INTENSIVE CARE MEDICINE: ELECTRONIC SUPPLEMENTARY MATERIAL

ARTICLE TITLE: Cerebrospinal fluid pentraxin 3 early after subarachnoid hemorrhage is associated with vasospasm

By: Elisa R Zanier1,2 MD, Giovanna Brandi2 MD, Giuseppe Peri3 PhD, Luca Longhi2 MD, Tommaso Zoerle2 MD, Mauro Tettamanti1 PhD, Cecilia Garlanda3 PhD, Anna Sigurtà2 MD, Serenella Valaperta3 PhD, Alberto Mantovani3,4 PhD, Maria Grazia De Simoni1 PhD, Nino Stocchetti2 MD.

1Mario Negri Institute for Pharmacological Research, Milano, Italy. 2Milano University , Neurosurgical Intensive Care Unit, Department of Anesthesia and Critical Care Medicine, Fondazione IRCCS Ospedale Maggiore Policlinico, Mangiagalli e Regina Elena, Milano, Italy. 3Istituto Clinico Humanitas, IRCCS, Rozzano, Milano, Italy. 4Department of Translational Medicine, Milano University, Italy.

Correspondence to:

Maria Grazia De Simoni, Ph.D.

Mario Negri Institute

Departement of Neuroscience, Laboratory of Inflammation and Nervous System Diseases

via Giuseppe La Masa, 19

20156 Milan Italy

Phone: +39 02 390 14 568

Fax: +39 02 390 01 916

Email:

SUPPLEMENTARY MATERIALS

Supplementary Methods

Supplementary Table 1

Supplementary Figures 1-4

Supplementary Methods: Clinical management

Our ICU was a 4 bed unit treating patients with acute brain damage (traumatic brain injury, cerebrovascular disease, status epilepticus) and postoperative patients following elective surgery for brain tumors and complex spine surgery.

In case of SAH, injury severity on admission was recorded according to the World Federation of Neurosurgical Societies (WFNS) grading scale. Based on WFNS we divided patients into Mild (score: 1-3) and Severe (score: 4-5) for analysis. Patients underwent four-vessel cerebral angiography within 24 h from admission. Management goals included the early clipping/coiling of the aneurysm (in this study, 78% of the ruptured intracranial aneurysm were secured within the first 24 h) and, where indicated, surgical evacuation of the intracranial hematoma. Hydrocephalus was drained using an intraventricular catheter and intracranial pressure (ICP) was monitored with the goal to maintain ICP levels below 20 mmHg and cerebral perfusion pressure around 60-70 mmHg. A jugular bulb catheter was inserted in all comatose patients as part of clinical routine to determine jugular saturation of oxygen.

Supplementary Table 1: Subarachnoid Hemorrhage Patient Clinical Information.

