Content uploaded by Dora Hopf
Author content
All content in this area was uploaded by Dora Hopf on Aug 24, 2020
Content may be subject to copyright.
Journal of Neuroendocrinology. 2020;32:e12887. wileyonlinelibrary.com/journal/jne
|
1 of 24
https://doi.org/10.1111/jne.12887
Received:16September2019
|
Revised:26M ay2020
|
Accepted:14June2020
DOI : 10.1111/j ne.1 2887
EDITOR INVITED REVIEW
Neuroendocrine mechanisms of grief and bereavement:
A systematic review and implications for future interventions
Dora Hopf | Monika Eckstein | Corina Aguilar-Raab | Marco Warth |
Beate Ditzen
Instit uteofMedicalPs ychology,Heidelberg
UniversityHospital,Ruprecht-Karls
UniversityHeidelberg,Heidelberg,Germany
Correspondence
BeateDit zen,Institu teofMedical
Psychology,HeidelbergUniversityHospital,
Ruprecht-KarlsUniversityHeidelberg,
BergheimerStr aße20,69115,Heidelb erg,
Germany.
Email:beate.ditzen@med.uni-heidelberg.de
Funding information
MarsiliusKollegatHeidel bergUniversity;
OlympiaMorat aProgr amatHeid elberg
University;FA ZITStif tung
Abstract
Bereavementisassociatedwithmanynegativebehavioural,psychologicalandphysi-
ological consequences and leads to an increased risk of mortality and morbidity.
However, studies specifically examining neuroendocrine mechanisms of grief and
bereavement h aveyet to b e reviewed. This sys tematic review is a synth esis of the
latest evidence in this field and aims to draw conclusions about the implications of
neurobiologicalfindingsonthedevelopmentofnewinterventions.PRISMAguidelines
for systematic reviews were used to search for articles assessing neuroendocrine cor-
relatesofgrief.Findingswerequalitativelysummarised.TheNationalHeart,Lung,and
BloodInstituteStudyAssessmentToolwasusedtoassessthequalityoftheincluded
studies.Out of460papers,20 mettheinclusion criteria.However,mostwereoffair
qualityonly.Asaneuroendocrinemarker,themajorityofthestudiesreportedcortisol
astheoutcomemeasureandfoundelevatedmeancortisollevels,flatteneddiurnalcor-
tisol slopes and higher morning cortisol in bereaved subjects. Cortisol alterations were
moderatedbyindividualdifferencessuch as emotional reaction to grief,depressive
symptoms,griefseverity,closeness to the deceased andageorgender.Research on
neuroendocrine mechanisms of grief is still in its early stages regarding grief measures
andtheuseandtimingofneuroendocrineassessments.Mostofthestudiesfocuson
cortisolasoutcome,andonlylimiteddataexistonotherbiomarkerssuchasoxytocin.
Future research might consider assessing a broader range of neuroendocrine markers
and use longitudinal designs with a focus on the psychobiological reactions to loss.
Basedonthis,individuallytailoredpsychosocialinterventions,possiblyinthepalliative
carecontext,mightbedevelopedtopreventprolongedgriefdisorder.
KEYWORDS
bereavement,cortisol,grief,hypothalamic-pituitary-adrenal-axis,oxytocin,stress,trauma
Thisisanop enaccessarti cleundertheter msoftheCreativeCommonsAttributionL icense,whichpe rmitsuse,dis tribu tionandreprod uctioninanymed ium,
provide d the original wor k is properly cited.
©2020TheAuthors.Journal of NeuroendocrinologypublishedbyJohnWiley&SonsLtdonbehalfofBritishSociet yforNeuroendo crinology
2 of 24
|
HOPF et al.
1 | INTRODUCTION
1.1 | Social loss and its consequences
Thelossofalovedpersonisoneofthemostdevastatingexperiences
inlifeandisassociatedwithpsychological,behaviouralandphysiologi-
cal changes in the survivingclose persons. The term loss is referred
totheactuallossevent,whereasgrief entails the subjective reactions
thata reassociatedw it hl os s.Al th ou gh gr ie falsooccu rsaf tera se pa ra-
tion, in this review,wefocus exclusively on griefafter an actual loss
of a loved one through death. Bereavement is defined as the state of
having suffered the loss of a loved one and entails the time after a loss
duringwhichgriefisexperienced.1Physiologicalre ac tionstobereave-
mentinclude neuroendocrine,immunologicaland somatic changes.2
Psychological consequences include insecurity, anxiety, aggression
and depr essive and (psych o-)s omatic sympto ms,3 which result in a
greatervulnerabilitytosomaticorpsychiatricproblems,suchascardi-
ovascular diseases4 or clinical depression.5,6 Some studies even associ-
atelosswithincreasedmortalityamongthesurvivors,7-10 highlighting
the massi ve effects of t his experien ce. More recentl y,fo r example,
systematic research revealed that social loss triggers the development
of Takotsubo cardiomyopathy, or “Broken Hear t Syndrome”. This
syndro me is a reversible, s tress-induced c ardiomyopathy th at mim-
ics acute myocardial infarction and occurs after intense emotional or
physical stress.11Abovet his,pat ientswithTakot suboc ar diomyopathy
have a higher prevalence of neurological or psychiatric disorders than
those with an acute coronary syndrome.12
Thepreviou sl yd escribe dn on -p at holog ic almourn in gp ro ce ssis an
adaptive r esponse and u sually has no lon g-term negative ef fects.13
If,however,grievingcontinues andsymptomsoccur,thatarebeyond
typical grief, Prolonged Grief Disorder (PGD) or Persistent Complex
Bereavement Disorder(PCBD)canbediagnosed.PGDischaracter-
isedby longingforand preoccupationwiththedeceased, along with
emotional distress and significant functional impairments that persist
beyond 6 months after the loss of a significant other.14Approximately
10%-20% of mourners developPGD/PCBD.15-1 8 The diagn osis has
only recently been added to the latest versions of the International
Classification ofDiseases (ICD-XI, PGD) and the Diagnostic Manual
forPsychiatricDisorders(DSM-5,PCBD),19 and led to debate about
the defining criteria and consequences.7,16,20,21ThetermComplicated
Grief(CG),whichwasoriginallydevelopedtodistinguishgrieffrom
depression,22doesnotrepresenttheofficialdiagnosisbut,instead,
comprises a larger category with diagnostic disordered grief encom-
passing a smaller group.23Thisdis ti nc tio nhastob ekep tinmindwhe n
interp re ti ngempir ic alstu di esongrief.Inth ef ol lo wi ng,wee mpl oyth e
original terms used in the studies in each case.
