and is likely superior in identifying the critical vessels of small caliber such as the anterior choroidal artery. Regardless of the method used to evaluate adequate clipping, judicious identification of the perforating arteries remains critical as a central tenet of aneurysm surgery, because compromise of these vessels in the operative field will only be appreciated with ICG-VA if they were first recognized during the microsurgical dissection. Other examples of advances in the use of ICG-VA include combining it with Doppler and endoscopy.23-25 This exciting new application of ICG-VA allows direct visualization of vasculature not otherwise readily seen on microscope-integrated ICG-VA. Another elegant method to visualize hidden arteries during ICGVA is the use of a Yasargil mirror.26 In a small series of patients, Schnell et al27 examined the feasibility of ICG-VA combined with intraoperative computed tomography angiography/perfusion as complementary methods. Furthermore, ICG-VA uniquely allows the surgeon to manipulate the clipped aneurysm while simultaneously assessing vessel and aneurysm patency in a manner that is not possible with IA. Intraoperative angiography is limited to the assessment of the vessels injected, whereas systemically injected ICG-VA will provide for fluorescence of the entire cerebral vasculature. This may become particularly advantageous when operating on midline aneurysms associated with the Acomm. A single ICG injection may allow for complete visualization of the communicating complex, whereas complete assessment by using IA could require bilateral carotid injections, depending on the patency of the Acomm. We did not perform preclipping ICG-VA in this study, although used by others, it is possible preclipping ICG-VA could improve its utility by providing a “control” image prior to aneurysm clipping. Given the overall tendency in our study for larger aneurysms to require clip adjustments regardless of imaging modality, ICG-VA may be useful in larger aneurysms that are typically more difficult to reconstruct and require a more extensive dissection. Another potential improvement in ICGVA is the quantitative rather than qualitative analysis of ICG-VA, which is reported in multiple studies.28-30 This approach offers a numerical analysis of the ICG-VA imaging rather than relying on the surgeon’s interpretation of the video images. However, 1 major limitation continues to be that only the microscope images directly included in the operative view can be analyzed. As such, it is difficult to view aneurysms in multiple locations. Intraoperative angiography at present is superior to ICG-VA for this indication. In our study, residual filling of 1 Pcomm aneurysm neck (Table 3, patient 4) was not identified by ICG-VA, because the field of view was focused on a different location (Acomm region for a second
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
INTRAOPERATIVE ANGIOGRAPHY AND INDOCYANINE GREEN
aneurysm). Although ICG-VA would likely have discovered this filling if the field of view were focused on this region, the inability to view multiple fields of view simultaneously presents a limitation of ICG-VA that is not encountered with IA.
CONCLUSION Intraoperative angiography is the gold standard for the intraoperative assessment of clipping of intracranial aneurysms. The use of ICG-VA has not replaced the need for IA. Despite the expectation that ICG-VA might decrease the need for clip adjustments following IA, our study fails to demonstrate a statistically significant decrease in clip adjustment rates with ICG-VA when IA is routinely performed. Given the complexity of interpreting the results in which the presence of 1 diagnostic test may influence the outcome of a subsequent test, a randomized study should be performed to determine the true utility of this exciting emerging imaging modality in comparison with intraoperative angiography alone. ICG-VA continues to have a valuable role in cerebrovascular microsurgery, but does not appear, in our study, to provide any significant decrease in post-IA clip adjustment rate when IA is routinely performed for all aneurysm surgery. As with any tool, it is essential to recognize the strengths and limitations of each in order to tailor its application to the individual and optimize patient outcomes. Further studies are needed to identify the specific instances when ICG-VA is most useful for aneurysm surgery. Disclosure Dr Radvany has received honoraria from Siemens Medical; Dr Colby, from Covidien, Microvention; and Dr Coon, from Covidien. The other authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES 1. Chiang VL, Gailloud P, Murphy KJ, Rigamonti D, Tamargo RJ. Routine intraoperative angiography during aneurysm surgery. J Neurosurg. 2002;96(6):988-992. 2. Klopfenstein JD, Spetzler RF, Kim LJ, et al. Comparison of routine and selective use of intraoperative angiography during aneurysm surgery: a prospective assessment. J Neurosurg. 2004;100(2):230-235. 3. Chalouhi N, Theofanis T, Jabbour P, et al. Safety and efficacy of intraoperative angiography in craniotomies for cerebral aneurysms and arteriovenous malformations: a review of 1093 consecutive cases. Neurosurgery. 2012;71(6):1162-1169. 4. Tang G, Cawley CM, Dion JE, Barrow DL. Intraoperative angiography during aneurysm surgery: a prospective evaluation of efficacy. J Neurosurg. 2002;96(6):993-999. 5. Raabe A, Beck J, Gerlach R, Zimmermann M, Seifert V. Near-infrared indocyanine green video angiography: a new method for intraoperative assessment of vascular flow. Neurosurgery. 2003;52(1):132-139; discussion 139. 6. Raabe A, Nakaji P, Beck J, et al. Prospective evaluation of surgical microscopeintegrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. J Neurosurg. 2005;103(6):982-989. 7. Washington CW, Zipfel GJ, Chicoine MR, et al. Comparing indocyanine green videoangiography to the gold standard of intraoperative digital subtraction angiography used in aneurysm surgery. J Neurosurg. 2013;118(2):420-427. 8. Jing Z, Ou S, Ban Y, Tong Z, Wang Y. Intraoperative assessment of anterior circulation aneurysms using the indocyanine green video angiography technique. J Clin Neurosci. 2010;17(1):26-28. 9. Ma CY, Shi JX, Wang HD, Hang CH, Cheng HL, Wu W. Intraoperative indocyanine green angiography in intracranial aneurysm surgery: microsurgical clipping and revascularization. Clin Neurol Neurosurg. 2009;111(10):840-846.
