- Email: [email protected]
Long-Term, Continuous Intra-Arterial Nimodipine Treatment of Severe Vasospasm After Aneurysmal Subarachnoid Hemorrhage
Accepted Manuscript Long-term, continuous intra-arterial nimodipine treatment of severe vasospasm following aneurysmal subarachnoid hemorrhage Konstantin Hockel, MD, Jennifer Diedler, MD, PhD, Jochen Steiner, MD, Ulrich Birkenhauer, MD, Sören Danz, MD, Ulrike Ernemann, MD, PhD, Martin U. Schuhmann, MD, PhD PII:
S1878-8750(15)01704-0
DOI:
10.1016/j.wneu.2015.11.081
Reference:
WNEU 3497
To appear in:
World Neurosurgery
Received Date: 1 July 2015 Revised Date:
23 November 2015
Accepted Date: 24 November 2015
Please cite this article as: Hockel K, Diedler J, Steiner J, Birkenhauer U, Danz S, Ernemann U, Schuhmann MU, Long-term, continuous intra-arterial nimodipine treatment of severe vasospasm following aneurysmal subarachnoid hemorrhage, World Neurosurgery (2016), doi: 10.1016/ j.wneu.2015.11.081. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Long-term,
continuous
intra-arterial
nimodipine
treatment
of
severe
vasospasm following aneurysmal subarachnoid hemorrhage
Konstantin Hockel, MD1
RI PT
Jennifer Diedler, MD, PhD1 Jochen Steiner, MD1 Ulrich Birkenhauer, MD1 Sören Danz, MD2 Ulrike Ernemann, MD, PhD2
SC
Martin U. Schuhmann, MD, PhD1 1
M AN U
Department of Neurosurgery, University Hospital Tübingen, University of Tübingen;
2
Diagnostic and Interventional Neuroradiology, Department of Radiology, University
Corresponding author:
EP
Konstantin Hockel
TE D
Hospital Tübingen, University of Tübingen, Germany.
Department of Neurosurgery
University Hospital Tübingen
AC C
Hoppe-Seyler-Str. 3 72076 Tübingen Germany
Phone: +49 7071 2986679 Fax: +49 7071 5245 Email: [email protected]
Key words: subarachnoid hemorrhage, vasospasm, nimodipine, brain tissue oxygen, cerebral autoregulation, continuous intra-arterial nimodipine, DCI.
Hockel et al.
ACCEPTED MANUSCRIPT Abbreviations CIN – continuous intra-arterial nimodipine CPP – cerebral perfusion pressure CTA – computed tomography angiography CTP – computed tomography perfusion
RI PT
DCI – delayed cerebral ischemia DSA – digital subtraction angiography EVD – external ventricular drainage GOS – Glasgow outcome scale
SC
ICA – internal carotid artery ICP – intracranial pressure
MCA – middle cerebral artery ORx – oxygen reactivity index PRx – pressure reactivity index SAH – subarachnoid hemorrhage
M AN U
MAP – mean arterial pressure
AC C
EP
TE D
TCD – transcranial Doppler sonography
2
Hockel et al.
ACCEPTED MANUSCRIPT Abstract Background:
Secondary
vasospasm
and
disturbances
in
cerebrovascular
autoregulation are associated with the development of delayed cerebral ischemia (DCI) following aneurysmal subarachnoid hemorrhage (SAH). An intra-arterial
RI PT
application of nimodipine has been shown to increase the vessel diameter, although this effect is transient. The feasibility of long-term, continuous, intra-arterial nimodipine (CIN) treatment and its effects on macrovasospasm, autoregulation
SC
parameters and outcome were evaluated in patients with refractory severe macrovasospasm.
M AN U
Methods: Ten patients were included with refractory macrovasospasm despite bolus nimodipine application (n=4) or with primary severe vasospasm (n=6). The patients were assessed with continuous multimodal neuromonitoring (MAP, ICP, CPP, pbrO2), daily TCD exams and CT-angiography/perfusion (CTA, CTP). Autoregulation indices,
TE D
the pressure reactivity index (PRx) and oxygen reactivity index (ORx) were calculated. Indwelling microcatheters were placed in the extracranial internal carotid
EP
arteries and 0.4 mg nimodipine was continuously infused at 50 ml/hour. Results: The duration of CIN treatment ranged from 9-15 days. During treatment ICP
AC C
remained stable, TCD flow velocity decreased and pbrO2 improved by 37%. Macrovasospasm, as assessed via CTA, had improved (n=5) or disappeared (n=5) at the end of treatment. Cerebrovascular autoregulation according to the PRx and ORx significantly worsened during treatment. All patients showed a favorable outcome (median GOS 5) at 3 months. Conclusions: In well-selected patients with prolonged severe macrovasospasm, continuous intra-arterial nimodipine treatment can be applied as a rescue therapy with relative safety for more than 2 weeks to prevent secondary cerebral ischemia.
3
Hockel et al.
ACCEPTED MANUSCRIPT
The induced impairment of cerebrovascular autoregulation during treatment seems to
AC C
EP
TE D
M AN U
SC
RI PT
have no negative effects.
4
Hockel et al.
ACCEPTED MANUSCRIPT Introduction
Delayed cerebral ischemia (DCI) is the secondary reason for an unfavorable outcome following aneurysmal subarachnoid hemorrhage (SAH). Secondary macrovascular spasm and disturbances in the cerebrovascular autoregulation,
RI PT
located in the precapillary microvessels, have been identified as associated risk factors for DCI.1,2 In clinical practice, the diagnosis and subsequent therapy of angiographically detected macrovasospasm remain keystones in DCI prevention,3
SC
especially in comatose or sedated patients where no feedback on neurological function can be obtained.
M AN U
Limited therapeutic strategies have been shown to effectively improve clinical outcome and mortality. Oral nimodipine and hemodynamic augmentation are wellaccepted
basic
treatment
regimens.4-6
In
refractory
cases
of
severe
macrovasospasm, endovascular interventions have been implemented as alternative
TE D
options; however, the scientific evidence remains scarce.
The most commonly used techniques are mechanical, e.g., transluminal balloon angioplasty, or pharmacological vasodilatation, e.g., papaverine and calcium channel
EP
blockers, such as nimodipine or verapamil.7-11 Rebound ischemia, the need for repeated interventions and the hazards of patient transport remain significant risk
AC C
factors.9,11
To overcome the drawbacks of single-time interventions, the continuous intra-arterial application of vasodilators, such as nimodipine, using endovascular microcatheters in the internal carotid artery (ICA) has been suggested. Singular reports in cases of refractory macrovasospasm have demonstrated the technical feasibility and the potential to improve a compromised cerebral perfusion.12-16 To date, nimodipine infusion has only been applied for a limited number of days; however, the critical period of macrovasospasm may last for up to two weeks.7,9 5
Hockel et al.
ACCEPTED MANUSCRIPT
In this pilot study, we report our experience with long-term, continuous intra-arterial nimodipine infusion (CIN) in thirteen patients following aneurysmal SAH with refractory severe macrovasospasm. We evaluated the feasibility and safety of longterm CIN treatment as primary end points. The effects on macrovasospasm, brain
RI PT
tissue oxygenation and cerebrovascular autoregulation indices, including the pressure reactivity index (PRx) and oxygen reactivity index (ORx), were assessed as secondary end points. Furthermore, patient outcome at discharge and at 3 months is
SC
reported.
M AN U
Patients and Methods
Patient inclusion and general management
Thirteen patients with aneurysmal SAH were selected for CIN within a 10-month period. Sixty patients with a diagnosis of SAH were treated in our department during
TE D
this time period.
This study comprises a retrospective data analysis of a cohort of patients with refractory macrovasospasm who were treated on an individual basis with intra-
EP
arterial continuous nimodipine application. In this specific cohort of SAH patients, who were threatened by cerebral infarction as a result of severe refractory
AC C
macrovasospasm, we performed an “individual treatment attempt” with CIN application via indwelling microcatheters. This protocol was not performed in a prospective study setting with a specific research question; instead, it comprised a last tier treatment option in individual cases. Thus, IRB approval was not required from our institute. However, IRB approval was obtained for computerized neuromonitoring and data collection for the retrospective data analysis of all patients in the intensive care unit (ICU). The patients’ relatives provided their informed consent for the individual treatment attempt using CIN. 6
Hockel et al.
ACCEPTED MANUSCRIPT
Following admission to the ICU, diagnostic imaging, including computed tomography angiography (CTA) and digital subtraction angiography (DSA), was performed. The patients underwent subsequent aneurysm occlusion via microsurgical clipping or
RI PT
endovascular coiling. In cases of acute hydrocephalus, an extraventricular drain was inserted within 24 hours of admission. If not primarily intubated, sedation (midazolam, sufentanyl) and mechanical ventilation were induced prior to aneurysm treatment.
SC
Further intensive care management was conducted according to our current ICU standards. Mechanical ventilation was guided to maintain the arterial pO2 at 110 ± 5
M AN U
mmHg and the arterial pCO2 between 35-40 mmHg. The fluid balance was aimed at normovolemia. Catecholamines (noradrenalin) were titrated in cases of arterial hypotension despite normovolemia to ensure cerebral perfusion pressures of ≥ 70 mmHg. Pre-emptive intravenous (iv) nimodipine was not routinely used in the
TE D
patients who required moderate to high doses of catecholamines; however, in normotensive or hypertensive patients, iv nimodipine was administered.
EP
Neuromonitoring and assessment of cerebral autoregulation All patients who could not be extubated within 24 hour after aneurysm treatment
AC C
(because of a primary poor initial clinical grade, significant intraoperative swelling or failure of the postoperative / postinterventional weaning attempt) received multimodal neuromonitoring, including intracranial pressure (ICP) monitoring, cerebral perfusion pressure calculation (CPP) and a brain tissue oxygen tension probe (pbrO2). A Neurovent-PTO probe (Raumedic AG, Helmbrechts, Germany) was used for the assessment of the ICP, pbrO2 and brain temperature; the probe was inserted into the frontal white matter of the hemisphere that was predominantly affected by the hemorrhage via a one-lumen bolt. The mean arterial pressure (MAP) was 7
Hockel et al.
ACCEPTED MANUSCRIPT
continuously monitored using a catheter inserted into the radial artery with the transducer referenced to the foramen of Monro. The ICP, MAP and pbrO2 signals were continuously recorded on a bedside-mounted device (Datalogger MPR2, Raumedic AG, Helmbrechts, Germany) and transmitted to
RI PT
the bedside hospital monitoring system. The three parameters were also digitally sampled at a rate of 100 Hz and forwarded to a notebook PC that was running ICM+ software (Cambridge Enterprise, Cambridge, UK). The CPP was calculated as the
SC
difference between the MAP and the ICP.
ICM+ software was used for both the online display of data and the retrospective
M AN U
analysis of the recorded monitoring parameters. For the retrospective analysis, the ICP and MAP data were subjected to manual artifact detection and removal. All parameters were calculated using a one-minute moving window. The parameters of cerebrovascular autoregulatory capacity, i.e., the PRx and ORx, were calculated as
TE D
previously described.17 In brief, the PRx was evaluated as the moving Pearson correlation coefficient between the averaged ICP and MAP calculated over the moving window length of 5 minutes. The ORx was evaluated as the moving Pearson
EP
correlation coefficient between the pbrO2 and CPP over an average period of 1 hour. The PR and OR indices may vary between -1 and 1. Intact cerebral autoregulation
AC C
can be assumed when the index values are close to zero, which indicates that no correlation between the ICP and MAP or the pbrO2 and CPP exists. Transcranial Doppler (TCD) examinations were performed daily with a hand-held 2MHz probe (EME Companion, Nicolet Vascular Ltd., Madison, USA) to assess the maximum blood flow velocities of the major intracranial arteries. For subsequent analyses, the mean values of the maximum blood flow velocity (cm/s) in the MCA on both sides were recorded.