1.2 | Psychobiological models of pair bond
formation and bond disruption
Thedeathofalovedonegoesalongwith severalpsychosocialcon-
sequen ces: loneliness , a disruption in da ily routines, a s ubstantial
loss of coherence, impaired sleep, and, most centrally,being sepa-
rated from the loved person. All ofthese factors individually have
beenassociatedwithpoorhealthoutcomes.Forexample,loneliness
enhancestheriskofmorbidity and mortalit y,24 elevates cardiovas-
cularactivation,25 leads to cortisol dysregulation26 -28 and is associ-
ated with a greater utilisation of health care institutions.29Alower
level of sense of coherence is associated with increased burden in
caregivers of patients with chronic illness.30Additionally,poorsleep
quality is associated with blunted cor tisol awakening responses. 31As
the above mentioned psychosocial consequences all come together
ingrieving sur vivors,itcanbeassumed thatthoseneuroendocrine
and psychological changes may be even more pronounced in those
who suffer intensely from the loss.
Inthiscontext,attachment andattachment disruption theories
give impor tant indications towards a better understanding of grief
anditsrole inphysicalandmental health.SbarraandHazan32 pos-
tulated that understanding the functionality and cause of human
adult attachment could give us deeper insights into human coreg-
ulation and biobehavioural reactions to loss. According to their
model,32 relationships function as interpersonal regulatory sys-
tems.Interpersonalregulationmeansthatcouplesco-regulatetheir
emotionaland behavioural responses, which serves asanadaptive
mechanism that is less effor tful and more automatic than individu-
ally regulating them.32,33Thedisruptionofarelationshipendsthese
regulatorybenefitsandleadstostress-relatedgriefresponses(dys-
regulation).The main t ask in coping with losswould be to manage
dysregulationbyusingbehavioural,emotionalorcognitivestrategies
(functional self-regulation), which then atte nuate the physiolog ical
consequ ences. Accord ing to the model , the initial re action to lo ss
not only involves psychological, but also physiological changes ac-
companied by psychological reactions.32The ref ore ,i tisim p ort a ntto
know the associated biological mechanisms of grief to predict neg-
ative psychological changes and to prevent grieving persons from
long-termnegativeeffectssuchasPGD/PCBD.
On the neuroendocrine level, grief might be primarily associ-
atedwith an unspecificneuroendocrine stress-reaction,especially
hypothalamic-pituitary-adrenal (HPA) axis activity. HPA axis acti-
vation leadstothe synthesis of corticotrophin-releasing hormones
(CRH)andvasopressin(VP),stimulatingthesecretionofadrenocor-
ticotrophic hormones (ACTH) into the peripheral circulation.34 As
aresult, ACTH induces glucocor ticoid(e.g., cortisol) release in the
adrenal gland, leading to a negative-feedback inhibiting HPA axis
activation in the brain.34,35 Cortisol secretion normally reaches its
peak 30 -45 minutes afterawakening(cortisolawakeningresponse
[CAR]),followedbyasubsequentdeclineduringthedayandreach-
ing its lowest pointbetweenmidnightand 5.00am36, 37 Besides its
stress-dependence,ahealthyHPAaxis function shows strongdiur-
nalpatterns,anddeviationsfromthetypicaldeclinethroughoutthe
day provide v aluable infor mation regard ing the role of the a xis in
disease processes. Cortisol can be measured in several ways. Basal
urinar y free cortisol is often used to interpret aggregated cortisol
levels. Hair or nail samples indicatehormonesecretionoverweeks
or even months.36 Recent studies have star ted to examine the
|
3 of 24
HOPF et al .
