Neuropeptides 48 (2014) 91–96

Contents lists available at ScienceDirect

Neuropeptides

journal homepage: www.elsevier .com/locate /npep

Vasopressin and oxytocin in CSF and plasma of patients with aneurysmalsubarachnoid haemorrhage

0143-4179/$ - see front matter � 2014 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.npep.2013.12.004

⇑ Corresponding author. Tel.: +49 89 41405471; fax: +49 89 41404886.E-mail address: j.martin@lrz.tum.de (J. Martin).

Jan Martin a,⇑, Simone M. Kagerbauer a, Tibor Schuster b, Manfred Blobner a, Eberhard F. Kochs a,Rainer Landgraf c

a Klinik für Anaesthesiologie, Technische Universität München, Klinikum rechts der Isar, Ismaninger Strasse 22, 81675 München, Germanyb Institut für Medizinische Statistik und Epidemiologie, Technische Universität München, Klinikum rechts der Isar, Ismaninger Strasse 22, 81675 München, Germanyc Max-Planck-Institut für Psychiatrie, Kraepelinstrasse 2, 80804 München, Germany

a r t i c l e i n f o a b s t r a c t

Article history:Received 23 May 2013Accepted 23 December 2013Available online 3 January 2014

Keywords:VasopressinOxytocinCSFPlasmaAneurysmalSubarachnoidHaemorrhage

Objective: Clinicopathological studies on patients succumbing to subarachnoid haemorrhage (SAH) dem-onstrated hypothalamic lesions. The implication of the hypothalamic neuropeptides arginine-vasopressin(AVP) and oxytocin (OXT) has not been linked to aneurysmal SAH yet. This study investigates AVP andOXT in CSF and plasma of patients with spontaneous aneurysmal SAH and their association with out-come.Methods: CSF and plasma samples of 12 patients with aneurysmal SAH were prospectively studied for2 weeks. AVP and OXT were measured by radioimmunoassay. Outcome was assessed on Glasgow-Out-come-Scale. Twenty-nine patients without neuropsychiatric disturbances served as controls. Differencesin neuropeptide concentration time courses were assessed by regression models. Group comparisonswere performed by Kruskal–Wallis and correlations by Spearman tests.Results: Regression of CSF levels between patients with poor and good outcome revealed significantlylower levels of AVP in patients with poor outcome (p = 0.012) while OXT showed a trend towards lowerlevels (p = 0.063). In plasma, no significant differences between outcome groups were found. Group com-parisons between poor outcome patients and controls revealed significant differences in CSF for AVP(p = 0.001) and OXT (p = 0.015). In plasma, AVP yielded significantly different results while OXT didnot. No differences were found between the good outcome group and controls. Plasma and CSF concen-trations showed no significant correlation.Conclusion: Patients with poor outcome after aneurysmal SAH have lower AVP and OXT levels in CSF thanpatients with good outcome while neuropeptide levels in plasma failed to reflect differences in outcome.The data indicate hypothalamic damage as an aetiologic factor for outcome after aneurysmal SAH.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The incidence of subarachnoid haemorrhage (SAH) is approxi-mately six to seven per 100,000 person-years in most populations(van Gijn et al., 2007; Linn et al., 1996). Ruptured cerebral aneu-rysms are the cause in 85% of patients (van Gijn et al., 2007). Halfthe patients are younger than 55 years at the time of haemorrhage,thus producing a large burden of premature morbidity and mortal-ity (Johnston et al., 1998; van Gijn et al., 2007). Despite improvedmethods of microsurgery, interventional neuroradiology and treat-ment in specialized intensive care units, the neurological outcomeafter aneurysmal SAH often remains disappointing (Hackett andAnderson, 2000; Mayer et al., 2002). The initial neurological statusof the patient after the ictus seems to have a crucial impact on the

outcome while the precise pathogenesis of a possibly evolving de-layed neurological deterioration and its causal links to the initialneurological condition remain unclear (Macdonald et al., 2007;Giraldo et al., 2012; Rosengart et al., 2007).

Early clinicopathological studies of patients succumbing to SAHdemonstrated hypothalamic lesions in histological examinations(Doshi and Neil-Dwyer, 1980; Neil-Dwyer et al., 1994; Crompton,1963). Since disturbances in sodium metabolism and osmolalitywere frequently reported in patients with SAH, the implication ofthe hypothalamic neuropeptides arginine-vasopressin (AVP) andoxytocin (OXT) was assessed in clinical studies focusing on electro-lyte-water balance, cerebral vasospasm and diurnal concentrationchanges in cerebrospinal fluid (CSF) (Mather et al., 1981; Kuboy-ama et al., 1988). In a rather small study, Kuboyama et al. analysedAVP and OXT in CSF of 4 patients for 2 days in the second weekafter aneurysmal SAH and found a preserved diurnal pattern butall patients were without neurological deficit (Kuboyama et al.,