NEUROSURGERY
10. Li J, Lan Z, He M, You C. Assessment of microscope-integrated indocyanine green angiography during intracranial aneurysm surgery: a retrospective study of 120 patients. Neurol India. 2009;57(4):453-459. 11. Lin J, Zhao J, Zhao Y, et al. Multiple intraoperative monitoring-assisted microneurosurgical treatment for anterior circulation cerebral aneurysm. J Int Med Res. 2011;39(3):891-903. 12. Khurana VG, Seow K, Duke D. Intuitiveness, quality and utility of intraoperative fluorescence videoangiography: Australian Neurosurgical Experience. Br J Neurosurg. 2010;24(2):163-172. 13. Wang S, Liu L, Zhao Y, Zhang D, Yang M, Zhao J. Evaluation of surgical microscope-integrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. Neurosurg Rev. 2011;34(2):209-215. 14. Moon HS, Joo SP, Seo BR, Jang JW, Kim JH, Kim TS. Value of indocyanine green videoangiography in deciding the completeness of cerebrovascular surgery. J Korean Neurosurg Soc. 2013;53(6):349-355. 15. Gruber A, Dorfer C, Standhardt H, Bavinzski G, Knosp E. Prospective comparison of intraoperative vascular monitoring technologies during cerebral aneurysm surgery. Neurosurgery. 2011;68(3):657-673; discussion 673. 16. Raabe A, Beck J, Seifert V. Technique and image quality of intraoperative indocyanine green angiography during aneurysm surgery using surgical microscope integrated nearinfrared video technology. Zentralbl Neurochir. 2005;66(1):1-6; discussion 7-8. 17. Martin NA, Bentson J, Viñuela F, et al. Intraoperative digital subtraction angiography and the surgical treatment of intracranial aneurysms and vascular malformations. J Neurosurg. 1990;73(4):526-533. 18. Heros RC. Intraoperative angiography. J Neurosurg. 2002;96(6):979-980; discussion 980. 19. Wicks RT, Pradilla G, Raza SM, et al. Impact of changes in intraoperative somatosensory evoked potentials on stroke rates after clipping of intracranial aneurysms. Neurosurgery. 2012;70(5):1114-1124. 20. Dashti R, Laakso A, Niemelä M, Porras M, Hernesniemi J. Microscope-integrated near-infrared indocyanine green videoangiography during surgery of intracranial aneurysms: the Helsinki experience. Surg Neurol. 2009;71(5):543-550; discussion 550. 21. Morcos JJ. Editorial: indocyanine green videoangiography or intraoperative angiography? J Neurosurg. 2013;118(2):417-418; discussion 418-419. 22. de Oliveira JG, Beck J, Seifert V, Teixeira MJ, Raabe A. Assessment of flow in perforating arteries during intracranial aneurysm surgery using intraoperative nearinfrared indocyanine green videoangiography. Neurosurgery. 2007;61(3 suppl): 63-72; discussion 72-73. 23. Imizu S, Kato Y, Sangli A, Oguri D, Sano H. Assessment of incomplete clipping of aneurysms intraoperatively by a near-infrared indocyanine green-video angiography (Niicg-Va) integrated microscope. Minim Invasive Neurosurg. 2008;51:199-203. 24. Bruneau M, Appelboom G, Rynkowski M, Van Cutsem N, Mine B, De Witte O. Endoscope-integrated ICG technology: first application during intracranial aneurysm surgery. Neurosurg Rev. 2013;36(1):77-84; discussion 84-85. 25. Nishiyama Y, Kinouchi H, Senbokuya N, et al. Endoscopic indocyanine green video angiography in aneurysm surgery: an innovative method for intraoperative assessment of blood flow in vasculature hidden from microscopic view. J Neurosurg. 2012;117(2):302-308. 26. Wilson J, Screven R, Volk J, Payner T. Use of a Yas¸argil mirror as an adjunct to indocyanine green angiography to evaluate the patency of elusive posterior communicating arteries during aneurysm clipping: case report. Neurosurgery. 2012; 71(1 suppl operative):195-197. 27. Schnell O, Morhard D, Holtmannspötter M, Reiser M, Tonn JC, Schichor C. Nearinfrared indocyanine green videoangiography (ICGVA) and intraoperative computed tomography (iCT): are they complementary or competitive imaging techniques in aneurysm surgery? Acta Neurochir (Wien). 2012;154(10):1861-1868. 28. Son YJ, Kim JE, Park SB, Lee SH, Chung YS, Yang HJ. Quantitative analysis of intraoperative indocyanine green video angiography in aneurysm surgery. J Cerebrovasc Endovasc Neurosurg. 2013;15(2):76-84. 29. Oda J, Kato Y, Chen SF, et al. Intraoperative near-infrared indocyanine greenvideoangiography (ICG-VA) and graphic analysis of fluorescence intensity in cerebral aneurysm surgery. J Clin Neurosci. 2011;18(8):1097-1100. 30. Chen SF, Kato Y, Oda J, et al. The application of intraoperative near-infrared indocyanine green videoangiography and analysis of fluorescence intensity in cerebrovascular surgery. Surg Neurol Int. 2011;2(1):42.