circadi an rhythm of co rtisol by eval uating a stro ng CAR and dai ly
pattern of pronounced cortisol decreases during the day as indica-
torsofahighlyfunctionalfeedback-sensitivityoftheHPAaxis.37
As an add itional neuro modulator, oxy tocin (OT) is a hypo tha-
lamic neuropeptide that, after secretion from the paraventricular
nucleus ofthehypothalamus (PVN) and supraoptic nucleus(SON),
is stored in the posterior pituitar y lobe38 and released into the pe-
ripheralbloodcircuitandintocentral-nervousbrainareas,asparts
oft he pa inne t wo rk an dt here ward-syst em ,39OTi nt er act swiththe
HPAaxissystembyaccompanyingitsresponsetoagiven stressor
andexer tingstress-reducingeffects,forexampleheartrate,blood
pressure and cortisol level decrease.40- 42 OT plays an important
role in the formation and maintenance of social relationships.43,44
In turn, the OT system is also altered after the disruption of a
relationship.45
1.3 | Neuroendocrine changes after social loss
in animals
In the history of research on neurobiological changes af ter social
lossinhumans,re searchersoftenre liedonanimalmo delsofsepara-
tion and lo ss. More spec ifically, they b egan to examine n eurobio-
logical factors of social loss in the prairie vole (Microtus ochraster),
which ser ves as an animal model of human social loss. In these mo-
nogamous rodents, theloss of a companionis associated with the
activation ofthe HPAaxiswith higherbasal plasmacorticosterone
concentrations46-48 and adrenal hypertrophy.49
Volemothersshowsignificantincreasesinthecorticotrophin-re-
leasingfactor(CRF)mRNAexpressioninthePVN,46 when separated
fr o mt h eir p ups . Int ere s t ing l y,the s tre s sre s pon s eto sep a r at i onc a nbe
reducedthroughtheperipheral,subcutaneousapplicationofOT.50, 51
Theseparationfromanadultattachmentfigureinvolesleadstode-
creased OTmRNAexpression inthePVN49 and increased density of
OT-imm uno rea cti vecel ls in the PV Nan dt h eS ON. Thela tte rh asbee n
interpretedasa consequenceof adecreased releaseandlimited OT
receptor activity in reaction to loss.48Furthermore,OTfibressignal-
lingtotheNAccshowdecreasedac tivationafterlossinvoles.44
Translating these effectsof OT to humanattachment,one can
assume that theOTsystem is alsoinvolved in socialloss in human
beings.NeuroendocrinemechanismsinvolvingOThavealreadybeen
discussed with relevance for different mental disorders.52Although
they might only serve as one of many response domains af ter the
deathofabelovedperson,theycouldbeakeymediatorintherela-
tionship between grief and the development of psychiatric disorders
suchasPGDorPCBD.32 Deviations from functional neuroendocrine
stress responses have already shown to be involved in response to
trauma53-55 and could possibly ser ve as a prognostic indicator for
the development of grief-related psychopathology. Furthermore,
important implications could be derived regarding preventive psy-
chosocial interventions before the death of the close person in order
toenhanceco-regulation,aswellastheawarenessoftheupcoming
relationship disruption.
Tod ate, a number of ar ticles exis t reviewing lite rature on the
neuroendocrine mechanisms of grief, although they either exclu-
sively focus on animal studies44,45 or on prolonged grief in the con-
textofonlyoneneuroendocrinemarker.56Therefore,theaimofthe
current wo rk is to extend t he existing li terature by sys tematical ly
reviewing studies investigating neuroendocrine mechanisms in the
early st age of grief with potential predic tive value for long-term
pathological reactions to loss.
2 | MATERIALS AND METHODS
2.1 | Search strategy and eligibility criteria
A systematic literature review was per formed according to the
PreferredReportingItemsforSystematicReviewsandMeta-Analysis
(PRISMA)guidelines.57ABooleansearchwasusedtofindthewide
rangeofstudiesreportingneuroendocrinemechanismsofgrief.The
search terms were (grief OR bereavement OR bereaved OR "bond
disruption"OR"socialloss"OR"bondloss"ORsorrowORmourning)
AND(neuroendocrineORendocrine ORneurobiol*ORpsychobiol*
ORpsychophysiol*ORbiomarker*).Theinitialsearchwasperformed
in 22 March 2019 and upda ted on 23 April wit hin four large d ata-
bases including Web of Science(http://webofknowledge.com),CINAHL
(https://www.ebscohost.com/nursing/products/cinahl-databases),
PubMed (https://pubmed.ncbi.nlm.nih.gov) and PsycINFO ( ht t p ://
www.apa.org/psycinfo). The authors repeated the search on 13
November2019byaddingmorespecificneuroendocrinewords(oxy-
tocinOROXTOROTORcort*ORinsulinORpro lactinORendorphin
ORcatecholamin*)tofindalltherelevantarticlesconcerningspecific
neuroendocrinechanges after bereavement. Additionally,reference
listsofrelevantreviews,primarystudies,andtheoreticalframeworks
were searched for potential ar ticles.6,17,32,43-45,58-62 Two independ-
entreaders(DHandHM)screenedthearticleabstractsandreadthe
selected full-text articles inorder to decide whether to include or
exclude the articles according topredefinedcriteria.Non-consistent
decisionswerediscusseduntilconsensuswasreached.Theeligibility
criteria for the studies were:
Inclusion criteria:
• Original study.
• Neuroendocrinemarkers (cortisol, epinephrine, norepinephrine,
OT,insulin,prolactin,endorphin)investigated.
• Population:humanadult s(>18years)wholostabelovedperson
(partner,familymember,closefriend).
• ArticleavailableinEnglish.
Exclusioncriteria:
• Experimentallyinducedgrief.
• Griefreactionsdidnotoccurasaresultofdeath(eg,griefrelated
todepressionorpost-traumaticstressdisorder(PTSD);griefafter
divorceorbreakup).
4 of 24
|
HOPF et al.
• ArticlenotavailableinEnglish.
• Noneuroendocrinemarkersassessed.
• Child or adolescent population (<18years).