8
Hockel et al.
ACCEPTED MANUSCRIPT Detection of cerebral macrovasospasm
Screening for macrovasospasm in sedated patients was performed via daily TCD examination. Neuromonitoring data were not used as a screening tool for macrovasospasm or possible DCI. In addition, the patients received diagnostic CTA
RI PT
and CTP imaging as a routine screening examination on day 5 after SAH or at the time point when an increase in the mean TCD flow velocity greater than 120 cm/s was identified for the first time. In cases of significant macrovasospasm according to
SC
the CTA/CTP criteria, i.e., significant vessel narrowing greater than 60% plus a respective perfusion deficit, conventional diagnostic angiography was performed to
M AN U
confirm macrovasospasm. In case of severe macrovasospasm (significant vessel narrowing greater than 60%), an intra-arterial bolus application of nimodipine, 3 mg over 30 minutes, was performed for each involved vascular territory in each patient. In awake patients which developed clinical signs of DCI (neurological deterioration)
TE D
conservative treatment was escalated according to our ICU standard protocol: the elevation and maintenance of a CPP ≥ 70 mmHg combined with iv nimodipine. In case neurological symptoms did not disappear, conventional digital angiography was
EP
performed. In case of macrovasospasm, an intra-arterial bolus application of nimodipine, 3 mg over 30 minutes, was performed for each involved vascular territory
AC C
in each patient.
Induction of CIN treatment CIN treatment was initiated if 1) a primary, severe macrovasospasm with respective perfusion deficits via CTP showed an insufficient response to the first bolus dose of nimodipine (same intervention) or 2) a recurrent vasospasm was identified via followup angiography (subsequent day) after the intra-arterial nimodipine bolus application and/or CT imaging showed evidence of new ischemic lesions. 9
Hockel et al.
ACCEPTED MANUSCRIPT
Over a femoral artery sheath, one or two microcatheters (14-Prowler Microcatheter, Cordis Corp., Hialeah, USA) were placed in the extracranial segment of the feeding artery of each involved vascular territory, i.e., typically the bilateral ICA or vertebral artery (VA, one patient). Intra-arterial nimodipine infusion was immediately initiated at
RI PT
0.4 mg/catheter/hour, with an infusion rate of 50 ml/hour. A comparable dose of nimodipine has been previously recommended.14 In our group, the nimodipine dosage ranged between 0.2-0.6 mg/catheter/hour. An infusion rate of 50 ml/hour was
SC
selected to prevent catheter obstruction and thrombus apposition. From the time point of catheter placement, of positioning maneuvers of the patient were reduced to
M AN U
a minimum. To minimize the risk of thromboembolic complications as a result of indwelling arterial catheters, anticoagulation was facilitated with unfractioned heparin to a partial thromboplastin time (PTT) of 60-80 seconds.
Following CIN induction, further treatment was continued in the ICU. Multimodal
TE D
neuromonitoring and daily TCD examinations were continued. During CIN treatment, no routine imaging studies, i.e., CT scan or angiography, were performed unless otherwise indicated by the individual course.
EP
There was no predefined endpoint for CIN treatment. The termination of CIN was decided on an individual basis, typically when the pbrO2 and TCD values indicated a
AC C
sustained improvement of brain perfusion or resolution of macrovasospasm, respectively. CIN was then paused for at least 6 hours, and subsequent CTA and CTP studies performed to confirm resolution of macrovasospasm. In this case and if CTA of the extracranial carotid artery indicated no signs of thrombus apposition, the indwelling arterial catheters were removed.
Statistical analysis
10
Hockel et al.
ACCEPTED MANUSCRIPT
For the final analysis of the physiological variables, i.e., the blood gases, ICP, MAP, CPP, and pbrO2, 24 hour mean values were calculated and presented for the days of interest prior to and during CIN treatment. Regarding the autoregulation indices PRx and ORx, a 12-hour window was evaluated, which focused on the quantitative
mean +/- standard error, if not otherwise indicated.
RI PT
changes during the start and end of CIN treatment. The data are represented as the
Statistical analysis was performed using SPSS statistical software (SPSS 21.0,
SC
SPSS Inc., Chicago, USA). The data were analyzed using paired t-tests and
M AN U
Wilcoxon signed rank tests. Statistical significance was assumed at p<0.05.
Results Clinical course
In two of the thirteen patients, CIN treatment was prematurely terminated after 2 and
TE D
8 days, respectively. In one patient, a postoperative hemorrhagic complication (intracerebral hemorrhage) occurred after a decompressive craniectomy was performed 3 days prior to the initiation of CIN. Another patient underwent a
EP
decompressive craniectomy on day 8 of CIN for re-hemorrhage, which resulted from an aneurysm previously occluded by a clip. The outcome of both patients was GOS 2
AC C
at discharge. In one additional patient, pbrO2 monitoring was not available prior to and during the first days of CIN treatment. These three patients were excluded from the final analysis of CIN effects on pbrO2, autoregulation indices and outcome. The distribution of gender, age, WFNS and Fisher grade, aneurysm location and treatment is shown in Table 1.
Seven of the ten included patients were admitted to the hospital within 48 hours after the initial hemorrhage. Three of the ten patients were awake after aneurysm 11
Hockel et al.
ACCEPTED MANUSCRIPT
treatment and presented with clinical symptoms of DCI, e.g., hemiparesis, aphasia. Subsequent angiography confirmed macrovasospasm. For the remaining seven patients, no clinical evaluation was feasible because of their poor initial clinical grade
RI PT
and continued sedation and ventilation after aneurysm treatment.
A significant angiographic vasospasm was identified after a mean of 5.7 days (days 4-9) following SAH (Table 2). The vasospasm was unilaterally localized in one case,
SC
bilaterally localized in the anterior circulation in seven patients and distributed in a generalized pattern in two patients. The vasospasm duration had a mean of 14.4
M AN U
days (10-20 days). In all cases, the vasospasm was classified as severe according to angiography criteria.
Four patients underwent an initial singular or repeat intra-arterial bolus treatment (mean 2.5 interventions) prior to the initiation of CIN (Table 2). Three of those four
TE D
patients developed new or further ischemic lesions visible on CT. In these four cases and in the six remaining patients with primarily severe macrovasospasm unresponsive to bolus treatment CIN treatment was initiated at a mean of 7.4 days
EP
(4-10 days) after SAH and continued for a mean of 12.0 days (9-15 days). Seven patients received the starting dosage of 0.4 mg/catheter/hour for the entire CIN
AC C
period. According to the behavior of the TCD flow velocities, nimodipine was increased to 0.6 mg/hour in two patients, whereas it was reduced to 0.2 mg/hour in one patient during the course of the treatment. During CIN, two patients developed further localized cerebral insults.
The noradrenalin dosage changed non-significantly from 0.20 ± 0.46 µg/kg/min to 0.27 ± 0.52 µg/kg/min during CIN treatment (p>0.05). The arterial blood pressure was not significantly affected by CIN treatment, and a CPP ≥ 75 mmHg was 12
Hockel et al.
ACCEPTED MANUSCRIPT
maintained (Table 4). CIN treatment led to an increased volume input by approximately 2.5 liters (50 ml/catheter/hour).
During the course of CIN treatment, CTA was performed after the nimodipine infusion
RI PT
was ceased for at least 6 hours at a time point as described above. Treatment was continued in eight of ten patients after the first attempt to stop CIN treatment, due to the presence of remaining significant vasospasm.
SC
The qualitative evaluation of the vasospasm degree, as determined via CTA by a dedicated neuroradiologist, indicated a stepwise decrease in the vasospasm severity
M AN U
until the end of CIN treatment (Table 3). Remaining vessel constrictions were considered of no or minor relevance for cerebral perfusion.
Catheter dislocation was detected via CTA in two of nineteen catheters at the end of CIN treatment. Both catheters were most likely proximally dislocated into the aortic
TE D
arch, which was not visible on CTA. The duration of the dislocation was unknown. In two other patients, CIN was terminated on days 12 and 14, respectively, because of
gastrointestinal
bleeding
and
heparin-induced
thrombocytopenia
(HIT),
(Table 2).
EP
complications that were presumably related to the applied continuous anticoagulation
AC C
Following the successful termination of CIN treatment, sedation was discontinued, and weaning from the respirator was achieved. At discharge, minor focal neurological deficits were detectable in some patients; however, the majority of the patients were able to ambulate independently, as indicated by a median GOS of 4 (range 3-5). At three months, the median GOS was 5 (range 4-5) (Table 2).
Neuromonitoring parameters and cerebral autoregulation
13
Hockel et al.
ACCEPTED MANUSCRIPT
TCD examinations of the MCA demonstrated an increase in the blood flow velocities to a mean value of 144 ± 9 cm/s prior to CIN treatment. Following the induction of CIN, the flow velocities continuously decreased over days. The TCD velocities significantly improved to a median of 99 ± 7 cm/s on day 10 (Fig. 1, p<0.05 compared
RI PT
with the time prior to CIN treatment); however, the velocities did not decrease below a threshold of 120 cm/s in any case.
SC
The total monitoring time according to ICM+ software was 4,856 hours. Premature failure of the pbrO2 probes during CIN treatment occurred in four patients, which
M AN U
occurred prior to day 10 of CIN treatment in two patients. The CPP, paO2 and paCO2 were maintained within the desired limits during the entire period of CIN treatment (Table 4). The mean ICP was 10 ± 1 mmHg prior to CIN treatment and changed nonsignificantly to a mean of 11 ± 1 mmHg during intra-arterial nimodipine infusion. The
TE D
mean brain tissue oxygen level decreased to 23.8 ± 3.5 mmHg on the day CIN treatment was initiated. During nimodipine infusion, the pbrO2, which represents a surrogate marker of cerebral blood flow, increased within 24 hours in all but 2
EP
patients; furthermore, it continuously increased to 32.7 ± 4.8 on day 10, which corresponds to a 37% improvement in pbrO2 during the course of CIN treatment
AC C
(p>0.05).
The indices of cerebrovascular autoregulation, PRx and ORx, were determined as 0.03 ± 0.02 and 0.05 ± 0.06, respectively, prior to the initiation of CIN treatment. Following the induction of CIN treatment, both indices simultaneously increased to 0.16 ± 0.05 and 0.13 ± 0.03, respectively, within 12 hours, which was statistically significant for the PRx. After the cessation of nimodipine infusion, both indices
14
Hockel et al.
ACCEPTED MANUSCRIPT
significantly decreased to 0.03 ± 0.03 and -0.07 ± 0.03, respectively, within 24 hours (p<0.05) (Fig. 2A + B).
Discussion
RI PT
In the current study, we demonstrate the feasibility and relative safety of long-term intra-arterial nimodipine infusion for the treatment of severe macrovasospasm in selected high-risk patients.