92 J. Martin et al. / Neuropeptides 48 (2014) 91–96

1988). Mather et al. suggested a relation between increased AVPconcentrations, cerebral oedema and worsening of the neurologicalstatus (Mather et al., 1981). Neither, however, did these authorsinvestigate the concentration course during the 2 weeks ofsub-acute illness after the onset of SAH nor was OXT measuredin conjunction with AVP. The latter is of particular interest, as bothneuropeptides have been shown to cross-react, due to the highextent of receptor homology (Sala et al., 2011). Since cerebralvasospasm received growing attention as a possible cause forneurological deterioration, the involvement of AVP in the develop-ment of cerebral vasospasm was suggested in rodent subarachnoidhaemorrhage models (Nishihashi et al., 2005; Trandafir et al., 2004;Delgado et al., 1988).

At present, the hypothalamic neuropeptides OXT and AVP aresubjected to ongoing research, in particular their functional impli-cation in promoting diverse aspects of socio-emotional behavioursincluding social preference and cognition, pair-bonding, aggressionand anxiety (Lukas et al., 2011; Bosch and Neumann, 2008; Don-aldson and Young, 2008; Jarcho et al., 2011; Gabor et al., 2012;Bielsky and Young, 2004; Gossen et al., 2012; Neumann and Landg-raf, 2012) but have not been linked to aneurysmal SAH yet. As thehypothalamus constitutes a possible focus of damage in aneurys-mal SAH (Neil-Dwyer et al., 1994; Doshi and Neil-Dwyer, 1980;Mather et al., 1981), alterations in neuropeptide concentrationsmerit further attention. AVP and OXT are released centrally andperipherally (Landgraf and Neumann, 2004; Neumann and Landg-raf, 2012) with no correlative concentrations in both compart-ments as recently shown in humans (Kagerbauer et al., 2013).The present study thus investigates the levels of the neuropeptidesAVP and OXT in both CSF and plasma of patients with spontaneousaneurysmal SAH during the 2 weeks of sub-acute illness after theinitial bleed. The association with the patients’ functional outcomeis evaluated.

2. Materials and methods

The study was approved by the Institutional Review Board ofthe Medical Faculty of the Technische Universität München. Weprospectively studied 12 patients (average age 55.1 years, range21–71) with aneurysmal SAH (Table 1). No patients with a historyof intracranial illness were included in the study. The patients’ pre-operative clinical condition was assessed by the grading scale ofthe World Federation of Neurological Surgeons (WFNS) and thesystem of Hunt and Hess (HH) (Hunt and Hess, 1968; Teasdaleet al., 1988). Computed tomographic classification was based onthe method of Fisher et al. (1980). Transcranial Doppler (TCD)was performed daily to measure blood flow velocities in the ante-rior, middle and posterior cerebral arteries. Cerebral infarction wasdiagnosed by computer tomography (CT). Cranial CT scans wereperformed on hospital admission, on neurological deterioration,before ICU dismissal and in comatose patients approximately twicea week during the first 2 weeks of sub-acute illness. Patient out-come was assessed on the Glasgow Outcome Scale (GOS) 3 monthsafter the ictus (Jennett and Bond, 1975). A GOS of 1–3 (death, veg-etative state, severe disability) was considered as poor outcome,whereas a GOS of 4–5 (moderate disability, good recovery) wasconsidered as good outcome.

A cohort of 29 patients (15 female, 14 male patients; averageage 52.8 years, range 19–81) without neuropsychiatric distur-bances served as control group. CSF and plasma samples were ob-tained concomitantly during spinal anaesthesia for elective minororthopaedic or urological surgery.

Samples of CSF and plasma were collected daily or every otherday via the arterial line and the external ventricular drainage for2 weeks after the onset of SAH (mean sampling period 12 days,

range 10–14 days). Per patient, eight to nine paired CSF and plasmasamples were drawn resulting in a total of 194 samples for analysisof AVP and OXT in each sample. All samples were collected in pre-chilled plastic EDTA tubes and centrifuged for 10 min at 1300g at4 �C immediately after sampling. CSF and plasma samples wereanalysed identically, i.e., extracted and assayed in the same batchat the same time. Briefly, samples (0.5 ml) were kept at �20 �C un-til extraction using LiChroprep� Si60 (Merck) heat-activated at700 �C for 3 h. Twenty milligram of LiChroprep� Si60 in 1 ml dis-tilled water were added to the sample, mixed for 30 min, washedtwice with distilled water and 0.01 N HCl and eluded with 60% ace-tone. The lyophilized extract was divided to assay both neuropep-tides in highly sensitive and specific radioimmunoassays(RIAgnosis, Munich, Germany). Assay sensitivities were in the0.5 pg range, cross-reactivities with related peptides <0.7% and in-tra- and inter-assay variability <10%.