2.2 | Data extraction
Relevant data of the incorporated studies, including publication
date, study design, sample characteristics, grief assessment tools,
neuroendocrinemeasureandresults,were extracted forqualitative
data analyses. Study quality was assessed independently by three au-
thors (DH,ME and CAR) using theNationalHeart ,Lung, andBlood
Institute(NHLBI)QualityAssessmentToolforObservationalCohort
andCross-SectionalStudies.63Thistoolconsistsof14items(includ-
ing18sub-items)assessingkeyissuesof thestudy'sinternalvalidity;
forexample,populationrecruitment,statisticalpowerconsiderations,
assessmentof exposureandoutcome variables andconsideration of
confoundi ng variable s. The crite ria can be met , not met, c annot be
determined,arenotapplicableor not repor ted. The ratersdiscussed
their ratings to resolve discrepancies and come to a final decision.
Studies wererated as “good”,“fair” or “poor” to describe therisk of
bias. A “good” quality rating indicates the least risk of bias. We de-
cided toratestudiesas“good”iftheymetmorethan 2/3ofthecri-
teria.A“fair”ratingindicatesthatthestudyshowshigherriskofbias
butnotenoughtoinvalidateresults.Studieswereratedas“fairifthey
metatleast half ofthe criteria.A“poor” ratingindicateshighriskof
bias that could significantly compromise the accuracy of the results.
Studieswereratedas“poor”iftheymetlessthanhalfofthecriteria.
3 | RESULTS
In total, 677 papers were found during the systematic search
(Figure1), from which469articlesremainedafterremovingdupli-
cates. After screening theabstracts, 39 articles remainedfor full-
texteligibilitysearch. Sixarticleswereexcludedbecausetherewas
no full-tex t available onl ine,64-69 one stud y was excluded b ecause
it exclusively examined heart rate variabilityand cytokine produc-
tion sys tem,70 one other study investigated receptor genes only71
and five studies investigated early parental loss in childhood and
FIGURE 1 PRISMAflowchartonthestudyselectionprocess
Records identified through database
searching: PubMed (n = 146);
PsychInfo (n = 183);
Web of Science (n = 188);
CINAHL (n = 57)
Records after duplicates removed
(n = 469)
IdentificationScreeningEligibilityIncluded
Records screened
(n = 469)
Full-text articles assessed
for eligibility
(n = 39)
Studies included in
qualitative synthesis
(n = 26)
Records excluded
(n = 430)
Full-text articles excluded (n = 13)
Reasons:
Full-text not available (n = 6)
Wrong outcome (n = 2)
Wrong study population (n = 5)
Additional records identified through
database searching: Specific
neuroendocrine measures entered,
reviews and similar articles search
(n = 103)
|
5 of 24
HOPF et al .
weretherefore excluded.72 -76Thefinalsampleconsistedof26ar-
ticles published between 1986 and 2019.According to the Quality
assessmentratings,five studies showedgoodqualit y,77-8 1 whereas
21 studies showed fair quality.82-1 02 Th er esult sareshown in Table1.
3.1 | Mean cortisol level
Five studies examined the association between bereavement and
mean cortisol levels.Jacobs et al84 compared56bereavedwithnon-
bereaved spouses, both1and2months afterhospitalisationoftheir
spouse. Itwashypothesisedthatadultswithrisingseparation anxiety
and dist ress during bere avement would show hig her cortisol l evels than
thosewithloweranxiety.Theycollec ted24-hoururinaryfreecortisol
on three separate days in the week before the second interview and
averagedthedailyvaluesofcortisol.Par ticipantswithhighseparation
anxietyshowedhigher cortisollevelsthanthosewhoseanxietylevel
fellfr om 1t o2mo nt hsaft erhos pi tal is ation .T her ew as nodif fe ren ce in
cortisol levels between the bereaved and the anticipatory bereaved.84
Irwin et al92 assessed cortisol weekly over 1 to 2 months in 28 recently
bereaved,anticipator ybereaved,ornon-bereavedwomen.Theyfound
significantly higher cortisol levels in the bereaved compared to the
other controls. Spratt and Denney93examinedtheeffectofsudden
childdeathoncor tisollevelsin18bereavedvsnon-bereavedparents.
Theyreport nodifferencesin cortisollevelsbetween thetwogroups.
One further study compared cortisol levels between 260 bereaved
and262non-bereavedmenandwomenatthesame time as control-
lingfordepressivesymptoms.Levelswereassessedatonetime-point
after the psychiatric interview. No significant differences between
the two groups were found.79 Mintonet al96 investigated changes in
physio logic alstr ess11, 12 an d13m ont hs af ter lo ssin47wi do ws.Th ey
compared mean morning and evening cortisol levels and hypothesised
that duri ng the firs t anniversa ry after t heir loss, phys iological s tress
levelwouldbethehighest.However,nosignificantdifferencesincor-
tisollevelswerefound.Theauthorssuggestthattheanniversarydoes
notrepresent an immediate stressor, not being sufficiently salientto
change neuroendocrine stress levels.96Andersen et al97 investigated
the psychological and physical health effects of repeated loss among
university students after clustered peer deaths. Cortisol was meas-
uredviahairsamples3monthsaftertheloss.Asignificantassociation
ofprior bereavementexperiences with hair cortisol level wasfound,
as well as a significant negative relationship between the number of
bereavement experiences and cortisol levels. The latter finding is in-
terpreted in the way that people with prior bereavement maintain av-
erage leve ls of cortisol a cross the ext ended perio d of loss, where as
thosewithnopriorexperiencedisplaydysregulatedcortisollevels.97
3.2 | Morning/evening cortisol
Twostudiesinvestigated bereavement inthecontext ofmorningcor-
tisol. Buckley et al80assessedmorningcortisolin62bereavedand50
non-bere aved men and wome n 2 weeks and 6 mon ths post-loss by
taking one sampleeachmorning.Theyfoundsignificantly highercor-
tisol morning levels in the bereaved compared to the controls at both
time-points.80 The second study examined whetherovernight basal
urinar y free corti sol 12 months aft er loss depend ed on gender, the
emotional reaction toloss (emotional numbness) and circumstances
ofspousalbereavement(prolongation)whichwereassessed6months
post-loss.Itwashypothesisedthatlongerforewarningofdeath(“How
longbeforeyourspouse'sdeathdidyourealisethats/hewasgoingto
die? ”)andh igheremot io na lnumbnesswouldbeassociate dwithhi gh er
cortisoldysregulation(highercortisollevels).78Asexpected,prolonged
forewarning was significantly associated with elevated cortisol levels.