SC
Angiographically detected macrovasospasm is only one of the presumed pathomechanisms that contribute to DCI after SAH
18,19
; however, endovascular
M AN U
procedures that directly target refractory macro-vasoconstriction remain widely employed. The reduction of macrovasospasm does not necessarily lead to an improvement in neurological outcome
20
; however, previous studies have shown that
the intra-arterial application of nimodipine significantly reversed macrovasospasm in
TE D
approximately 50-65% of cases 7 and led to consecutive improvement in the cerebral perfusion in CTP, which lasted for up to 24 hours.9 In these studies, an immediate clinical improvement following a single treatment of intra-arterial nimodipine occurred
EP
in up to 75% of patients.
The vasospasm period, however, may last up to 14 days after the hemorrhage.7,9
AC C
Thus, in cases of recurring macrovasospasm, multiple interventions are necessary to prevent ischemia. Daily interventions result in daily patient transport from the ICU to the angio suite, which bears risks for hypothermia, ICP increase, ventilatory problems and hemodynamic instability. Nimodipine administration is associated with a reduction in blood pressure and, consequently, the cerebral perfusion pressure.21 Specifically, an intra-arterial bolus application, as performed as a single-time treatment, can lead to a profound decrease in the MAP and CPP, which requires high doses of counterbalancing catecholamines to maintain the cerebral perfusion within a 15
Hockel et al.
ACCEPTED MANUSCRIPT
desired range. Although the single-time nimodipine effect is more long-lasting compared with papaverine,9,11 in general, uncertainty exists regarding whether the single-time treatment with nimodipine will provide sufficient protection until the next intervention is scheduled. The continuous application of nimodipine over indwelling
RI PT
catheters placed proximal to the major cerebral arteries has been previously published.12-16 Previous studies have demonstrated the feasibility of the CIN
SC
procedure for a maximum duration of 6 days.
The present study is the first investigation of long-term CIN treatment over a period of
M AN U
9-15 days, which thus covers the entire time during which patients were at risk for macrovasospasm. Microcatheter placement in each internal carotid artery is most suitable and will affect the entire anterior circulation. Nevertheless, vertebral artery catheters that target posterior circulation vasospasm are also feasible. In contrast to
TE D
a bolus application, we demonstrated that continuous nimodipine infusion only mildly decreased the arterial blood pressure, and catecholamine doses were not significantly increased during treatment.
EP
Monitoring of the therapeutic effects during CIN treatment remains critical. Repeated imaging studies would contradict one aim of CIN, which is to decrease the number of
AC C
patient transports. Thus, continuous monitoring of CBF surrogate parameters should be performed during the entire treatment period.14,16 Our findings imply that the effect of continuous nimodipine on macrovasospasm is not immediately evident. The TCD flow velocities decreased slowly over days, and brain tissue oxygen monitoring was able to demonstrate a continuous but also rather slow increase in the pbrO2, which improved by 37% at the end of treatment. The levels of sedation, CPP and paO2 were stable during the treatment period; thus, the improvement in the pbrO2 in parallel to decreasing
TCD
flow
velocities
is
most 16
likely
reflecting
a
decrease
of
Hockel et al.
ACCEPTED MANUSCRIPT
macrovasospasms plus an improvement in the cerebral blood flow. In addition to the reduction of macrovasospasm and the improvement of cerebral perfusion, the rate of cerebral infarctions under CIN was low. Before initiation of CIN 4 of 10 patients under maximum standard therapy, including bolus intra-arterial nimodipine treatment,
additional but small infarct during CIN treatment.
RI PT
developed new infractions on CT, whereas only 2 of 10 patients suffered an
Furthermore, our findings support the hypothesis that nimodipine is also effective in
SC
dilating the distal cerebral vasculature, i.e., the precapillary arterioles not visible on angiography.22 The autoregulation indices PRx and ORx, which describe the
M AN U
cerebrovascular autoregulatory capacity and thus the functionality of the precapillary arterioles, exhibited a close time-effect-relationship to nimodipine infusion. After the initiation of CIN, the PRx and ORx worsened within 12 hours indicating an impairment of autoregulation; however, the postulated thresholds for loss of
TE D
autoregulation, i.e., 0.2 for PRx and 0.25 for ORx,17,23 were not reached in every patient. In our opinion, small precapillary vessel vasodilatation is the most likely explanation for the direct effect of nimodipine on the ICP-based PRx and the brain
EP
tissue oxygenation-based ORx. This phenomenon of a pharmacologically induced worsening in the autoregulation indices must be distinguished from the spontaneous
AC C
impairment of cerebrovascular autoregulation (increase in the indices) associated with DCI and an unfavorable outcome in SAH.2,24,25 Jaeger et al. demonstrated that in patients with high-grade SAH, a consistent increase in the ORx values was predictive of the development of cerebral infarction.2 None of the patients in our group exhibited a pathologically increased PRx or ORx prior to CIN treatment; thus, autoregulation was intact. The monitoring of autoregulation indices may therefore be a way to trace the therapeutic effectiveness of nimodipine treatment on the precapillary resistance vessels. 17
Hockel et al.
ACCEPTED MANUSCRIPT
The potential adverse effects of long-term CIN treatment must be thoroughly considered. These are prolonged analgosedation, a strict supine position without positioning maneuvers to avoid catheter dislocation, heparinization, and an increased
RI PT
volume load as a result of the continuous nimodipine infusion. These potentially negative effects seem acceptable for a cohort of selected, high-risk, younger SAH patients.
SC
Heparinization is of special concern in these patients. The necessity of therapeutic anticoagulation is undisputed to prevent occlusion of or thromboembolism elicited by
M AN U
the intra-arterial catheter in the ICA. Although heparinization may have an additional positive effect via prevention of micro-thromboembolism, which is thought to contribute to secondary ischemia following SAH,18 hemorrhagic complications can occur as our data demonstrate.
TE D
We assume that the aneurysm re-rupture we experienced occurred independent of the applied anticoagulation because this is a primary clip occlusion failure. However, the severity of any postoperative hemorrhagic complication is affected by
EP
anticoagulation, and we cannot exclude a contribution to the final outcome in these patients. Therefore, we exclude meanwhile patients who underwent previous
AC C
intracranial surgery within a 7-day period, e.g., craniotomy for aneurysm clipping, or those who have any other contraindication for continuous anticoagulation (see algorithm for patient selection, Appendix). CIN treatment must be terminated when a surgical intervention becomes necessary, e.g., a decompressive craniectomy for ICP control, as observed in one case. Therefore, all necessary surgical procedures, such as the implantation of ICP probes or EVD catheters, must be accomplished prior to the initiation of CIN. Nimodipine infusion, which leads to vasodilatation and thus an increase in cerebral blood volume, 18
Hockel et al.
ACCEPTED MANUSCRIPT
did not significantly increase the ICP in our patient cohort. Nevertheless, we exclude patients with massive brain swelling, an increased ICP or other indices of compromised intracranial compliance from CIN treatment.
RI PT
Patients who may benefit from long-term intra-arterial nimodipine treatment must be carefully identified considering the invasiveness of the technique and the described potential hazards. Although the long-term outcome in our cohort was homogeneously
SC
favorable (median GOS of 5 at 3 months), over-treatment for some patients cannot be excluded. DCI was clinically diagnosed in only three of ten patients; in the other
M AN U
patients, we strongly assumed, that the severity of macrovasospasm and the perfusion deficits detected by CTP would have correlated with clinical DCI. We demonstrated that the conduction of continuous intra-arterial infusion was feasible beyond two weeks; however, a superior effect on neurological outcome and
TE D
the prevention of cerebral ischemia compared with standard treatment cannot be demonstrated from this preliminary retrospective non-controlled cohort. Therefore, a controlled larger study with defined inclusion criteria that compare basic treatment
EP
protocols (hemodynamics and nimodipine) to more aggressive interventional regimens, i.e., bolus and continuous intra-arterial nimodipine treatment, with respect
AC C
to neurological outcome, the incidence of cerebral infarction and the complication rate remains necessary.
Conclusion CIN treatment may prevent the development of secondary ischemia within a window of 14 days in patients with prolonged and severe macrovasospasm associated with CT perfusion deficits or secondary clinical deterioration prior to sedation and intubation. When patients can be identified, in which conservative therapy and single19
Hockel et al.
ACCEPTED MANUSCRIPT
time bolus interventions with intra-arterial nimodipine fail to achieve a sustained improvement of cerebral perfusion, CIN treatment appears to be justified as a rescue therapy in well-selected cases based on these preliminary data. A randomized controlled study with a larger patient cohort is necessary to confirm the overall
RI PT
efficacy of CIN in the prevention of DCI and to further characterize the associated risks.
SC
Conflict of interest
AC C
EP
TE D
M AN U
None.
20
Hockel et al.
ACCEPTED MANUSCRIPT Reference List 1.
Crowley RW, Medel R, Dumont AS, Ilodigwe D, Kassell NF, Mayer SA, Ruefenacht D, Schmiedek P, Weidauer S, Pasqualin A, Macdonald RL. Angiographic vasospasm is strongly correlated with cerebral infarction after
2.
RI PT
subarachnoid hemorrhage. Stroke. 2011;42:919–923.
Jaeger M, Soehle M, Schuhmann MU, Meixensberger J. Clinical significance of impaired cerebrovascular autoregulation after severe aneurysmal
3.
SC
subarachnoid hemorrhage. Stroke. 2012;43:2097–2101.
Bederson JB, Connolly ES, Batjer HH, Dacey RG, Dion JE, Diringer MN,
M AN U
Duldner JE, Harbaugh RE, Patel AB, Rosenwasser RH. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke. 2009;40:994–1025. Dorhout Mees SM, Rinkel GJE, Feigin VL, Algra A, van den Bergh WM,
TE D
4.
Vermeulen M, van Gijn J. Calcium antagonists for aneurysmal subarachnoid hemorrhage. Stroke. 2008;39:514–515. Pickard JD, Murray GD, Illingworth R, Shaw MD, Teasdale GM, Foy PM,
EP
5.
Humphrey PR, Lang DA, Nelson R, Richards P. Effect of oral nimodipine on
AC C
cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. BMJ. 1989;298:636–642.
6.
Raabe A, Beck J, Keller M, Vatter H, Zimmermann M, Seifert V. Relative importance of hypertension compared with hypervolemia for increasing cerebral oxygenation in patients with cerebral vasospasm after subarachnoid hemorrhage. J Neurosurg. 2005;103:974–981.
7.
Biondi A, Ricciardi GK, Puybasset L, Abdennour L, Longo M, Chiras J, Van Effenterre R. Intra-arterial nimodipine for the treatment of symptomatic cerebral 21
Hockel et al.
ACCEPTED MANUSCRIPT
vasospasm after aneurysmal subarachnoid hemorrhage: preliminary results. AJNR Am J Neuroradiol. 2004;25:1067–1076. 8.
Feng L, Fitzsimmons B-F, Young WL, Berman MF, Lin E, Aagaard BDL, Duong H, Pile-Spellman J. Intraarterially administered verapamil as adjunct therapy for
RI PT
cerebral vasospasm: safety and 2-year experience. AJNR Am J Neuroradiol. 2002;23:1284–1290. 9.
Hänggi D, Turowski B, Beseoglu K, Yong M, Steiger HJ. Intra-arterial
SC
nimodipine for severe cerebral vasospasm after aneurysmal subarachnoid
hemorrhage: influence on clinical course and cerebral perfusion. AJNR Am J
10.
M AN U
Neuroradiol. 2008;29:1053–1060.
Hoelper BM, Hofmann E, Sporleder R, Soldner F, Behr R. Transluminal balloon angioplasty improves brain tissue oxygenation and metabolism in severe vasospasm after aneurysmal subarachnoid hemorrhage: case report.