Generalized linear regression models were employed to assessgroup differences in the time course of AVP and OXT levels inCSF and plasma as well as in the time course of TCD values. Forthe time course analysis of AVP and OXT levels, a logarithmic linkfunction was chosen due to the right-skewed distribution. Toapproximate a smooth functional course of measurement valuesover time in the outcome groups, B-spline functions were fittedto the data in the regression analysis. This procedure allowed thestatistical assessment of mean group differences between thetwo outcome groups over the course of time (test of main effect)(McBurney and White, 2004).

Group comparisons of marginal (time-averaged) OXT and AVPlevels in plasma and CSF between outcome groups and healthycontrols were performed by Kruskal–Wallis testing with Bonfer-roni correction.

Strength of bivariate monotonous correlation of quantitativedata was assessed by the Spearman rank correlation coefficient(rho). An absolute value of the correlation coefficient of <0.39 indi-cates a weak correlation, 0.40–0.69 a modest correlation and a va-lue of >0.70 a strong correlation.

All statistical tests were conducted two-sided. A p-value <0.05was considered to indicate statistical significance. No adjustmentof p-values was made to correct for multiple testing, except forthe Bonferroni corrections performed within the Kruskal–Wallistesting. Statistical analysis was performed by using R software ver-sion 2.11.1 (R Development Core Team, R Foundation for StatisticalComputing, 2011, Vienna, Austria).

3. Results

The demographics of the study group are presented in Table 1.Twelve patients with aneurysmal SAH were studied. Poor outcome(GOS 1–3) was seen in seven patients, good outcome (GOS 4–5) infive patients. Twenty-nine CSF and plasma samples served as con-trols (median CSF OXT 12.7 pg/ml, range 4.3–35.1, interquartilerange (IQR) 6.8–20.1; median plasma OXT 6.0 pg/ml, range1.9–97.5, IQR 4.0–11.4; median CSF AVP 17.8 pg/ml, range 4.4–61.0,IQR 10.6–26.9; median plasma AVP 3.4 pg/ml, range 0.8–20.9,IQR 1.2–9.3).

The regression analysis of CSF levels between patients with poorand good outcome over the course of time revealed significantlylower levels of AVP in the former (Fig. 1, test of main effect:p = 0.012). OXT in CSF of patients with poor outcome showed a trendtowards lower levels (Fig. 2, test of main effect: p = 0.063). In plas-ma, the time course regression analysis between the outcomegroups revealed no significant differences for AVP (Fig. 3, test ofmain effect: p = 0.46) or OXT (Fig. 4, test of main effect: p = 0.078).

Pairwise comparisons of both outcome groups and controlgroup by Kruskal–Wallis analysis revealed statistically significant

Table 1Demographics of aSAH patients.

Age/sex Aneurysm WFNS HH Fisher CI LOS ICU GOS

1 65/F BA 4 4 3 No 25 32 71/F ACOMA 4 4 4 Yes 29 23 69/M ACOMA 5 5 4 No 13 14 62/F MCA 5 5 4 Yes 28 25 67/F ACOMA 2 2 3 No 9 46 52/M SCA 3 3 3 No 18 47 50/M MCA 1 2 3 No 9 58 21/F PCA 1 2 3 No 16 59 43/F ACOMA 5 5 4 Yes 23 210 63/F ICA 3 3 4 Yes 11 111 43/F MCA 2 2 3 No 14 512 55/F ACOMA 3 3 3 No 19 3

ACOMA, anterior communicating artery; aSAH, aneurysmal subarachnoid haem-orrhage; BA, basilar artery; CI, cerebral infarction; F, female; GCS, Glascow comascale score; GOS, Glascow outcome scale score at 3 months after the ictus; HH, Hunt& Hess classification; ICA, internal carotid artery; LOS, length of stay (days); MCA,middle cerebral artery; PCA, posterior cerebral artery; ICU, intensive care unit; M,male; SCA, superior cerebellar artery; WFNS, World Federation of Neurosurgeonsscale score.

Fig. 1. The concentration time course of AVP in CSF is presented on a logarithmicscale. Patients with good outcome (circles) are represented by the solid line,patients with poor outcome (triangles) by the broken line. The horizontal dottedline indicates the respective median of the control cohort.

Fig. 2. The concentration time course of OXT in CSF is presented on a logarithmicscale. Patients with good outcome (circles) are represented by the solid line,patients with poor outcome (triangles) by the broken line. The horizontal dottedline indicates the respective median of the control cohort.

Fig. 3. The concentration time course of AVP in plasma is presented on alogarithmic scale. Patients with good outcome (circles) are represented by thesolid line, patients with poor outcome (triangles) by the broken line. The horizontaldotted line indicates the respective median of the control cohort.

Fig. 4. The concentration time course of OXT in plasma is presented on alogarithmic scale. Patients with good outcome (circles) are represented by thesolid line, patients with poor outcome (triangles) by the broken line. The horizontaldotted line indicates the respective median of the control cohort.