Acknowledgment The authors thank Xiaobu Ye, MD, for her assistance with the statistical analysis.
VOLUME 75 | NUMBER 4 | OCTOBER 2014 | 443
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
CAPLAN ET AL
COMMENTS
T
he authors report a retrospective single-surgeon series of 94 patients who underwent 97 craniotomies for 128 aneurysms at 2 campuses of the same center over a 21-month period. In 37 of the 97 craniotomies considered, clip placement was assessed by using both indocyanine green angiography (ICGA) and conventional intraoperative digital subtraction angiography (DSA), whereas in the remaining 60 craniotomies DSA was the only technique available. The authors compare clip repositioning rates between the 2 groups and find an insignificant difference of 10.8% (4/37) for the ICGA1DSA group vs 11.7% (7/60) for the DSA-only group (P = .89). In the first group, ICGA was performed after aneurysm clipping and indicated the need for clip repositioning in 3/37 cases (8.1%). Thereafter, DSA was performed and disclosed the need for clip modification in another 4/37 cases (10.8%). Therefore, in a total of 7/37 cases (18.9%) aneurysm misclipping was discovered by using ICGA and DSA in conjunction. In the second group, DSA was used alone and disclosed a misclipping rate of 11.7% (7/60). The key information of this article, however, is not that ICGA does not provide a significant benefit when used in conjunction with conventional DSA in decreasing clip repositioning rates compared with DSA alone (10.8% vs 11.7%), but that within the group of patients receiving both ICGA and DSA, ICGA resulted in a 43% relative risk reduction for aneurysm misclipping (18.9% vs 10.8%). Engelbert Knosp Andreas Gruber Vienna, Austria The study compared the rate of false-negative results after aneurysm clipping of 97 patients. In 37 craniotomies ICG-VA and IA was carried out and in other 60 craniotomies only IA was performed. In 8% of the craniotomies the ICG-VA detected a residual aneurysm filling or vessel occlusion, which could be solved immediately. A subsequently
444 | VOLUME 75 | NUMBER 4 | OCTOBER 2014
performed IA showed a vessel stenosis in two cases, a residual aneurysm filling in one case and also a residual aneurysm filling of an aneurysm, which was not focused in ICG-VA. This leads to a rate of 8%-11%, depending on whether 3 or 4 cases were calculated as false-negative. In the “IA-only” group the clip position was adjusted in 12% of the cases. These findings are similar to our own prospective results.1 Unfortunately, complications were not presented in detail. Considering the self-cited publication about IA at JHBMC,2 2.6% of the IA had complications (including stroke; vs 0.1% complication rate for ICG-VA1). It might be possible that some of these complications can be reduced using ICG-VA routinely (time required for ICG-VA 1-2 min vs 15-35 min for IA). This publication also detected a false-negative rate of 8% for IA proven by postoperative angiography; and these false-negative patients subsequently rebled. Altogether, nowadays every clipped aneurysm should be controlled intraoperatively, either by (several) ICG-VA or by IA. Also a postoperative 3D angiography should be performed in any patient to identify falsenegative cases (with ICG-VA and IA approx. 10%) and if necessary to treat them. Despite the advantages and disadvantages of ICG-VA or IA, a local standard and setting should be created, ensuring the safest method and offering the most comfortable possibility to the surgeon to check the complete occlusion of the aneurysm during the operation. Volker Seifert Frankfurt, Germany
1. Raabe A, Nakaji P, Beck J, et al. Prospective evaluation of surgical microscopeintegrated intraoperative near-infrared indocyanine green videoangiography during aneurysm surgery. J Neurosurg. 2005;103(6):982-989. 2. Chiang VL, Gailloud P, Murphy KJ, Rigamonti D, Tamargo RJ. Routine intraoperative angiography during aneurysm surgery. J Neurosurg. 2009;96(6): 988-992.
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