11.
TE D
Neurosurgery. 2003;52:970–976.
Vajkoczy P, Horn P, Bauhuf C, Munch E, Hubner U, Ing D, Thome C, PoecklerSchoeninger C, Roth H, Schmiedek P. Effect of intra-arterial papaverine on
EP
regional cerebral blood flow in hemodynamically relevant cerebral vasospasm. Stroke. 2001;32:498–505.
Doukas A, Petridis AK, Barth H, Jansen O, Mehdorn HM. Continuous intra-
AC C
12.
arterial infusion of nimodipine at the onset of resistant vasospasm in aneurysmal subarachnoidal haemorrhage. Technical report. Neurol Res. 2011;33:290–294.
13.
Mayer TE, Dichgans M, Straube A, Birnbaum T, Müller-Schunk S, Hamann GF, Schulte-Altedorneburg G. Continuous intra-arterial nimodipine for the treatment of cerebral vasospasm. Cardiovasc Intervent Radiol. 2008;31:1200–1204.
22
Hockel et al.
ACCEPTED MANUSCRIPT 14.
Musahl C, Henkes H, Vajda Z, Coburger J, Hopf N. Continuous local intraarterial nimodipine administration in severe symptomatic vasospasm after subarachnoid hemorrhage. Neurosurgery. 2011;68:1541–1547.
15.
Ott S, Jedlicka S, Wolf S, Peter M, Pudenz C, Merker P, Schürer L, Lumenta
RI PT
CB. Continuous selective intra-arterial application of nimodipine in refractory cerebral vasospasm due to aneurysmal subarachnoid hemorrhage. BioMed Res Int. 2014;2014:970741.
Wolf S, Martin H, Landscheidt JF, Rodiek SO, Schürer L, Lumenta CB.
SC
16.
Continuous selective intraarterial infusion of nimodipine for therapy of
17.
M AN U
refractory cerebral vasospasm. Neurocrit Care. 2010;12:346–351. Jaeger M, Schuhmann MU, Soehle M, Meixensberger J. Continuous assessment of cerebrovascular autoregulation after traumatic brain injury using brain tissue oxygen pressure reactivity. Crit Care Med. 2006;34:1783–1788. Sehba FA, Bederson JB. Mechanisms of acute brain injury after subarachnoid
TE D
18.
hemorrhage. Neurol Res. 2006;28:381–398. 19.
Vergouwen MD, Etminan N, Ilodigwe D, Macdonald RL. Lower incidence of
EP
cerebral infarction correlates with improved functional outcome after aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab.
20.
AC C
2011;31:1545–1553.
Etminan N, Vergouwen MD, Ilodigwe D, Macdonald RL. Effect of pharmaceutical treatment on vasospasm, delayed cerebral ischemia, and clinical outcome in patients with aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Cereb Blood Flow Metab. 2011;31:1443–1451.
21.
Choi HA, Ko S, Chen H, Gilmore E, Carpenter AM, Lee D, Claassen J, Mayer SA, Schmidt JM, Lee K, Connelly ES, Paik M, Badjatia N. Acute effects of 23
Hockel et al.
ACCEPTED MANUSCRIPT
nimodipine on cerebral vasculature and brain metabolism in high grade subarachnoid hemorrhage patients. Neurocrit Care. 2012;16:363–367. 22.
Canova D, Roatta S, Micieli G, Bosone D. Cerebral oxygenation and haemodynamic effects induced by nimodipine in healthy subjects. Funct
23.
RI PT
Neurol. 2012;27:169–176.
Czosnyka M, Smielewski P, Kirkpatrick P, Laing RJ, Menon D, Pickard JD. Continuous assessment of the cerebral vasomotor reactivity in head injury.
24.
SC
Neurosurgery. 1997;41:11–19.
Bijlenga P, Czosnyka M, Budohoski KP, Soehle M, Pickard JD, Kirkpatrick PJ,
M AN U
Smielewski P. “Optimal cerebral perfusion pressure” in poor grade patients after subarachnoid hemorrhage. Neurocrit Care. 2010;13:17–23. Rasulo FA, Girardini A, Lavinio A, De Peri E, Stefini R, Cenzato M, Nodari I, Latronico N. Are optimal cerebral perfusion pressure and cerebrovascular
TE D
autoregulation related to Long-term outcome in patients with aneurysmal
EP
subarachnoid hemorrhage? J Neurosurg Anesthesiol. 2012;24:3–8.
AC C
25.
24
Hockel et al.
ACCEPTED MANUSCRIPT Figure legends
Figure 1: Time course of the flow velocities via TCD in the right and left MCA. When vasospasms were diagnosed, the flow velocities had increased to approximately 150
SC
day 10 (p<0.05) compared with the pre-CIN TCD exams.
RI PT
cm/s, median value. After CIN induction, the velocities significantly decreased until
Figure 2A and B: Time course of changes in the PRx (Fig. 2A) and ORx (Fig. 2B),
M AN U
displayed as 12-hour means before and after the induction and before and after the termination of CIN treatment. Both indices indicate a close time relationship with CIN treatment, and the autoregulation parameters were significantly altered during
AC C
EP
TE D
treatment, p<0.05.
25
ACCEPTED MANUSCRIPT
Algorithm for selection patients for CIN
within 48 hours
Monitoring
ICP, pbrO2, microdialysis
Screening
neuro exams, TCD, CTA/CTP (5, 10, 15thd)
Prophylaxis
CPP > 70mmHg, IV NDP
confirm macrovasospasm
angiography
consider endovasc. treatment
IA bolus NDP (+ increase IV NDP)
24 hours
CTA/CTP, angiography
TE D
repeat diagnostics
M AN U
yes
SC
neurological deterioration, radiographic DCI
Exclusion criteria for CIN: • major territorial infarction or intracerebral hemorrhage • previous intracranial surgery, i.e. craniotomy, aneurysm clipping within 5-7days • increased ICP, significant brain swelling • any contraindication for continuous anticoagulation
RI PT
aneurysm occlusion CSF drainage
SAH
consider CIN if • continued radiographic DCI • absent exclusion criteria
• IA continuous NDP (0.4mg/h per catheter) • Heparin PTT 60-80 sec
AC C
EP
repeat IA bolus NDP if • exclusion criteria for CIN • repeat diagnostics every 24h
Monitoring/Screening
IA bolus NDP (+ increase IV NDP Continue IV NDP repeat diagnostics
ACCEPTED MANUSCRIPT
Table 1: Patient characteristics. sex (f/m)
age (yrs.)
Hunt & Hess Fisher aneurysm treatment grade° grade° location (clip/coil)
EVdrain (y/n)
DCI on diagnosis of admission vasospasm (y/n) (days post SAH)
1
f
46
4
4
MCA left
clip
y
n
5
2
f
29
3
3
AComA
clip
y
n
5
3
f
46
3
4
PComA
coil
y
n
8
4
f
38
3
3
AComA
coil
n
y
6
5
f
27
2
4
MCA right clip
n
n
4
6
m
45
3
2
AComA
clip
n
y
9
7
f
47
3
3
AComA
clip
y
n
7
8
f
45
5
3
ACI right
coil
y
n
4
9
m
25
5
4
AComA
clip
10
f
46
3
4
ICA left
coil
SC
RI PT
Patient No°
y
n
5
y
n
4
AC C
EP
TE D
M AN U
MCA – middle cerebral artery, AComA – anterior communicating artery, PComA – posterior communicating artery, ICA – internal carotid artery, time – days after the initial hemorrhage; ev -extraventricular.
ACCEPTED MANUSCRIPT Table 2: Patient outcome statistics. single-time start duration intra-artrial of CIN of CIN infarcts nimodipine (day) (days) on CT complications
GOS at GOS at discharge 3 months
1
gen.
2
7
15
y
-
4
4
2
ACA/ICA bilat
-
5
12
n
-
5
5
3
MCA./ICA bilat. -
8
13
y
-
4
-
4
ACA/ICA bilat.
3
9
11
y
-
4
5
5
MCA/ICA right
4
10
14
y
-
6
ACA/MCA bilat. -
10
14
n
7
ACA/MCA bilat. -
7
9
8
MCA bilat.
1
5
9
gen.
-
10
ACA/MCA bilat. -
4
5
cath. disl.
3
-
n
-
4
5
14
n
GI bleed
4
-
9
11
n
cath. disl.
4
5
4
11
n
SC
RI PT
Patient site of No° vasospasm
5
HIT
5
AC C
EP
TE D
M AN U
gen. – generalized, MCA – middle cerebral artery, ACA – anterior cerebral artery, ICA – internal carotid artery, bilat. – bilateral, cath disl. – catheter dislocation, time – days after the initial hemorrhage.
ACCEPTED MANUSCRIPT Table 3: CT-angiographic grading of vasospasm. grading, total n=10
prior CIN
during CIN*
after CIN
high grade vasospasm
10
2
0
moderate vasospasm
0
5
2
light vasospasm
0
1
3
no vasospasm
0
0
5
AC C
EP
TE D
M AN U
SC
RI PT
*not every patient received CTA under CIN (n=9)
ACCEPTED MANUSCRIPT Table 4: Physiological parameters during CIN. start monit.
prior CIN
CIN day 0
CIN day 5
CIN day 10
ABP (mmHg)
89 ± 2
90 ± 2
92 ± 2
92 ± 3
90 ± 2
CPP (mmHg)
77 ± 2
79 ± 2
80 ± 1
83 ± 2
83 ± 2
ICP (mmHg)
11 ± 1
10 ± 1
12 ± 1
9±1
8±1
pbrO2 (mmHg)
29.1 ± 3.1
27.1 ± 3.1
23.8 ± 3.5
27.7 ± 4.7
32.7 ± 4.8
paO2 (mmHg)
119 ± 1.7
111.8 ± 4.7
111.6 ± 7.6
113.1 ± 5.3
114.2 ± 4.4
paCO2 (mmHg)
37.6 ± 0.6
36.0 ± 0.6
37.8 ± 1.3
37.9 ± 0.5
AC C
EP
TE D
M AN U
SC
24 hour mean ± SEM; p>0.05.