J. Martin et al. / Neuropeptides 48 (2014) 91–96 93

differences for both AVP and OXT in CSF when comparing the pooroutcome group with the healthy controls (Table 2). In plasma, AVP

also yielded significantly different results while OXT did not(Table 2). No statistically significant differences were foundbetween the good outcome group and healthy controls.

The regression analysis of intracranial blood flow velocitiesshowed significantly higher values in patients with poor outcomecompared to patients with good outcome (test of main effect:p = 0.028). The correlation analysis comparing neuropeptide levelswith cerebral blood flow velocities revealed modest to strong in-verse correlations between TCD measurements and CSF parame-ters (CSF OXT: rho = �0.588; CSF AVP: rho = �0.702). Plasma OXT(rho = +0.178) and plasma AVP (rho = +0.371) showed weak corre-lations with TCD data.

The analysis of AVP and OXT levels in CSF vs plasma did not re-veal any significant correlations between the two compartments,neither in patients with aneurysmal SAH nor in healthy controls(Table 3).

4. Discussion

In patients with aneurysmal SAH, the implication of AVP andOXT, synthesized in the hypothalamic supraoptic and paraventric-ular nuclei, has not been comprehensively assessed. Given an

Table 2Group comparisons of outcome groups vs controls.

AVP CSF OXT CSF AVP plasma OXT plasma

Poor outcome vs healthy controls 0.001 0.015 0.021 0.663Good outcome vs healthy controls 0.366 0.588 0.366 0.999

Kruskal–Wallis-testing with Bonferroni adjusted p-values, group comparisons of marginal (time-averaged) AVP and OXT levels in CSF and plasma between outcome groupsand healthy controls; level of significance, p < 0.05; AVP, arginine vasopressin; CSF, cerebrospinal fluid; OXT, oxytocin.

Table 3Correlation analysis of AVP and OXT in CSF vs plasma.

Spearman’s rho

Patients with aSAHAVP CSF vs plasma �0.100OXT CSF vs plasma +0.301

Control groupAVP CSF vs plasma +0.113OXT CSF vs plasma �0.142

Spearman rank correlation analysis of CSF and plasma levels revealed no strongcorrelation for AVP and OXT. aSAH, aneurysmal subarachnoid haemorrhage; AVP,arginine vasopressin; CSF, cerebrospinal fluid; OXT, oxytocin.

94 J. Martin et al. / Neuropeptides 48 (2014) 91–96

increasing number of studies on AVP and OXT in the convergingfields of neuroendocrinology and behavioural neuroscience, theidea that both neuropeptides may reflect processes linked to thefunctional outcome in aneurysmal SAH remains to be investigatedin more detail. In this pilot study, we demonstrate that patientswith poor outcome after aneurysmal SAH have lower AVP andOXT levels in CSF than patients with good outcome. These findingsare illustrated by the neuropeptide level slopes of the two outcomegroups (Figs. 1 and 2). In contrast to CSF, the slopes of plasma OXT(Fig. 3) and plasma AVP (Fig. 4) failed to discriminate between pa-tients with good vs poor outcome implying that differences infunctional outcome are better reflected by neuropeptide levels inthe CSF.

The pairwise comparisons between the poor outcome groupand healthy controls support the results of time course analyses(Figs. 1–4) as significantly lower levels of AVP and OXT were foundin CSF of the poor outcome group (Table 2). In contrast, no differ-ences were detectable between patients with good outcome andhealthy controls for both neuropeptides in either compartment.With regard to plasma levels, it has to be recognized that AVPwas higher in patients with poor outcome compared to controls.This finding in the plasma compartment meets in part the resultsof an early study by Mather et al., who examined AVP levels inplasma and CSF of 42 patients immediately after aneurysmal SAHand found increased AVP levels in either or both compartmentsof 10 out of 22 patients with rather poor neurological status (Math-er et al., 1981). However, the elevated AVP levels in CSF describedby Mather et al. are not consistent with the lower CSF levels seen inour study, particularly in patients with poor outcome. In rodentmodels of SAH, the involvement of AVP in the development of cere-bral vasospasm was suggested (Trandafir et al., 2004; Nishihashiet al., 2005; Delgado et al., 1988). In our patients, the correlationbetween plasma AVP and TCD was rather weak (rho = +0.371)not favouring an effect of plasma AVP on cerebral vasospasm. Mod-est inverse correlations were seen between TCD measurementsand CSF neuropeptide levels reflecting higher cerebral blood flowvelocities in patients with poor outcome but a causal link is ratherunlikely. A different approach pursued by Zhu et al. to predict theclinical course of SAH patients used plasma copeptin that derivesfrom the same precursor as AVP, demonstrating prognostic valuefor the prediction of poor functional outcome but not for cerebralvasospasm (Zhu et al., 2011).