During 6-12 months, cort isol levels increased in widowers and de-
creased in widows. Bereaved men with emotional numbness at time
1 had higher cortisol levels at time 2 compared to men without emo-
tionalnumbness.Thisassociationwasnotfoundinwomen.78
3.3 | Diurnal patterns of cortisol
Four stud ies examined d iurnal cort isol patter ns. Ong et al90 com-
pared morningcortisol,CAR and cortisol slopesacross the daybe-
tween22bereavedand22non-bereavedadults.Theyhypothesised
that affect moderated the relationship bet ween bereavement and
HPAaxisdysfunction.90Significantly lowercortisolwake-uplevels
and flatter diurnal cortisol slopes were found in the bereaved com-
paredtothenon-bereavedadults.Theresultswerealsopartlyinline
withthemediationhypothesis:pre-topost-losschangesinpositive
affect accounted for 29% of the ef fect of spousal loss on diurnal cor-
tisol slopes.90 Similar results were found in a small sample (n =12)of
studyparticipants suffering fromCG.81 Only participants with CG
showedflatteneddiurnalcortisolslopes,whereasparticipantsexpe-
riencing normal grief did not. It was proposed that only individuals
experiencingaprolongedreactiontolossmightdeveloppermanent
HPAaxisdysregulation.81Bycontrast,Hollandetal82 found dysreg-
ulatedHPAaxisfunctionindependentofgrieflevel.Theycompared
diurnalcortisollevelsbetween56depressedcontrolsanddepressed
bereavedmenandwomenwithorwithoutelevatedPGDsymptoms.
Significantlylowercortisolwake-uplevelsandflatterdiurnalslopes
werefoundinthedepressedbereavedPGDgroupcomparedtothe
depressedcontrols. The differencesincortisollevels between the
depressedbereavedwithPGDandthedepressedbereavedwithout
PGD were no t statist ically signi ficant. O n a descript ive level, men
and women who had lost a spouse showed greater cortisol dys-
regulation than those who lost someone else than their partner.82
Itwassuggestthat,accordingtotheresults,thelossofalovedone
is predictive of more dysregulated cortisol, irrespective of one's
level of PGD s ymptoms. Pe réz et al77 investigated diurnal cortisol
levelsboth2yearsand5 yearspost-lossin CGsuffererscompared
to normal g rieving men and wo men and contro ls in a population -
based sample of208 4adult s. Significantlylower morning cortisol
andoverallcortisollevels(representedbytheareaunderthecurve)
werefoundintheCGgroupcomparedtothehealthygrieversattime
1.Nosignificantdifferenceswerefoundregardingthediurnalslope.
6 of 24
|
HOPF et al.
TABLE 1 Result s of the qualitative systematic review grouped by outcome
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Mean cortisol level
Jacobs
etal(1987)84
Longitudinal:t wotime-points:first
interview 1 month after hospitalisation
ofpartner(1),secondinterview
2monthsaf terhospitalisation(2)
2 × 2 groups
Bereaved vs anticipatory bereaved
Risingseparationanxietyvsdeclining
separationanxietyfrom(1)to(2)
Spouse N =56
n(bereaved)=40
n (anticipatory
bereaved)= 16
(50)
62.6(NR) SA 24-hoururinaryfreecortisol
Assessmenttimes:threeseparate
daysintheweekbeforetime-
point(2)
Group,inwhichseparationanxietyrose
from(1)to(2)hadsig.highercortisol
levelsthangroupinwhichSAdiminished
or dropped
sig.Highercortisollevelswerefoundboth
for the bereaved and the anticipatory
bereaved subjects
Fair
Irwinetal(1988)92 Longitudinal:
weekly assessment of cortisol in a
1-2monthperiod
<6monthspost-loss
(1)Recentlybereavedwomenvs
(2)womenwithterminallyill
husbands vs
(3)womenwit hhealthyhusbands
Spouse N = 28
n(1)= 9
n(2)= 11
n (39 = 8
(100 )
52.2(3.4) None Plasmacortisol sig.Highermeanplasmacortisollevelsin
group(1)comparedtogroup(3)
Notsignificant:Plasmacortisollevelin
group(2)and(3)
Fair
Spratt&Denney
(1991) 93
Longitudinal:
fourtime-points:2,4,6and8months
post-loss(suddenloss)
Suddenlybereaved(1)vsnon-
bereaved(matchedtobereaved)
Child N = 18
n(1)= 9
(66)
(1)=49
(38 -61)
None Plasmacortisol
Assessmenttime:between9.30am
andnoon)att hefourdefined
time-points
Nosignificantdifferencesinplasma
cortisol between the t wo groups
Fair
Andersenetal
(2013)97
Cross-sectional Undergraduate students who
experiencedrepeatedpeerdeat hs
Friends
classmate
N =122(61.48) 20.13(1.14) None
Other variables: relationship to
thedeceasedmediaexposure
to deaths mental health history
prioradverseexperiences
distress responses to peer
deaths
Haircortisolassessmenttimes:
once 3 months after the loss
experience
Priorbereavementexperiences(eg,
deathoffriendorfamilymember)are
significantly associated with hair cortisol
level
sig. negative relationship between number
ofbereavementexperiencesandcor tisol
levels during the period of peer deaths
single most important predic tor of cor tisol
response is whether or not a student
hadpreviouslyexperiencedthelossofa
friend or family member
Cohen
etal(2015)79
Cross-sectional(partofalarger