RI PT
variable
37.1 ± 0.7
ACCEPTED MANUSCRIPT
v MCA (cm/s)
mean ± SEM; n=10; * p<0.05
*
SC M AN U TE D
150
EP
100
50
AC C
v TCD (cm/s)
200
RI PT
250
0
Sc
r
n ee
ing r CIN a y 0 d io Npr I C
2 N
CI
a -d
y5
3 N CI
-
y da
10
ACCEPTED MANUSCRIPT
mean ± SEM; n=10; * p<0.05
RI PT
0.3
SC
*
*
TE D
M AN U
0.2
EP
0.1
AC C
pressure reactivity index (PRx)
PRx
0.0
CIN infusion -0.1 12 h
12 h
12 h
12 h
12 h
12 h
12 h
12 h
ACCEPTED MANUSCRIPT
mean ± SEM; n=10; * p<0.05
*
SC
RI PT
0.3
M AN U
0.2
EP
TE D
0.1
AC C
oxygen reactivity index (ORx)
ORx
0.0
CIN infusion -0.1 12 h
12 h
12 h
12 h
12 h
12 h
12 h
12 h
ACCEPTED MANUSCRIPT Highlights We investigated a patient series with severe refractory macrovasospasm after subarachnoid hemorrhage Long-term continuous intra-arterial nimodipine treatment as rescue therapy was retrospectively evaluated for feasibility, safety and effectiveness
RI PT
Continuous improvement of macrovasospasm severity and brain tissue oxygenation were found after 12 days (mean) of treatment
AC C
EP
TE D
M AN U
SC
Associated risks have to be weighed up thoroughly against potential benefits
S1878-8750(15)01704-0
DOI:
10.1016/j.wneu.2015.11.081
Reference:
WNEU 3497
To appear in:
World Neurosurgery
Received Date: 1 July 2015 Revised Date:
23 November 2015
Accepted Date: 24 November 2015
Please cite this article as: Hockel K, Diedler J, Steiner J, Birkenhauer U, Danz S, Ernemann U, Schuhmann MU, Long-term, continuous intra-arterial nimodipine treatment of severe vasospasm following aneurysmal subarachnoid hemorrhage, World Neurosurgery (2016), doi: 10.1016/ j.wneu.2015.11.081. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Long-term,
continuous
intra-arterial
nimodipine
treatment
of
severe
vasospasm following aneurysmal subarachnoid hemorrhage
Konstantin Hockel, MD1
RI PT
Jennifer Diedler, MD, PhD1 Jochen Steiner, MD1 Ulrich Birkenhauer, MD1 Sören Danz, MD2 Ulrike Ernemann, MD, PhD2
SC
Martin U. Schuhmann, MD, PhD1 1
M AN U
Department of Neurosurgery, University Hospital Tübingen, University of Tübingen;
2
Diagnostic and Interventional Neuroradiology, Department of Radiology, University
Corresponding author:
EP
Konstantin Hockel
TE D
Hospital Tübingen, University of Tübingen, Germany.
Department of Neurosurgery
University Hospital Tübingen
AC C
Hoppe-Seyler-Str. 3 72076 Tübingen Germany
Phone: +49 7071 2986679 Fax: +49 7071 5245 Email: [email protected]
Key words: subarachnoid hemorrhage, vasospasm, nimodipine, brain tissue oxygen, cerebral autoregulation, continuous intra-arterial nimodipine, DCI.
Hockel et al.
ACCEPTED MANUSCRIPT Abbreviations CIN – continuous intra-arterial nimodipine CPP – cerebral perfusion pressure CTA – computed tomography angiography CTP – computed tomography perfusion
RI PT
DCI – delayed cerebral ischemia DSA – digital subtraction angiography EVD – external ventricular drainage GOS – Glasgow outcome scale
SC
ICA – internal carotid artery ICP – intracranial pressure
MCA – middle cerebral artery ORx – oxygen reactivity index PRx – pressure reactivity index SAH – subarachnoid hemorrhage
M AN U
MAP – mean arterial pressure
AC C
EP
TE D
TCD – transcranial Doppler sonography
2
Hockel et al.
ACCEPTED MANUSCRIPT Abstract Background:
Secondary
vasospasm
and
disturbances
in
cerebrovascular
autoregulation are associated with the development of delayed cerebral ischemia (DCI) following aneurysmal subarachnoid hemorrhage (SAH). An intra-arterial
RI PT
application of nimodipine has been shown to increase the vessel diameter, although this effect is transient. The feasibility of long-term, continuous, intra-arterial nimodipine (CIN) treatment and its effects on macrovasospasm, autoregulation
SC
parameters and outcome were evaluated in patients with refractory severe macrovasospasm.
M AN U
Methods: Ten patients were included with refractory macrovasospasm despite bolus nimodipine application (n=4) or with primary severe vasospasm (n=6). The patients were assessed with continuous multimodal neuromonitoring (MAP, ICP, CPP, pbrO2), daily TCD exams and CT-angiography/perfusion (CTA, CTP). Autoregulation indices,
TE D
the pressure reactivity index (PRx) and oxygen reactivity index (ORx) were calculated. Indwelling microcatheters were placed in the extracranial internal carotid
EP
arteries and 0.4 mg nimodipine was continuously infused at 50 ml/hour. Results: The duration of CIN treatment ranged from 9-15 days. During treatment ICP
AC C
remained stable, TCD flow velocity decreased and pbrO2 improved by 37%. Macrovasospasm, as assessed via CTA, had improved (n=5) or disappeared (n=5) at the end of treatment. Cerebrovascular autoregulation according to the PRx and ORx significantly worsened during treatment. All patients showed a favorable outcome (median GOS 5) at 3 months. Conclusions: In well-selected patients with prolonged severe macrovasospasm, continuous intra-arterial nimodipine treatment can be applied as a rescue therapy with relative safety for more than 2 weeks to prevent secondary cerebral ischemia.
3
Hockel et al.
ACCEPTED MANUSCRIPT
The induced impairment of cerebrovascular autoregulation during treatment seems to
AC C
EP
TE D
M AN U
SC
RI PT
have no negative effects.
4
Hockel et al.
ACCEPTED MANUSCRIPT Introduction
Delayed cerebral ischemia (DCI) is the secondary reason for an unfavorable outcome following aneurysmal subarachnoid hemorrhage (SAH). Secondary macrovascular spasm and disturbances in the cerebrovascular autoregulation,
RI PT
located in the precapillary microvessels, have been identified as associated risk factors for DCI.1,2 In clinical practice, the diagnosis and subsequent therapy of angiographically detected macrovasospasm remain keystones in DCI prevention,3
SC
especially in comatose or sedated patients where no feedback on neurological function can be obtained.
M AN U
Limited therapeutic strategies have been shown to effectively improve clinical outcome and mortality. Oral nimodipine and hemodynamic augmentation are wellaccepted
basic
treatment
regimens.4-6
In
refractory
cases
of
severe
macrovasospasm, endovascular interventions have been implemented as alternative
TE D
options; however, the scientific evidence remains scarce.
The most commonly used techniques are mechanical, e.g., transluminal balloon angioplasty, or pharmacological vasodilatation, e.g., papaverine and calcium channel
EP
blockers, such as nimodipine or verapamil.7-11 Rebound ischemia, the need for repeated interventions and the hazards of patient transport remain significant risk
AC C
factors.9,11
To overcome the drawbacks of single-time interventions, the continuous intra-arterial application of vasodilators, such as nimodipine, using endovascular microcatheters in the internal carotid artery (ICA) has been suggested. Singular reports in cases of refractory macrovasospasm have demonstrated the technical feasibility and the potential to improve a compromised cerebral perfusion.12-16 To date, nimodipine infusion has only been applied for a limited number of days; however, the critical period of macrovasospasm may last for up to two weeks.7,9 5
Hockel et al.
ACCEPTED MANUSCRIPT
In this pilot study, we report our experience with long-term, continuous intra-arterial nimodipine infusion (CIN) in thirteen patients following aneurysmal SAH with refractory severe macrovasospasm. We evaluated the feasibility and safety of longterm CIN treatment as primary end points. The effects on macrovasospasm, brain
RI PT
tissue oxygenation and cerebrovascular autoregulation indices, including the pressure reactivity index (PRx) and oxygen reactivity index (ORx), were assessed as secondary end points. Furthermore, patient outcome at discharge and at 3 months is
SC
reported.
M AN U
Patients and Methods
Patient inclusion and general management
Thirteen patients with aneurysmal SAH were selected for CIN within a 10-month period. Sixty patients with a diagnosis of SAH were treated in our department during
TE D
this time period.
This study comprises a retrospective data analysis of a cohort of patients with refractory macrovasospasm who were treated on an individual basis with intra-
EP
arterial continuous nimodipine application. In this specific cohort of SAH patients, who were threatened by cerebral infarction as a result of severe refractory
AC C
macrovasospasm, we performed an “individual treatment attempt” with CIN application via indwelling microcatheters. This protocol was not performed in a prospective study setting with a specific research question; instead, it comprised a last tier treatment option in individual cases. Thus, IRB approval was not required from our institute. However, IRB approval was obtained for computerized neuromonitoring and data collection for the retrospective data analysis of all patients in the intensive care unit (ICU). The patients’ relatives provided their informed consent for the individual treatment attempt using CIN. 6
Hockel et al.
ACCEPTED MANUSCRIPT
Following admission to the ICU, diagnostic imaging, including computed tomography angiography (CTA) and digital subtraction angiography (DSA), was performed. The patients underwent subsequent aneurysm occlusion via microsurgical clipping or
RI PT
endovascular coiling. In cases of acute hydrocephalus, an extraventricular drain was inserted within 24 hours of admission. If not primarily intubated, sedation (midazolam, sufentanyl) and mechanical ventilation were induced prior to aneurysm treatment.
SC
Further intensive care management was conducted according to our current ICU standards. Mechanical ventilation was guided to maintain the arterial pO2 at 110 ± 5
M AN U
mmHg and the arterial pCO2 between 35-40 mmHg. The fluid balance was aimed at normovolemia. Catecholamines (noradrenalin) were titrated in cases of arterial hypotension despite normovolemia to ensure cerebral perfusion pressures of ≥ 70 mmHg. Pre-emptive intravenous (iv) nimodipine was not routinely used in the
TE D
patients who required moderate to high doses of catecholamines; however, in normotensive or hypertensive patients, iv nimodipine was administered.
EP
Neuromonitoring and assessment of cerebral autoregulation All patients who could not be extubated within 24 hour after aneurysm treatment
AC C
(because of a primary poor initial clinical grade, significant intraoperative swelling or failure of the postoperative / postinterventional weaning attempt) received multimodal neuromonitoring, including intracranial pressure (ICP) monitoring, cerebral perfusion pressure calculation (CPP) and a brain tissue oxygen tension probe (pbrO2). A Neurovent-PTO probe (Raumedic AG, Helmbrechts, Germany) was used for the assessment of the ICP, pbrO2 and brain temperature; the probe was inserted into the frontal white matter of the hemisphere that was predominantly affected by the hemorrhage via a one-lumen bolt. The mean arterial pressure (MAP) was 7
Hockel et al.
ACCEPTED MANUSCRIPT
continuously monitored using a catheter inserted into the radial artery with the transducer referenced to the foramen of Monro. The ICP, MAP and pbrO2 signals were continuously recorded on a bedside-mounted device (Datalogger MPR2, Raumedic AG, Helmbrechts, Germany) and transmitted to
RI PT
the bedside hospital monitoring system. The three parameters were also digitally sampled at a rate of 100 Hz and forwarded to a notebook PC that was running ICM+ software (Cambridge Enterprise, Cambridge, UK). The CPP was calculated as the
SC
difference between the MAP and the ICP.
ICM+ software was used for both the online display of data and the retrospective
M AN U
analysis of the recorded monitoring parameters. For the retrospective analysis, the ICP and MAP data were subjected to manual artifact detection and removal. All parameters were calculated using a one-minute moving window. The parameters of cerebrovascular autoregulatory capacity, i.e., the PRx and ORx, were calculated as
TE D
previously described.17 In brief, the PRx was evaluated as the moving Pearson correlation coefficient between the averaged ICP and MAP calculated over the moving window length of 5 minutes. The ORx was evaluated as the moving Pearson
EP
correlation coefficient between the pbrO2 and CPP over an average period of 1 hour. The PR and OR indices may vary between -1 and 1. Intact cerebral autoregulation
AC C
can be assumed when the index values are close to zero, which indicates that no correlation between the ICP and MAP or the pbrO2 and CPP exists. Transcranial Doppler (TCD) examinations were performed daily with a hand-held 2MHz probe (EME Companion, Nicolet Vascular Ltd., Madison, USA) to assess the maximum blood flow velocities of the major intracranial arteries. For subsequent analyses, the mean values of the maximum blood flow velocity (cm/s) in the MCA on both sides were recorded.