Our results suggest that CSF levels of AVP and OXT are reducedin patients with poor outcome after aneurysmal SAH indicatingthat hypothalamic damage occurs in aneurysmal SAH therebyaffecting neuropeptide levels in the central compartment. Histo-logical examinations in patients succumbing to SAH demonstratedhypothalamic lesions in 24 of 48 patients compared to patients dy-ing from other causes, including six associated with elevated intra-cranial pressure (Neil-Dwyer et al., 1994). Neil-Dwyer et al. furthernote that there may be a more subtle disturbance of hypothalamicfunction that could be demonstrated by other methods(Neil-Dwyer et al., 1994). This hypothalamic damage possibly re-flected by CSF but not plasma levels might be part of the immedi-ate pathophysiological events of early brain injury that occur afterthe initial bleed. As demonstrated by clinical and experimentalstudies, at the time of aneurysmal rupture, there is an acute riseof intracranial pressure, the extent of which reflects the severityof the bleeding, and a resultant decrease in cerebral perfusion pres-sure (Voldby and Enevoldsen, 1982; Fisher, 1975). There is sub-stantial interest in early brain damage and the mechanisms ofinjury during the acute phase of aneurysmal SAH although re-search has focused predominantly on cerebral vasospasm and cor-tical lesions (Cahill et al., 2006; Sehba and Bederson, 2006; Kusakaet al., 2004; Kassell et al., 1990). Several studies demonstrated thataneurysmal SAH induces diverse acute and long-term cognitive,psychosocial and neuropsychiatric disturbances (McKenna et al.,1989; Ogden et al., 1993; Uski et al., 2000; Hutter and Gilsbach,1993; Fertl et al., 1999; Caeiro et al., 2011; Powell et al., 2002).Possibly, the initial SAH insult constitutes the essential factor inthe production of persistent cognitive deficits (Berry et al., 1997;Ogden et al., 1993). However, it is comprehensible that outcomeassessment in a complex disease like SAH remains controversialand might never be precise (Macdonald, 2012; Giraldo et al.,2012). Biomarkers beyond clinical and radiological variables thatare used for the initial clinical grading at admission are neededto further improve the accuracy of outcome prediction.

The implication of neuropeptide alterations in the brain afteraneurysmal SAH is increasingly recognised and investigated (Barryet al., 2011; Schebesch et al., 2011; Uski et al., 2000; Ang et al.,2005; Dohi et al., 2005). OXT and AVP are involved in a wide rangeof behavioural and neuroendocrine regulation (Landgraf and Neu-mann, 2004). Central OXT exerts prosocial, anxiolytic and antide-pressive effects whereas AVP tends to show anxiogenic anddepressive actions (Neumann and Landgraf, 2012). An altered orunbalanced activity of both brain neuropeptide systems as seenin this study may contribute to poor functional outcome, cognitivedeficits or inappropriate socio-emotional behaviours. However, theestablishment of causal links between altered neuropeptide levelsand specific neurological impairments will be challenging and sub-tle gradations of neurological outcome may be difficult to explaingiven the plethora of neuropeptide and neurological aberrations.In a cohort of patients with rather good neurological outcome, Uskiet al. analysed AVP levels in CSF several months after the onset ofaneurysmal SAH with no significant correlation between scores onstandardized neuropsychological tests and AVP levels (Uski et al.,2000). Rather than a functional contribution to SAH-related symp-toms, the similar decrease in both AVP and OXT points towards po-tential biomarkers reflecting the extent of hypothalamic damage. It

J. Martin et al. / Neuropeptides 48 (2014) 91–96 95

is well known that central neuropeptide release and peripheralsecretion may be regulated independently (Landgraf and Neu-mann, 2004; Neumann and Landgraf, 2012); accordingly, a recentstudy in humans showed no significant correlative concentrationsin CSF and plasma (Kagerbauer et al., 2013). However, the failure todetect any correlation in SAH patients (Table 3) and particularlythe opposite alterations in CSF and plasma AVP require additionalexplanation beyond independent regulatory capacity. Perhaps CSFlevels are more sensitive to hypothalamic damage than plasma lev-els mainly due to the posterior pituitary as major AVP and OXTstore (Leng and Ludwig, 2008), but this ‘‘buffer’’ hypothesis needsconfirmation.

The strength of the present study is its clinical approach as ro-dent models of experimental SAH may not reliably represent thepathophysiological processes of spontaneous aneurysmal SAH inhumans. A further strength lies in the prolonged analysis periodof 14 days of sub-acute illness after the index bleed. To our knowl-edge, these issues have not been investigated before. In addition,analyses of AVP and OXT were performed in both CSF and plasmathus not relying solely on analyses of the plasma compartment,which may not adequately describe neuropeptide activity in theCSF (Kleindienst et al., 2004; Kagerbauer et al., 2013). However, ageneral dilemma of human studies on central neuropeptide releaseremains that the brain is not directly accessible; thus even concen-trations in CSF are coupled with the uncertainty that they justroughly reflect brain neuropeptide concentrations. A further limi-tation of this study is its pilot character and thus the small numberof 12 patients although more than 200 samples were analysed toadequately describe the concentration time courses. No conclu-sions can be drawn from this study on a neuropeptide concentra-tion course in patients who severely deteriorated during thecritical 2 week period after the index bleed (e.g., as a result of cere-bral infarctions, in particular ischaemic lesions of the cortex) sinceno patients with a good clinical grade on admission but poor neu-rological outcome were investigated (Table 1). Further studies willbe necessary to clarify this aspect.