biomarkersstudy):inaverage1.04years
post-loss
Bereaved(1)vsnon-bereaved(2) NR N =529
n(1)= 260
(50)
(1)=54.27
(11.72 )
(2)=53.23
(11.0 5)
One/twoquestions:Had
someone close died since
Project1?-ifyes,numberof
persons close to the participant
who had died since the last
interview
Urinary cortisol
Assessmenttimes:afterinter view
during2-dayvisit
Nosignificantdifferencesinurinary
cortisol between the t wo groups
asig. association between number of
bereavements and levels of cortisol
Good
Morning/evening cortisol
Buckley
etal(2009)80
Prospectivecontrolledcohor tstudy,
longitudinal:
twotime-points:2weeksand6months
post-loss
Bereaved(1)vsnon-bereaved(2) Spouse(94)
Child(6)
N = 112
n(1)= 62
n(2)=50
(66)
(1)=65.2
(33-84)
(2)= 61. 6
(36-87)
None Plasmacortisol
Assessmenttime:morning
sig. higher morning cortisol levels in group
(1)comparedtogroup(2)attime-point
1 and 2
Notsignificant:cortisollevelsand
depression
Higheralcoholint akeisassociatedwith
higher cortisol levels
Good
Minton
etal(2009)96
Exploratorylongitudinalcorrelational
study11,12and13monthspost-loss
Widows Partner(100) N =47(100) 74.1(6.3) None Morningandeveningsalivary
cortisol (each averaged over
3days)
Assessmenttime:45minute safter
awakening and 12 hour s later
three consecutive days
Nosignificantdifferencesincortisollevels
betweenmonths11,12and13
Richard-son
etal(2015)78
Prospectivemulti-wavestudy,
longitudinal
Threetime-points(only1and2used
inbiomarkeranalysis):6months(1),
18months(2)and48mont hs(3)after
the death
Bereavement vs no bereavement Spouse Widowers:
n(1)=64
n(2)= 61
controls:
n =1545(only
subsampleused)
(NR)
Bereavement
group:
70(6.25)
no
bereavement
group:NR
None Overnight basal urinar y free
cortisol
Assessmenttime:
morning
Cortisollevelsincreasedfrom(1)to(2)in
widowed men and decreased in widowed
women
Prolongedforewarningassig.predictorof
cortisol levels
Bereaved men who reported emotional
numbnessat(1)hadhighercortisollevels
at(2)comparedtobereavedwomen
Good
|
7 of 24
HOPF et al .
TABLE 1 Result s of the qualitative systematic review grouped by outcome
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Mean cortisol level
Jacobs
etal(1987)84
Longitudinal:t wotime-points:first
interview 1 month after hospitalisation
ofpartner(1),secondinterview
2monthsaf terhospitalisation(2)
2 × 2 groups
Bereaved vs anticipatory bereaved
Risingseparationanxietyvsdeclining
separationanxietyfrom(1)to(2)
Spouse N =56
n(bereaved)=40
n (anticipatory
bereaved)= 16
(50)
62.6(NR) SA 24-hoururinaryfreecortisol
Assessmenttimes:threeseparate
daysintheweekbeforetime-
point(2)
Group,inwhichseparationanxietyrose
from(1)to(2)hadsig.highercortisol
levelsthangroupinwhichSAdiminished
or dropped
sig.Highercortisollevelswerefoundboth
for the bereaved and the anticipatory
bereaved subjects
Fair
Irwinetal(1988)92 Longitudinal:
weekly assessment of cortisol in a
1-2monthperiod
<6monthspost-loss
(1)Recentlybereavedwomenvs
(2)womenwithterminallyill
husbands vs
(3)womenwit hhealthyhusbands
Spouse N = 28
n(1)= 9
n(2)= 11
n (39 = 8
(100 )
52.2(3.4) None Plasmacortisol sig.Highermeanplasmacortisollevelsin
group(1)comparedtogroup(3)
Notsignificant:Plasmacortisollevelin
group(2)and(3)
Fair
Spratt&Denney
(1991) 93
Longitudinal:
fourtime-points:2,4,6and8months
post-loss(suddenloss)
Suddenlybereaved(1)vsnon-
bereaved(matchedtobereaved)
Child N = 18
n(1)= 9
(66)
(1)=49
(38 -61)
None Plasmacortisol
Assessmenttime:between9.30am
andnoon)att hefourdefined
time-points
Nosignificantdifferencesinplasma
cortisol between the t wo groups
Fair
Andersenetal
(2013)97
Cross-sectional Undergraduate students who
experiencedrepeatedpeerdeat hs
Friends
classmate
N =122(61.48) 20.13(1.14) None
Other variables: relationship to
thedeceasedmediaexposure
to deaths mental health history
prioradverseexperiences
distress responses to peer
deaths
Haircortisolassessmenttimes:
once 3 months after the loss
experience
Priorbereavementexperiences(eg,
deathoffriendorfamilymember)are
significantly associated with hair cortisol
level
sig. negative relationship between number
ofbereavementexperiencesandcor tisol
levels during the period of peer deaths
single most important predic tor of cor tisol
response is whether or not a student
hadpreviouslyexperiencedthelossofa
friend or family member
Cohen
etal(2015)79
Cross-sectional(partofalarger
biomarkersstudy):inaverage1.04years
post-loss
Bereaved(1)vsnon-bereaved(2) NR N =529
n(1)= 260
(50)
(1)=54.27
(11.72 )
(2)=53.23
(11.0 5)
One/twoquestions:Had
someone close died since
Project1?-ifyes,numberof
persons close to the participant
who had died since the last
interview
Urinary cortisol
Assessmenttimes:afterinter view
during2-dayvisit
Nosignificantdifferencesinurinary
cortisol between the t wo groups