8
Hockel et al.
ACCEPTED MANUSCRIPT Detection of cerebral macrovasospasm
Screening for macrovasospasm in sedated patients was performed via daily TCD examination. Neuromonitoring data were not used as a screening tool for macrovasospasm or possible DCI. In addition, the patients received diagnostic CTA
RI PT
and CTP imaging as a routine screening examination on day 5 after SAH or at the time point when an increase in the mean TCD flow velocity greater than 120 cm/s was identified for the first time. In cases of significant macrovasospasm according to
SC
the CTA/CTP criteria, i.e., significant vessel narrowing greater than 60% plus a respective perfusion deficit, conventional diagnostic angiography was performed to
M AN U
confirm macrovasospasm. In case of severe macrovasospasm (significant vessel narrowing greater than 60%), an intra-arterial bolus application of nimodipine, 3 mg over 30 minutes, was performed for each involved vascular territory in each patient. In awake patients which developed clinical signs of DCI (neurological deterioration)
TE D
conservative treatment was escalated according to our ICU standard protocol: the elevation and maintenance of a CPP ≥ 70 mmHg combined with iv nimodipine. In case neurological symptoms did not disappear, conventional digital angiography was
EP
performed. In case of macrovasospasm, an intra-arterial bolus application of nimodipine, 3 mg over 30 minutes, was performed for each involved vascular territory
AC C
in each patient.
Induction of CIN treatment CIN treatment was initiated if 1) a primary, severe macrovasospasm with respective perfusion deficits via CTP showed an insufficient response to the first bolus dose of nimodipine (same intervention) or 2) a recurrent vasospasm was identified via followup angiography (subsequent day) after the intra-arterial nimodipine bolus application and/or CT imaging showed evidence of new ischemic lesions. 9
Hockel et al.
ACCEPTED MANUSCRIPT
Over a femoral artery sheath, one or two microcatheters (14-Prowler Microcatheter, Cordis Corp., Hialeah, USA) were placed in the extracranial segment of the feeding artery of each involved vascular territory, i.e., typically the bilateral ICA or vertebral artery (VA, one patient). Intra-arterial nimodipine infusion was immediately initiated at
RI PT
0.4 mg/catheter/hour, with an infusion rate of 50 ml/hour. A comparable dose of nimodipine has been previously recommended.14 In our group, the nimodipine dosage ranged between 0.2-0.6 mg/catheter/hour. An infusion rate of 50 ml/hour was
SC
selected to prevent catheter obstruction and thrombus apposition. From the time point of catheter placement, of positioning maneuvers of the patient were reduced to
M AN U
a minimum. To minimize the risk of thromboembolic complications as a result of indwelling arterial catheters, anticoagulation was facilitated with unfractioned heparin to a partial thromboplastin time (PTT) of 60-80 seconds.
Following CIN induction, further treatment was continued in the ICU. Multimodal
TE D
neuromonitoring and daily TCD examinations were continued. During CIN treatment, no routine imaging studies, i.e., CT scan or angiography, were performed unless otherwise indicated by the individual course.
EP
There was no predefined endpoint for CIN treatment. The termination of CIN was decided on an individual basis, typically when the pbrO2 and TCD values indicated a
AC C
sustained improvement of brain perfusion or resolution of macrovasospasm, respectively. CIN was then paused for at least 6 hours, and subsequent CTA and CTP studies performed to confirm resolution of macrovasospasm. In this case and if CTA of the extracranial carotid artery indicated no signs of thrombus apposition, the indwelling arterial catheters were removed.
Statistical analysis
10
Hockel et al.
ACCEPTED MANUSCRIPT
For the final analysis of the physiological variables, i.e., the blood gases, ICP, MAP, CPP, and pbrO2, 24 hour mean values were calculated and presented for the days of interest prior to and during CIN treatment. Regarding the autoregulation indices PRx and ORx, a 12-hour window was evaluated, which focused on the quantitative
mean +/- standard error, if not otherwise indicated.
RI PT
changes during the start and end of CIN treatment. The data are represented as the
Statistical analysis was performed using SPSS statistical software (SPSS 21.0,
SC
SPSS Inc., Chicago, USA). The data were analyzed using paired t-tests and
M AN U
Wilcoxon signed rank tests. Statistical significance was assumed at p<0.05.
Results Clinical course
In two of the thirteen patients, CIN treatment was prematurely terminated after 2 and
TE D
8 days, respectively. In one patient, a postoperative hemorrhagic complication (intracerebral hemorrhage) occurred after a decompressive craniectomy was performed 3 days prior to the initiation of CIN. Another patient underwent a
EP
decompressive craniectomy on day 8 of CIN for re-hemorrhage, which resulted from an aneurysm previously occluded by a clip. The outcome of both patients was GOS 2
AC C
at discharge. In one additional patient, pbrO2 monitoring was not available prior to and during the first days of CIN treatment. These three patients were excluded from the final analysis of CIN effects on pbrO2, autoregulation indices and outcome. The distribution of gender, age, WFNS and Fisher grade, aneurysm location and treatment is shown in Table 1.
Seven of the ten included patients were admitted to the hospital within 48 hours after the initial hemorrhage. Three of the ten patients were awake after aneurysm 11
Hockel et al.
ACCEPTED MANUSCRIPT
treatment and presented with clinical symptoms of DCI, e.g., hemiparesis, aphasia. Subsequent angiography confirmed macrovasospasm. For the remaining seven patients, no clinical evaluation was feasible because of their poor initial clinical grade
RI PT
and continued sedation and ventilation after aneurysm treatment.
A significant angiographic vasospasm was identified after a mean of 5.7 days (days 4-9) following SAH (Table 2). The vasospasm was unilaterally localized in one case,
SC
bilaterally localized in the anterior circulation in seven patients and distributed in a generalized pattern in two patients. The vasospasm duration had a mean of 14.4
M AN U
days (10-20 days). In all cases, the vasospasm was classified as severe according to angiography criteria.
Four patients underwent an initial singular or repeat intra-arterial bolus treatment (mean 2.5 interventions) prior to the initiation of CIN (Table 2). Three of those four
TE D
patients developed new or further ischemic lesions visible on CT. In these four cases and in the six remaining patients with primarily severe macrovasospasm unresponsive to bolus treatment CIN treatment was initiated at a mean of 7.4 days
EP
(4-10 days) after SAH and continued for a mean of 12.0 days (9-15 days). Seven patients received the starting dosage of 0.4 mg/catheter/hour for the entire CIN
AC C
period. According to the behavior of the TCD flow velocities, nimodipine was increased to 0.6 mg/hour in two patients, whereas it was reduced to 0.2 mg/hour in one patient during the course of the treatment. During CIN, two patients developed further localized cerebral insults.
The noradrenalin dosage changed non-significantly from 0.20 ± 0.46 µg/kg/min to 0.27 ± 0.52 µg/kg/min during CIN treatment (p>0.05). The arterial blood pressure was not significantly affected by CIN treatment, and a CPP ≥ 75 mmHg was 12
Hockel et al.
ACCEPTED MANUSCRIPT
maintained (Table 4). CIN treatment led to an increased volume input by approximately 2.5 liters (50 ml/catheter/hour).
During the course of CIN treatment, CTA was performed after the nimodipine infusion
RI PT
was ceased for at least 6 hours at a time point as described above. Treatment was continued in eight of ten patients after the first attempt to stop CIN treatment, due to the presence of remaining significant vasospasm.
SC
The qualitative evaluation of the vasospasm degree, as determined via CTA by a dedicated neuroradiologist, indicated a stepwise decrease in the vasospasm severity
M AN U
until the end of CIN treatment (Table 3). Remaining vessel constrictions were considered of no or minor relevance for cerebral perfusion.
Catheter dislocation was detected via CTA in two of nineteen catheters at the end of CIN treatment. Both catheters were most likely proximally dislocated into the aortic
TE D
arch, which was not visible on CTA. The duration of the dislocation was unknown. In two other patients, CIN was terminated on days 12 and 14, respectively, because of
gastrointestinal
bleeding
and
heparin-induced
thrombocytopenia
(HIT),
(Table 2).
EP
complications that were presumably related to the applied continuous anticoagulation
AC C
Following the successful termination of CIN treatment, sedation was discontinued, and weaning from the respirator was achieved. At discharge, minor focal neurological deficits were detectable in some patients; however, the majority of the patients were able to ambulate independently, as indicated by a median GOS of 4 (range 3-5). At three months, the median GOS was 5 (range 4-5) (Table 2).
Neuromonitoring parameters and cerebral autoregulation
13
Hockel et al.
ACCEPTED MANUSCRIPT
TCD examinations of the MCA demonstrated an increase in the blood flow velocities to a mean value of 144 ± 9 cm/s prior to CIN treatment. Following the induction of CIN, the flow velocities continuously decreased over days. The TCD velocities significantly improved to a median of 99 ± 7 cm/s on day 10 (Fig. 1, p<0.05 compared
RI PT
with the time prior to CIN treatment); however, the velocities did not decrease below a threshold of 120 cm/s in any case.
SC
The total monitoring time according to ICM+ software was 4,856 hours. Premature failure of the pbrO2 probes during CIN treatment occurred in four patients, which
M AN U
occurred prior to day 10 of CIN treatment in two patients. The CPP, paO2 and paCO2 were maintained within the desired limits during the entire period of CIN treatment (Table 4). The mean ICP was 10 ± 1 mmHg prior to CIN treatment and changed nonsignificantly to a mean of 11 ± 1 mmHg during intra-arterial nimodipine infusion. The
TE D
mean brain tissue oxygen level decreased to 23.8 ± 3.5 mmHg on the day CIN treatment was initiated. During nimodipine infusion, the pbrO2, which represents a surrogate marker of cerebral blood flow, increased within 24 hours in all but 2
EP
patients; furthermore, it continuously increased to 32.7 ± 4.8 on day 10, which corresponds to a 37% improvement in pbrO2 during the course of CIN treatment
AC C
(p>0.05).
The indices of cerebrovascular autoregulation, PRx and ORx, were determined as 0.03 ± 0.02 and 0.05 ± 0.06, respectively, prior to the initiation of CIN treatment. Following the induction of CIN treatment, both indices simultaneously increased to 0.16 ± 0.05 and 0.13 ± 0.03, respectively, within 12 hours, which was statistically significant for the PRx. After the cessation of nimodipine infusion, both indices
14
Hockel et al.
ACCEPTED MANUSCRIPT
significantly decreased to 0.03 ± 0.03 and -0.07 ± 0.03, respectively, within 24 hours (p<0.05) (Fig. 2A + B).
Discussion
RI PT
In the current study, we demonstrate the feasibility and relative safety of long-term intra-arterial nimodipine infusion for the treatment of severe macrovasospasm in selected high-risk patients.
SC
Angiographically detected macrovasospasm is only one of the presumed pathomechanisms that contribute to DCI after SAH
18,19
; however, endovascular
M AN U
procedures that directly target refractory macro-vasoconstriction remain widely employed. The reduction of macrovasospasm does not necessarily lead to an improvement in neurological outcome
20
; however, previous studies have shown that
the intra-arterial application of nimodipine significantly reversed macrovasospasm in
TE D
approximately 50-65% of cases 7 and led to consecutive improvement in the cerebral perfusion in CTP, which lasted for up to 24 hours.9 In these studies, an immediate clinical improvement following a single treatment of intra-arterial nimodipine occurred
EP
in up to 75% of patients.