In conclusion, we here present a novel relationship between theneuropeptides AVP and OXT and the functional outcome afteraneurysmal SAH indicating hypothalamic damage as an aetiologicfactor. Patients with poor outcome have lower neuropeptide levelsin the CSF than patients with good outcome while levels in plasmafail to reflect differences in outcome. The link between lower CSFlevels and damaged hypothalamus needs further clarification, forinstance by functional NMR in living patients. Further research willshed more light on the complex pathophysiological processes ofaneurysmal SAH including the functional implication of AVP andOXT, possibly contributing to the development of new therapeuticapproaches.

Conflicts of interest

None.

Financial disclosures

None.

References

Ang, B.T., Tan, W.L., Lim, J., Ng, I., 2005. Cerebrospinal fluid orexin in aneurysmalsubarachnoid haemorrhage – a pilot study. J. Clin. Neurosci. Official J.Neurosurg. Soc. Australasia 12, 758–762.

Barry, C.M., Helps, S.C., den Heuvel, C., Vink, R., 2011. Characterizing the role of theneuropeptide substance P in experimental subarachnoid hemorrhage. BrainRes. 1389, 143–151.

Berry, E., Jones, R.A., West, C.G., Brown, J.D., 1997. Outcome of subarachnoidhaemorrhage. An analysis of surgical variables, cognitive and emotionalsequelae related to SPECT scanning. Br. J. Neurosurg. 11, 378–387.

Bielsky, I.F., Young, L.J., 2004. Oxytocin, vasopressin, and social recognition inmammals. Peptides 25, 1565–1574.

Bosch, O.J., Neumann, I.D., 2008. Brain vasopressin is an important regulator ofmaternal behavior independent of dams’ trait anxiety. Proc. Natl. Acad. Sci. USA105, 17139–17144.

Caeiro, L., Santos, C.O., Ferro, J.M., Figueira, M.L., 2011. Neuropsychiatricdisturbances in acute subarachnoid haemorrhage. Eur. J. Neurol. Official J. Eur.Fed. Neurol. Soc. 18, 857–864.

Cahill, J., Calvert, J.W., Zhang, J.H., 2006. Mechanisms of early brain injury aftersubarachnoid hemorrhage. J. Cerebral Blood Flow Metab. official J. Int. Soc.Cerebral Blood Flow Metab. 26, 1341–1353.

Crompton, M.R., 1963. Hypothalamic lesions following the rupture of cerebral berryaneurysms. Brain J. Neurol. 86, 301–314.

Delgado, T.J., Arbab, M.A., Warberg, J., Svendgaard, N.A., 1988. The role ofvasopressin in acute cerebral vasospasm. Effect on spasm of a vasopressinantagonist or vasopressin antiserum. J. Neurosurg. 68, 266–273.

Dohi, K., Ripley, B., Fujiki, N., Ohtaki, H., Shioda, S., Aruga, T., Nishino, S., 2005. CSFhypocretin-1/orexin-A concentrations in patients with subarachnoidhemorrhage (SAH). Peptides 26, 2339–2343.

Donaldson, Z.R., Young, L.J., 2008. Oxytocin, vasopressin, and the neurogenetics ofsociality. Science 322, 900–904.

Doshi, R., Neil-Dwyer, G., 1980. A clinicopathological study of patients following asubarachnoid hemorrhage. J. Neurosurg. 52, 295–301.

Fertl, E., Killer, M., Eder, H., Linzmayer, L., Richling, B., Auff, E., 1999. Long-termfunctional effects of aneurysmal subarachnoid haemorrhage with specialemphasis on the patient’s view. Acta Neurochir. 141, 571–577.

Fisher, C.M., 1975. Clinical syndromes in cerebral thrombosis, hypertensivehemorrhage, and ruptured saccular aneurysm. Clin. Neurosurg. 22, 117–147.

Fisher, C.M., Kistler, J.P., Davis, J.M., 1980. Relation of cerebral vasospasm tosubarachnoid hemorrhage visualized by computerized tomographic scanning.Neurosurgery 6, 1–9.

Gabor, C.S., Phan, A., Clipperton-Allen, A.E., Kavaliers, M., Choleris, E., 2012.Interplay of oxytocin, vasopressin, and sex hormones in the regulation ofsocial recognition. Behav. Neurosci. 126, 97–109.