asig. association between number of
bereavements and levels of cortisol
Good
Morning/evening cortisol
Buckley
etal(2009)80
Prospectivecontrolledcohor tstudy,
longitudinal:
twotime-points:2weeksand6months
post-loss
Bereaved(1)vsnon-bereaved(2) Spouse(94)
Child(6)
N = 112
n(1)= 62
n(2)=50
(66)
(1)=65.2
(33-84)
(2)= 61. 6
(36-87)
None Plasmacortisol
Assessmenttime:morning
sig. higher morning cortisol levels in group
(1)comparedtogroup(2)attime-point
1 and 2
Notsignificant:cortisollevelsand
depression
Higheralcoholint akeisassociatedwith
higher cortisol levels
Good
Minton
etal(2009)96
Exploratorylongitudinalcorrelational
study11,12and13monthspost-loss
Widows Partner(100) N =47(100) 74.1(6.3) None Morningandeveningsalivary
cortisol (each averaged over
3days)
Assessmenttime:45minute safter
awakening and 12 hour s later
three consecutive days
Nosignificantdifferencesincortisollevels
betweenmonths11,12and13
Richard-son
etal(2015)78
Prospectivemulti-wavestudy,
longitudinal
Threetime-points(only1and2used
inbiomarkeranalysis):6months(1),
18months(2)and48mont hs(3)after
the death
Bereavement vs no bereavement Spouse Widowers:
n(1)=64
n(2)= 61
controls:
n =1545(only
subsampleused)
(NR)
Bereavement
group:
70(6.25)
no
bereavement
group:NR
None Overnight basal urinar y free
cortisol
Assessmenttime:
morning
Cortisollevelsincreasedfrom(1)to(2)in
widowed men and decreased in widowed
women
Prolongedforewarningassig.predictorof
cortisol levels
Bereaved men who reported emotional
numbnessat(1)hadhighercortisollevels
at(2)comparedtobereavedwomen
Good
(Continues)
8 of 24
|
HOPF et al.
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Cortisol diurnal pattern
Ongetal(2012)90 Cross-sectional:inaver age17.5months
post-loss
Bereaved(1)vsnon-bereaved(2) Spouse N =44
n(1)= 22
(86)
65.8(48-8 0) None Salivary cortisol
Assessmenttimes:threeto
6 months af ter questionnaire
assessment on four successive
days
awakening ,30minutesafter
awakening ,beforelunch,at
bed-time
Sig. lower average wakeup levels of
salivarycor tisolingroup(1)comparedto
group(2)
sig. flat ter diurnal cortisol slope cur ve
amonggroup(1)comparedtogroup(2)
Notsignificant:effectofspousallosson
CARresponse
Pre-topost-losschangesinpositive
emotion accounted for 29% of the effec t
of spousal loss on diurnal cortisol slopes
*mediatingeffectofpositiveemotion,
even if controlling for confounding
factors
Fair
O'Connor
etal(2012)81
Cross-sectional:uptofiveyearspost-loss CGvsNG Mother(NR)
Sister(NR)
N =24
n(CG)= 12
(100 )
CG=42.67
(10.54)
NG=46.91
(9. 32 )
InterviewforComplicatedGrie Salivary cortisol
Assessmenttimes(Diur nal
pattern):
waking,45minutespostwaking,
4.00pm and 9.00 pm
sig.slopedifferencesbetweenCGandNG
groups:diurnalslopeoftheCGgroup
was lower in the morning and higher in
theevening--> flatter slope
Sig.lowercortisollevel45minutespost-
wakeinCGcomparedtoNG
Sig.highercortisollevelsat4.00pminCG
comparedtoNG
Good
Holland
etal(2014)82
Cross-sectional:inaverage3.1years
post-loss
(1)Depressednonbereavedvs
(2)depressedbereavedwithout
elevatedP GDvs
(3)depressedbereavedwithelevated
PGDsymptoms
Spouse/
partner(33)
Parent(16.7)
Sibling(12.5)
child(4.2)
N =56
n(1)= 32
n(2)=15
n(3)= 9
(60.7 )
69.9(7.6) ProlongedGriefDisorderScale
(P G -1 3)
Salivary cortisol
Assessmenttimes:
awakening-5.00pm-9.00p m
Twoconsecutivedays(combined
foranalysis)
log-transformedvaluesas
independent variable
Sig.lowerlevelsoflog-cortisollevelsat
wake and flatter diurnal slopes in group
(3)comparedtogroup(1)
Notsignificant:differencesb etween
group(2)and(3),althoughdescriptively
flatterprofileingroup(3)comparedto
group(2)
Bereavement independently of its
streng th is associated with dysregulated
cortisol levels
Subsidiary analysis:
Thosewhomostrecentlylosta
spouse showed sig. greater cortisol
dysregulation(higherlog-levelsatwake
andflat terslope)thanthosewholost
someone else than the partner
Notsignificant:
continuousPGdidnotpredictlog-cortisol
Fair
Perézetal(2017)77 Population-basedcohortstudy
(1)twoyearspost-loss
(2)betweentwoandfiveyearspost-loss
CGvsNGvsnogrief Partner(NR )
Child(NR)
Parent(NR)
Brother/
sister(NR)
Others(NR)
N =2084
n(NG)= 131
n(CG)= 31
n(nogrief)= 1922
(55)
64.9(5.5) InventoryofComplicatedGrief
(ICG),Dutchversion
Salivary cortisol
Assessmenttimes:
awakening-3 0minutesafter
awakening-5.00pm
-bedtime
(1)
Sig.lowerlevelsofmorningcor tisolinCG
vsNG
Sig. lower overall diurnal cor tisol levels
(AUCg)inCGvsNG
Sig.lowerlevelsofmorningcor tisolinCG
vs control
Notsignificant:slopedifferencebet ween
CGandNG
Notsignificant:cortisoldif ferences
betweenNGandcontrols
(2)
aSig.AUCgandcortisolmorningresponse
differencesbetweenCG(2-5years)and
CG(<2years)
aSig.higherscoresinICGareassociated
withlower morning cortisol
Good
TABLE 1 (Continued)
|
9 of 24
HOPF et al .