The vasospasm period, however, may last up to 14 days after the hemorrhage.7,9
AC C
Thus, in cases of recurring macrovasospasm, multiple interventions are necessary to prevent ischemia. Daily interventions result in daily patient transport from the ICU to the angio suite, which bears risks for hypothermia, ICP increase, ventilatory problems and hemodynamic instability. Nimodipine administration is associated with a reduction in blood pressure and, consequently, the cerebral perfusion pressure.21 Specifically, an intra-arterial bolus application, as performed as a single-time treatment, can lead to a profound decrease in the MAP and CPP, which requires high doses of counterbalancing catecholamines to maintain the cerebral perfusion within a 15
Hockel et al.
ACCEPTED MANUSCRIPT
desired range. Although the single-time nimodipine effect is more long-lasting compared with papaverine,9,11 in general, uncertainty exists regarding whether the single-time treatment with nimodipine will provide sufficient protection until the next intervention is scheduled. The continuous application of nimodipine over indwelling
RI PT
catheters placed proximal to the major cerebral arteries has been previously published.12-16 Previous studies have demonstrated the feasibility of the CIN
SC
procedure for a maximum duration of 6 days.
The present study is the first investigation of long-term CIN treatment over a period of
M AN U
9-15 days, which thus covers the entire time during which patients were at risk for macrovasospasm. Microcatheter placement in each internal carotid artery is most suitable and will affect the entire anterior circulation. Nevertheless, vertebral artery catheters that target posterior circulation vasospasm are also feasible. In contrast to
TE D
a bolus application, we demonstrated that continuous nimodipine infusion only mildly decreased the arterial blood pressure, and catecholamine doses were not significantly increased during treatment.
EP
Monitoring of the therapeutic effects during CIN treatment remains critical. Repeated imaging studies would contradict one aim of CIN, which is to decrease the number of
AC C
patient transports. Thus, continuous monitoring of CBF surrogate parameters should be performed during the entire treatment period.14,16 Our findings imply that the effect of continuous nimodipine on macrovasospasm is not immediately evident. The TCD flow velocities decreased slowly over days, and brain tissue oxygen monitoring was able to demonstrate a continuous but also rather slow increase in the pbrO2, which improved by 37% at the end of treatment. The levels of sedation, CPP and paO2 were stable during the treatment period; thus, the improvement in the pbrO2 in parallel to decreasing
TCD
flow
velocities
is
most 16
likely
reflecting
a
decrease
of
Hockel et al.
ACCEPTED MANUSCRIPT
macrovasospasms plus an improvement in the cerebral blood flow. In addition to the reduction of macrovasospasm and the improvement of cerebral perfusion, the rate of cerebral infarctions under CIN was low. Before initiation of CIN 4 of 10 patients under maximum standard therapy, including bolus intra-arterial nimodipine treatment,
additional but small infarct during CIN treatment.
RI PT
developed new infractions on CT, whereas only 2 of 10 patients suffered an
Furthermore, our findings support the hypothesis that nimodipine is also effective in
SC
dilating the distal cerebral vasculature, i.e., the precapillary arterioles not visible on angiography.22 The autoregulation indices PRx and ORx, which describe the
M AN U
cerebrovascular autoregulatory capacity and thus the functionality of the precapillary arterioles, exhibited a close time-effect-relationship to nimodipine infusion. After the initiation of CIN, the PRx and ORx worsened within 12 hours indicating an impairment of autoregulation; however, the postulated thresholds for loss of
TE D
autoregulation, i.e., 0.2 for PRx and 0.25 for ORx,17,23 were not reached in every patient. In our opinion, small precapillary vessel vasodilatation is the most likely explanation for the direct effect of nimodipine on the ICP-based PRx and the brain
EP
tissue oxygenation-based ORx. This phenomenon of a pharmacologically induced worsening in the autoregulation indices must be distinguished from the spontaneous
AC C
impairment of cerebrovascular autoregulation (increase in the indices) associated with DCI and an unfavorable outcome in SAH.2,24,25 Jaeger et al. demonstrated that in patients with high-grade SAH, a consistent increase in the ORx values was predictive of the development of cerebral infarction.2 None of the patients in our group exhibited a pathologically increased PRx or ORx prior to CIN treatment; thus, autoregulation was intact. The monitoring of autoregulation indices may therefore be a way to trace the therapeutic effectiveness of nimodipine treatment on the precapillary resistance vessels. 17
Hockel et al.
ACCEPTED MANUSCRIPT
The potential adverse effects of long-term CIN treatment must be thoroughly considered. These are prolonged analgosedation, a strict supine position without positioning maneuvers to avoid catheter dislocation, heparinization, and an increased
RI PT
volume load as a result of the continuous nimodipine infusion. These potentially negative effects seem acceptable for a cohort of selected, high-risk, younger SAH patients.
SC
Heparinization is of special concern in these patients. The necessity of therapeutic anticoagulation is undisputed to prevent occlusion of or thromboembolism elicited by
M AN U
the intra-arterial catheter in the ICA. Although heparinization may have an additional positive effect via prevention of micro-thromboembolism, which is thought to contribute to secondary ischemia following SAH,18 hemorrhagic complications can occur as our data demonstrate.
TE D
We assume that the aneurysm re-rupture we experienced occurred independent of the applied anticoagulation because this is a primary clip occlusion failure. However, the severity of any postoperative hemorrhagic complication is affected by
EP
anticoagulation, and we cannot exclude a contribution to the final outcome in these patients. Therefore, we exclude meanwhile patients who underwent previous
AC C
intracranial surgery within a 7-day period, e.g., craniotomy for aneurysm clipping, or those who have any other contraindication for continuous anticoagulation (see algorithm for patient selection, Appendix). CIN treatment must be terminated when a surgical intervention becomes necessary, e.g., a decompressive craniectomy for ICP control, as observed in one case. Therefore, all necessary surgical procedures, such as the implantation of ICP probes or EVD catheters, must be accomplished prior to the initiation of CIN. Nimodipine infusion, which leads to vasodilatation and thus an increase in cerebral blood volume, 18
Hockel et al.
ACCEPTED MANUSCRIPT
did not significantly increase the ICP in our patient cohort. Nevertheless, we exclude patients with massive brain swelling, an increased ICP or other indices of compromised intracranial compliance from CIN treatment.
RI PT
Patients who may benefit from long-term intra-arterial nimodipine treatment must be carefully identified considering the invasiveness of the technique and the described potential hazards. Although the long-term outcome in our cohort was homogeneously
SC
favorable (median GOS of 5 at 3 months), over-treatment for some patients cannot be excluded. DCI was clinically diagnosed in only three of ten patients; in the other
M AN U
patients, we strongly assumed, that the severity of macrovasospasm and the perfusion deficits detected by CTP would have correlated with clinical DCI. We demonstrated that the conduction of continuous intra-arterial infusion was feasible beyond two weeks; however, a superior effect on neurological outcome and
TE D
the prevention of cerebral ischemia compared with standard treatment cannot be demonstrated from this preliminary retrospective non-controlled cohort. Therefore, a controlled larger study with defined inclusion criteria that compare basic treatment
EP
protocols (hemodynamics and nimodipine) to more aggressive interventional regimens, i.e., bolus and continuous intra-arterial nimodipine treatment, with respect
AC C
to neurological outcome, the incidence of cerebral infarction and the complication rate remains necessary.
Conclusion CIN treatment may prevent the development of secondary ischemia within a window of 14 days in patients with prolonged and severe macrovasospasm associated with CT perfusion deficits or secondary clinical deterioration prior to sedation and intubation. When patients can be identified, in which conservative therapy and single19
Hockel et al.
ACCEPTED MANUSCRIPT
time bolus interventions with intra-arterial nimodipine fail to achieve a sustained improvement of cerebral perfusion, CIN treatment appears to be justified as a rescue therapy in well-selected cases based on these preliminary data. A randomized controlled study with a larger patient cohort is necessary to confirm the overall
RI PT
efficacy of CIN in the prevention of DCI and to further characterize the associated risks.
SC
Conflict of interest
AC C
EP
TE D
M AN U
None.
20
Hockel et al.
ACCEPTED MANUSCRIPT Reference List 1.
Crowley RW, Medel R, Dumont AS, Ilodigwe D, Kassell NF, Mayer SA, Ruefenacht D, Schmiedek P, Weidauer S, Pasqualin A, Macdonald RL. Angiographic vasospasm is strongly correlated with cerebral infarction after
2.
RI PT
subarachnoid hemorrhage. Stroke. 2011;42:919–923.
Jaeger M, Soehle M, Schuhmann MU, Meixensberger J. Clinical significance of impaired cerebrovascular autoregulation after severe aneurysmal
3.
SC
subarachnoid hemorrhage. Stroke. 2012;43:2097–2101.
Bederson JB, Connolly ES, Batjer HH, Dacey RG, Dion JE, Diringer MN,
M AN U
Duldner JE, Harbaugh RE, Patel AB, Rosenwasser RH. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke. 2009;40:994–1025. Dorhout Mees SM, Rinkel GJE, Feigin VL, Algra A, van den Bergh WM,
TE D
4.
Vermeulen M, van Gijn J. Calcium antagonists for aneurysmal subarachnoid hemorrhage. Stroke. 2008;39:514–515. Pickard JD, Murray GD, Illingworth R, Shaw MD, Teasdale GM, Foy PM,
EP
5.
Humphrey PR, Lang DA, Nelson R, Richards P. Effect of oral nimodipine on
AC C
cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. BMJ. 1989;298:636–642.
6.
Raabe A, Beck J, Keller M, Vatter H, Zimmermann M, Seifert V. Relative importance of hypertension compared with hypervolemia for increasing cerebral oxygenation in patients with cerebral vasospasm after subarachnoid hemorrhage. J Neurosurg. 2005;103:974–981.
7.
Biondi A, Ricciardi GK, Puybasset L, Abdennour L, Longo M, Chiras J, Van Effenterre R. Intra-arterial nimodipine for the treatment of symptomatic cerebral 21
Hockel et al.
ACCEPTED MANUSCRIPT
vasospasm after aneurysmal subarachnoid hemorrhage: preliminary results. AJNR Am J Neuroradiol. 2004;25:1067–1076. 8.
Feng L, Fitzsimmons B-F, Young WL, Berman MF, Lin E, Aagaard BDL, Duong H, Pile-Spellman J. Intraarterially administered verapamil as adjunct therapy for
RI PT
cerebral vasospasm: safety and 2-year experience. AJNR Am J Neuroradiol. 2002;23:1284–1290. 9.
Hänggi D, Turowski B, Beseoglu K, Yong M, Steiger HJ. Intra-arterial
SC
nimodipine for severe cerebral vasospasm after aneurysmal subarachnoid
hemorrhage: influence on clinical course and cerebral perfusion. AJNR Am J
10.
M AN U
Neuroradiol. 2008;29:1053–1060.
Hoelper BM, Hofmann E, Sporleder R, Soldner F, Behr R. Transluminal balloon angioplasty improves brain tissue oxygenation and metabolism in severe vasospasm after aneurysmal subarachnoid hemorrhage: case report.
11.
TE D
Neurosurgery. 2003;52:970–976.
Vajkoczy P, Horn P, Bauhuf C, Munch E, Hubner U, Ing D, Thome C, PoecklerSchoeninger C, Roth H, Schmiedek P. Effect of intra-arterial papaverine on
EP
regional cerebral blood flow in hemodynamically relevant cerebral vasospasm. Stroke. 2001;32:498–505.