Giraldo, E.A., Mandrekar, J.N., Rubin, M.N., Dupont, S.A., Zhang, Y., Lanzino, G.,Wijdicks, E.F., Rabinstein, A.A., 2012. Timing of clinical grade assessment andpoor outcome in patients with aneurysmal subarachnoid hemorrhage. J.Neurosurg. 117, 15–19.

Gossen, A., Hahn, A., Westphal, L., Prinz, S., Schultz, R.T., Grunder, G., Spreckelmeyer,K.N., 2012. Oxytocin plasma concentrations after single intranasal oxytocinadministration – a study in healthy men. Neuropeptides 46, 211–215.

Hackett, M.L., Anderson, C.S., 2000. Health outcomes 1 year after subarachnoidhemorrhage: an international population-based study. The AustralianCooperative Research on Subarachnoid Hemorrhage Study Group. Neurology55, 658–662.

Hunt, W.E., Hess, R.M., 1968. Surgical risk as related to time of intervention in therepair of intracranial aneurysms. J. Neurosurg. 28, 14–20.

Hutter, B.O., Gilsbach, J.M., 1993. Which neuropsychological deficits are hiddenbehind a good outcome (Glasgow = I) after aneurysmal subarachnoidhemorrhage. Neurosurgery 33, 999–1005 (discussion 1005–1006).

Jarcho, M.R., Mendoza, S.P., Mason, W.A., Yang, X., Bales, K.L., 2011. Intranasalvasopressin affects pair bonding and peripheral gene expression in maleCallicebus cupreus. Genes Brain Behav. 10, 375–383.

Jennett, B., Bond, M., 1975. Assessment of outcome after severe brain damage.Lancet 1, 480–484.

Johnston, S.C., Selvin, S., Gress, D.R., 1998. The burden, trends, and demographics ofmortality from subarachnoid hemorrhage. Neurology 50, 1413–1418.

Kagerbauer, S.M., Martin, J., Schuster, T., Blobner, M., Kochs, E.F., Landgraf, R.,2013. Plasma oxytocin and vasopressin do not predict neuropeptideconcentrations in the human cerebrospinal fluid. J. Neuroendocrinol. 25,668–673.

Kassell, N.F., Torner, J.C., Haley Jr., E.C., Jane, J.A., Adams, H.P., Kongable, G.L., 1990.The international cooperative study on the timing of aneurysm surgery. Part 1:Overall management results. J. Neurosurg. 73, 18–36.

Kleindienst, A., Hildebrandt, G., Kroemer, S.A., Franke, G., Gaab, M.R., Landgraf, R.,2004. Hypothalamic neuropeptide release after experimental subarachnoidhemorrhage: in vivo microdialysis study. Acta Neurol. Scand. 109, 361–368.

Kuboyama, T., Hashimoto, H., Ueguchi, T., Yamaki, T., Hirakawa, K., Noto, T.,Nakajima, T., 1988. Diurnal changes in vasopressin and oxytocin levels incerebrospinal fluid of post-operative patients with intracranial aneurysms.Endocrinologia Japonica 35, 249–254.

Kusaka, G., Ishikawa, M., Nanda, A., Granger, D.N., Zhang, J.H., 2004. Signalingpathways for early brain injury after subarachnoid hemorrhage. J. CerebralBlood Flow Metab. Official J. Int. Soc. Cerebral Blood Flow Metab. 24,916–925.

Landgraf, R., Neumann, I.D., 2004. Vasopressin and oxytocin release within thebrain: a dynamic concept of multiple and variable modes of neuropeptidecommunication. Front. Neuroendocrinol. 25, 150–176.

Leng, G., Ludwig, M., 2008. Neurotransmitters and peptides: whispered secrets andpublic announcements. J. Physiol. 586, 5625–5632.

Linn, F.H., Rinkel, G.J., Algra, A., van Gijn, J., 1996. Incidence of subarachnoidhemorrhage: role of region, year, and rate of computed tomography: a meta-analysis. Stroke J. Cerebral Circ. 27, 625–629.

96 J. Martin et al. / Neuropeptides 48 (2014) 91–96

Lukas, M., Toth, I., Reber, S.O., Slattery, D.A., Veenema, A.H., Neumann, I.D., 2011.The neuropeptide oxytocin facilitates pro-social behavior and prevents socialavoidance in rats and mice. Neuropsychopharmacol. Official Publ. Am. Coll.Neuropsychopharmacol. 36, 2159–2168.

Macdonald, R.L., 2012. Outcome after aneurysmal subarachnoid hemorrhage. J.Neurosurg. 117, 12–14 (discussion 14).

Macdonald, R.L., Pluta, R.M., Zhang, J.H., 2007. Cerebral vasospasm aftersubarachnoid hemorrhage: the emerging revolution. Nat. Clin. Pract. Neurol.3, 256–263.