Patient / Age (yrs) / Sex / WFNS / Fisher Grade / Aneurysm Location / Treatment / VS / 6-months GOS
1 / 56 / F / 5 / 3 / MCA / coil / No / 3
2 / 51 / F / 5 / 3 / PCA / coil / No / 1
3 / 48 / F / 5 / 3 / ICA / none / No / 1
4 / 65 / M / 5 / 3 / VA / none / No / 1
5 / 39 / F / 5 / 3 / ACom / coil / No / 1
6 / 68 / F / 5 / 3 / No angiogram / none / No / 1
7 / 73 / F / 4 / 3 / ACom / coil / Yes / 1
8 / 68 / F / 4 / 4 / ACom / coil / Yes / 3
9 / 58 / F / 4 / 4 / MCA / clip / No / 3
10 / 78 / F / 4 / 3 / ACA/pericallosal / clip / Yes / 1
11 / 38 / F / 1 / 3 / Ach / coil / Yes / 5
12 / 56 / F / 5 / 3 / PICA / coil / No / 5
13 / 47 / F / 1 / 4 / ACom / coil / Yes / 1
14 / 54 / F / 4 / 3 / BA / coil / No / 3
15 / 62 / F / 5 / 4 / MCA / clip / No / 3
16 / 73 / F / 1 / 4 / ACom / coil / Yes / 1
17 / 37 / F / 1 / 3 / PCom / coil / No / 5
18 / 63 / F / 2 / 4 / BA / coil / No / 5
19 / 63 / F / 1 / 3 / SCA / coil / No / 5
20 / 30 / F / 2 / 3 / ICA / coil / Yes / 4
21 / 41 / M / 2 / 4 / PICA / coil / No / 5
22 / 67 / F / 4 / 4 / MCA / coil / Yes / 4
23 / 45 / F / 4 / 3 / ACA / coil / Yes / 5
24 / 68 / F / 1 / 4 / PCA / coil / No / N.A.
25 / 63 / M / 5 / 3 / ACom / clip / Yes / 1
26 / 49 / F / 3 / 4 / ACA / coil / No / 3
27 / 42 / M / 5 / 3 / ACA / coil / No / 3
28 / 67 / F / 5 / 3 / ICA / coil / Yes / 3
29 / 37 / F / 1 / 2 / ICA / coil / No / 5
30 / 52 / F / 5 / 3 / ACA / coil / No / 1
31 / 45 / F / 1 / 3 / ACA / coil / No / 5
32 / 49 / F / 5 / 4 / PICA / coil / No / 3
33 / 42 / F / 4 / 3 / VA / coil / Yes / 1
34 / 55 / M / 5 / 3 / No angiogram / none / No / 1
35 / 39 / M / 2 / 3 / MCA / clip / Yes / 1
36 / 56 / M / 5 / 3 / MCA / clip / No / 1
37 / 51 / F / 1 / 3 / PCom / coil / Yes / 1
38 / 57 / F / 5 / 3 / ACA / coil / No / 1

Abbreviations: WFNS: World Federation of Neurological Surgeons grade. VS: Vasospasm. GOS: Glasgow Outcome Scale. MCA: Middle cerebral artery. PCA: Posterior cerebral artery. ICA: Internal carotid artery. VA: Vertebral artery. ACom: Anterior communicating artery. Ach: Anterior choroidal artery. PICA: Posterior inferior cerebellar artery. BA: Basilar artery. PCom: Posterior communicating artery. SCA: Superior cerebellar artery. ACA: Anterior cerebral artery. N.A.: Not available.

S1

Supplementary Figure 1: Relationship between acute CSF and arterial PTX3 peak concentrations and acute clinical status. Patients were categorized into mild (WFNS 1-3, n=24) and severe (WFNS 4-5, n=14). There were no significant differences in CSF (A) and arterial (B) PTX3 acute peak concentrations between groups.

S2

Supplementary Figure 2: Dynamics of PTX3 in CSF and plasma of patients without clinical vasospasm (n=16). A) CSF peak PTX3 levels on days 1-2 (acute phase), on days 3-4 (subacute phase) and on days 4-14 post-SAH (late phase) are plotted, an overall decrease over time was observed in these patients (Friedman test p=0.004, no significant post-hoc differences were detected). B) A similar behavior was also observed for arterial peak PTX3 levels (Friedman test p=0.0001) that where significantly lower in the late phase compared to the acute phase (p<0.05). Data are presented as Box plots and 10th and 90th percentiles.

S3

Supplementary Figure 3: Comparison of CSF PTX3 levels in patients with (n=12) or without (n=12) vasospasm during the late phase (4-14 days post-SAH). In the late phase patients who did not develop vasospasm showed significantly lower CSF PTX3 levels than patients with this complication (Mann-Whitney U test p=0.012). Data are presented as Box plots and 10th and 90th percentiles.

S4

Supplementary Figure 4: Relationship of CSF PTX levels to 6-month Glasgow Outcome Scale (GOS). Based on 6 months GOS we divided patients into favorable (GOS 4-5, n=11) and unfavorable (GOS 1-2-3, n=27) outcome; there was considerable overlap between groups and no differences could be detected. Data are presented as Box plots, box represents the upper and lower quartiles, range represented by the whiskers is the 10th and 90th percentiles.