Doukas A, Petridis AK, Barth H, Jansen O, Mehdorn HM. Continuous intra-
AC C
12.
arterial infusion of nimodipine at the onset of resistant vasospasm in aneurysmal subarachnoidal haemorrhage. Technical report. Neurol Res. 2011;33:290–294.
13.
Mayer TE, Dichgans M, Straube A, Birnbaum T, Müller-Schunk S, Hamann GF, Schulte-Altedorneburg G. Continuous intra-arterial nimodipine for the treatment of cerebral vasospasm. Cardiovasc Intervent Radiol. 2008;31:1200–1204.
22
Hockel et al.
ACCEPTED MANUSCRIPT 14.
Musahl C, Henkes H, Vajda Z, Coburger J, Hopf N. Continuous local intraarterial nimodipine administration in severe symptomatic vasospasm after subarachnoid hemorrhage. Neurosurgery. 2011;68:1541–1547.
15.
Ott S, Jedlicka S, Wolf S, Peter M, Pudenz C, Merker P, Schürer L, Lumenta
RI PT
CB. Continuous selective intra-arterial application of nimodipine in refractory cerebral vasospasm due to aneurysmal subarachnoid hemorrhage. BioMed Res Int. 2014;2014:970741.
Wolf S, Martin H, Landscheidt JF, Rodiek SO, Schürer L, Lumenta CB.
SC
16.
Continuous selective intraarterial infusion of nimodipine for therapy of
17.
M AN U
refractory cerebral vasospasm. Neurocrit Care. 2010;12:346–351. Jaeger M, Schuhmann MU, Soehle M, Meixensberger J. Continuous assessment of cerebrovascular autoregulation after traumatic brain injury using brain tissue oxygen pressure reactivity. Crit Care Med. 2006;34:1783–1788. Sehba FA, Bederson JB. Mechanisms of acute brain injury after subarachnoid
TE D
18.
hemorrhage. Neurol Res. 2006;28:381–398. 19.
Vergouwen MD, Etminan N, Ilodigwe D, Macdonald RL. Lower incidence of
EP
cerebral infarction correlates with improved functional outcome after aneurysmal subarachnoid hemorrhage. J Cereb Blood Flow Metab.
20.
AC C
2011;31:1545–1553.
Etminan N, Vergouwen MD, Ilodigwe D, Macdonald RL. Effect of pharmaceutical treatment on vasospasm, delayed cerebral ischemia, and clinical outcome in patients with aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Cereb Blood Flow Metab. 2011;31:1443–1451.
21.
Choi HA, Ko S, Chen H, Gilmore E, Carpenter AM, Lee D, Claassen J, Mayer SA, Schmidt JM, Lee K, Connelly ES, Paik M, Badjatia N. Acute effects of 23
Hockel et al.
ACCEPTED MANUSCRIPT
nimodipine on cerebral vasculature and brain metabolism in high grade subarachnoid hemorrhage patients. Neurocrit Care. 2012;16:363–367. 22.
Canova D, Roatta S, Micieli G, Bosone D. Cerebral oxygenation and haemodynamic effects induced by nimodipine in healthy subjects. Funct
23.
RI PT
Neurol. 2012;27:169–176.
Czosnyka M, Smielewski P, Kirkpatrick P, Laing RJ, Menon D, Pickard JD. Continuous assessment of the cerebral vasomotor reactivity in head injury.
24.
SC
Neurosurgery. 1997;41:11–19.
Bijlenga P, Czosnyka M, Budohoski KP, Soehle M, Pickard JD, Kirkpatrick PJ,
M AN U
Smielewski P. “Optimal cerebral perfusion pressure” in poor grade patients after subarachnoid hemorrhage. Neurocrit Care. 2010;13:17–23. Rasulo FA, Girardini A, Lavinio A, De Peri E, Stefini R, Cenzato M, Nodari I, Latronico N. Are optimal cerebral perfusion pressure and cerebrovascular
TE D
autoregulation related to Long-term outcome in patients with aneurysmal
EP
subarachnoid hemorrhage? J Neurosurg Anesthesiol. 2012;24:3–8.
AC C
25.
24
Hockel et al.
ACCEPTED MANUSCRIPT Figure legends
Figure 1: Time course of the flow velocities via TCD in the right and left MCA. When vasospasms were diagnosed, the flow velocities had increased to approximately 150
SC
day 10 (p<0.05) compared with the pre-CIN TCD exams.
RI PT
cm/s, median value. After CIN induction, the velocities significantly decreased until
Figure 2A and B: Time course of changes in the PRx (Fig. 2A) and ORx (Fig. 2B),
M AN U
displayed as 12-hour means before and after the induction and before and after the termination of CIN treatment. Both indices indicate a close time relationship with CIN treatment, and the autoregulation parameters were significantly altered during
AC C
EP
TE D
treatment, p<0.05.
25
ACCEPTED MANUSCRIPT
Algorithm for selection patients for CIN
within 48 hours
Monitoring
ICP, pbrO2, microdialysis
Screening
neuro exams, TCD, CTA/CTP (5, 10, 15thd)
Prophylaxis
CPP > 70mmHg, IV NDP
confirm macrovasospasm
angiography
consider endovasc. treatment
IA bolus NDP (+ increase IV NDP)
24 hours
CTA/CTP, angiography
TE D
repeat diagnostics
M AN U
yes
SC
neurological deterioration, radiographic DCI
Exclusion criteria for CIN: • major territorial infarction or intracerebral hemorrhage • previous intracranial surgery, i.e. craniotomy, aneurysm clipping within 5-7days • increased ICP, significant brain swelling • any contraindication for continuous anticoagulation
RI PT
aneurysm occlusion CSF drainage
SAH
consider CIN if • continued radiographic DCI • absent exclusion criteria
• IA continuous NDP (0.4mg/h per catheter) • Heparin PTT 60-80 sec
AC C
EP
repeat IA bolus NDP if • exclusion criteria for CIN • repeat diagnostics every 24h
Monitoring/Screening
IA bolus NDP (+ increase IV NDP Continue IV NDP repeat diagnostics
ACCEPTED MANUSCRIPT
Table 1: Patient characteristics. sex (f/m)
age (yrs.)
Hunt & Hess Fisher aneurysm treatment grade° grade° location (clip/coil)
EVdrain (y/n)
DCI on diagnosis of admission vasospasm (y/n) (days post SAH)
1
f
46
4
4
MCA left
clip
y
n
5
2
f
29
3
3
AComA
clip
y
n
5
3
f
46
3
4
PComA
coil
y
n
8
4
f
38
3
3
AComA
coil
n
y
6
5
f
27
2
4
MCA right clip
n
n
4
6
m
45
3
2
AComA
clip
n
y
9
7
f
47
3
3
AComA
clip
y
n
7
8
f
45
5
3
ACI right
coil
y
n
4
9
m
25
5
4
AComA
clip
10
f
46
3
4
ICA left
coil
SC
RI PT
Patient No°
y
n
5
y
n
4
AC C
EP
TE D
M AN U
MCA – middle cerebral artery, AComA – anterior communicating artery, PComA – posterior communicating artery, ICA – internal carotid artery, time – days after the initial hemorrhage; ev -extraventricular.
ACCEPTED MANUSCRIPT Table 2: Patient outcome statistics. single-time start duration intra-artrial of CIN of CIN infarcts nimodipine (day) (days) on CT complications
GOS at GOS at discharge 3 months
1
gen.
2
7
15
y
-
4
4
2
ACA/ICA bilat
-
5
12
n
-
5
5
3
MCA./ICA bilat. -
8
13
y
-
4
-
4
ACA/ICA bilat.
3
9
11
y
-
4
5
5
MCA/ICA right
4
10
14
y
-
6
ACA/MCA bilat. -
10
14
n
7
ACA/MCA bilat. -
7
9
8
MCA bilat.
1
5
9
gen.
-
10
ACA/MCA bilat. -
4
5
cath. disl.
3
-
n
-
4
5
14
n
GI bleed
4
-
9
11
n
cath. disl.
4
5
4
11
n
SC
RI PT
Patient site of No° vasospasm
5
HIT
5
AC C
EP
TE D
M AN U
gen. – generalized, MCA – middle cerebral artery, ACA – anterior cerebral artery, ICA – internal carotid artery, bilat. – bilateral, cath disl. – catheter dislocation, time – days after the initial hemorrhage.
ACCEPTED MANUSCRIPT Table 3: CT-angiographic grading of vasospasm. grading, total n=10
prior CIN
during CIN*
after CIN
high grade vasospasm
10
2
0
moderate vasospasm
0
5
2
light vasospasm
0
1
3
no vasospasm
0
0
5
AC C
EP
TE D
M AN U
SC
RI PT
*not every patient received CTA under CIN (n=9)
ACCEPTED MANUSCRIPT Table 4: Physiological parameters during CIN. start monit.
prior CIN
CIN day 0
CIN day 5
CIN day 10
ABP (mmHg)
89 ± 2
90 ± 2
92 ± 2
92 ± 3
90 ± 2
CPP (mmHg)
77 ± 2
79 ± 2
80 ± 1
83 ± 2
83 ± 2
ICP (mmHg)
11 ± 1
10 ± 1
12 ± 1
9±1
8±1
pbrO2 (mmHg)
29.1 ± 3.1
27.1 ± 3.1
23.8 ± 3.5
27.7 ± 4.7
32.7 ± 4.8
paO2 (mmHg)
119 ± 1.7
111.8 ± 4.7
111.6 ± 7.6
113.1 ± 5.3
114.2 ± 4.4
paCO2 (mmHg)
37.6 ± 0.6
36.0 ± 0.6
37.8 ± 1.3
37.9 ± 0.5
AC C
EP
TE D
M AN U
SC
24 hour mean ± SEM; p>0.05.
RI PT
variable
37.1 ± 0.7
ACCEPTED MANUSCRIPT
v MCA (cm/s)
mean ± SEM; n=10; * p<0.05
*
SC M AN U TE D
150
EP
100
50
AC C
v TCD (cm/s)
200
RI PT
250
0
Sc
r
n ee
ing r CIN a y 0 d io Npr I C
2 N
CI
a -d
y5
3 N CI
-
y da
10
ACCEPTED MANUSCRIPT
mean ± SEM; n=10; * p<0.05
RI PT
0.3
SC
*
*
TE D
M AN U
0.2
EP
0.1
AC C
pressure reactivity index (PRx)
PRx
0.0
CIN infusion -0.1 12 h
12 h
12 h
12 h
12 h
12 h
12 h
12 h
ACCEPTED MANUSCRIPT
mean ± SEM; n=10; * p<0.05
*
SC
RI PT
0.3
M AN U
0.2
EP
TE D
0.1
AC C
oxygen reactivity index (ORx)
ORx
0.0
CIN infusion -0.1 12 h
12 h
12 h
12 h
12 h
12 h
12 h
12 h
ACCEPTED MANUSCRIPT Highlights We investigated a patient series with severe refractory macrovasospasm after subarachnoid hemorrhage Long-term continuous intra-arterial nimodipine treatment as rescue therapy was retrospectively evaluated for feasibility, safety and effectiveness
RI PT
Continuous improvement of macrovasospasm severity and brain tissue oxygenation were found after 12 days (mean) of treatment
AC C
EP
TE D
M AN U
SC
Associated risks have to be weighed up thoroughly against potential benefits