Mather, H.M., Ang, V., Jenkins, J.S., 1981. Vasopressin in plasma and CSF of patientswith subarachnoid haemorrhage. J. Neurol. Neurosurg. Psychiatry 44, 216–219.

Mayer, S.A., Kreiter, K.T., Copeland, D., Bernardini, G.L., Bates, J.E., Peery, S., Claassen,J., Du, Y.E., Connolly Jr., E.S., 2002. Global and domain-specific cognitiveimpairment and outcome after subarachnoid hemorrhage. Neurology 59, 1750–1758.

Mcburney, D.M., White, T.L., 2004. Research Methods. Wadsworth, Belmont, CA.McKenna, P., Willison, J.R., Phil, B., Lowe, D., Neil-Dwyer, G., 1989. Cognitive

outcome and quality of life 1 year after subarachnoid haemorrhage.Neurosurgery 24, 361–367.

Neil-Dwyer, G., Lang, D.A., Doshi, B., Gerber, C.J., Smith, P.W., 1994. Delayed cerebralischaemia: the pathological substrate. Acta Neurochir. 131, 137–145.

Neumann, I.D., Landgraf, R., 2012. Balance of brain oxytocin and vasopressin:implications for anxiety, depression, and social behaviors. Trends Neurosci. 35,649–659.

Nishihashi, T., Trandafir, C.C., Wang, A., Ji, X., Kurahashi, K., 2005. Enhancedreactivity to vasopressin in rat basilar arteries during vasospasm aftersubarachnoid hemorrhage. Eur. J. Pharmacol. 513, 93–100.

Ogden, J.A., Mee, E.W., Henning, M., 1993. A prospective study of impairment ofcognition and memory and recovery after subarachnoid hemorrhage.Neurosurgery 33, 572–586 (discussion 586–587).

Powell, J., Kitchen, N., Heslin, J., Greenwood, R., 2002. Psychosocial outcomes at 3and 9 months after good neurological recovery from aneurysmal subarachnoidhaemorrhage: predictors and prognosis. J. Neurol. Neurosurg. Psychiatry 72,772–781.

Rosengart, A.J., Schultheiss, K.E., Tolentino, J., Macdonald, R.L., 2007. Prognosticfactors for outcome in patients with aneurysmal subarachnoid hemorrhage.Stroke J. Cerebral Circ. 38, 2315–2321.

Sala, M., Braida, D., Lentini, D., Busnelli, M., Bulgheroni, E., Capurro, V., Finardi, A.,Donzelli, A., Pattini, L., Rubino, T., Parolaro, D., Nishimori, K., Parenti, M., Chini,B., 2011. Pharmacologic rescue of impaired cognitive flexibility, social deficits,increased aggression, and seizure susceptibility in oxytocin receptor null mice:a neurobehavioral model of autism. Biol. Psychiatry 69, 875–882.

Schebesch, K.M., Brawanski, A., Kagerbauer, S.M., Martin, J., Bele, S., Herbst, A., Feigl,G., Stoerr, E.M., Lohmeier, A., Proescholdt, M., 2011. The possible role ofneuropeptide Y after spontaneous subarachnoid hemorrhage. ActaNeurochirurgica 153, 1663–1668 (discussion 1668).

Sehba, F.A., Bederson, J.B., 2006. Mechanisms of acute brain injury aftersubarachnoid hemorrhage. Neurol. Res. 28, 381–398.

Teasdale, G.M., Drake, C.G., Hunt, W., Kassell, N., Sano, K., Pertuiset, B., de Villiers,J.C., 1988. A universal subarachnoid hemorrhage scale: report of a committee ofthe World Federation of Neurosurgical Societies. J. Neurol. Neurosurg.Psychiatry 51, 1457.

Trandafir, C.C., Nishihashi, T., Wang, A., Murakami, S., Ji, X., Kurahashi, K., 2004.Participation of vasopressin in the development of cerebral vasospasm in a ratmodel of subarachnoid haemorrhage. Clin. Exp. Pharmacol. Physiol. 31, 261–266.

Uski, T.K., Lilja, A., Saveland, H., Ekman, R., Sonesson, B., Brandt, L., 2000. Cognitivefunctioning and cerebrospinal fluid concentrations of neuropeptides forpatients with good neurological outcomes after aneurysmal subarachnoidhemorrhage. Neurosurgery 47, 812–818.

van Gijn, J., Kerr, R.S., Rinkel, G.J., 2007. Subarachnoid haemorrhage. Lancet 369,306–318.

Voldby, B., Enevoldsen, E.M., 1982. Intracranial pressure changes followinganeurysm rupture. Part 1: Clinical and angiographic correlations. J. Neurosurg.56, 186–196.

Zhu, X.D., Chen, J.S., Zhou, F., Liu, Q.C., Chen, G., Zhang, J.M., 2011. Detection ofcopeptin in peripheral blood of patients with aneurysmal subarachnoidhemorrhage. Crit. Care 15, R288.