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Cortisol diurnal pattern
Ongetal(2012)90 Cross-sectional:inaver age17.5months
post-loss
Bereaved(1)vsnon-bereaved(2) Spouse N =44
n(1)= 22
(86)
65.8(48-8 0) None Salivary cortisol
Assessmenttimes:threeto
6 months af ter questionnaire
assessment on four successive
days
awakening ,30minutesafter
awakening ,beforelunch,at
bed-time
Sig. lower average wakeup levels of
salivarycor tisolingroup(1)comparedto
group(2)
sig. flat ter diurnal cortisol slope cur ve
amonggroup(1)comparedtogroup(2)
Notsignificant:effectofspousallosson
CARresponse
Pre-topost-losschangesinpositive
emotion accounted for 29% of the effec t
of spousal loss on diurnal cortisol slopes
*mediatingeffectofpositiveemotion,
even if controlling for confounding
factors
Fair
O'Connor
etal(2012)81
Cross-sectional:uptofiveyearspost-loss CGvsNG Mother(NR)
Sister(NR)
N =24
n(CG)= 12
(100 )
CG=42.67
(10.54)
NG=46.91
(9. 32 )
InterviewforComplicatedGrie Salivary cortisol
Assessmenttimes(Diur nal
pattern):
waking,45minutespostwaking,
4.00pm and 9.00 pm
sig.slopedifferencesbetweenCGandNG
groups:diurnalslopeoftheCGgroup
was lower in the morning and higher in
theevening--> flatter slope
Sig.lowercortisollevel45minutespost-
wakeinCGcomparedtoNG
Sig.highercortisollevelsat4.00pminCG
comparedtoNG
Good
Holland
etal(2014)82
Cross-sectional:inaverage3.1years
post-loss
(1)Depressednonbereavedvs
(2)depressedbereavedwithout
elevatedP GDvs
(3)depressedbereavedwithelevated
PGDsymptoms
Spouse/
partner(33)
Parent(16.7)
Sibling(12.5)
child(4.2)
N =56
n(1)= 32
n(2)=15
n(3)= 9
(60.7 )
69.9(7.6) ProlongedGriefDisorderScale
(P G -1 3)
Salivary cortisol
Assessmenttimes:
awakening-5.00pm-9.00p m
Twoconsecutivedays(combined
foranalysis)
log-transformedvaluesas
independent variable
Sig.lowerlevelsoflog-cortisollevelsat
wake and flatter diurnal slopes in group
(3)comparedtogroup(1)
Notsignificant:differencesb etween
group(2)and(3),althoughdescriptively
flatterprofileingroup(3)comparedto
group(2)
Bereavement independently of its
streng th is associated with dysregulated
cortisol levels
Subsidiary analysis:
Thosewhomostrecentlylosta
spouse showed sig. greater cortisol
dysregulation(higherlog-levelsatwake
andflat terslope)thanthosewholost
someone else than the partner
Notsignificant:
continuousPGdidnotpredictlog-cortisol
Fair
Perézetal(2017)77 Population-basedcohortstudy
(1)twoyearspost-loss
(2)betweentwoandfiveyearspost-loss
CGvsNGvsnogrief Partner(NR )
Child(NR)
Parent(NR)
Brother/
sister(NR)
Others(NR)
N =2084
n(NG)= 131
n(CG)= 31
n(nogrief)= 1922
(55)
64.9(5.5) InventoryofComplicatedGrief
(ICG),Dutchversion
Salivary cortisol
Assessmenttimes:
awakening-3 0minutesafter
awakening-5.00pm
-bedtime
(1)
Sig.lowerlevelsofmorningcor tisolinCG
vsNG
Sig. lower overall diurnal cor tisol levels
(AUCg)inCGvsNG
Sig.lowerlevelsofmorningcor tisolinCG
vs control
Notsignificant:slopedifferencebet ween
CGandNG
Notsignificant:cortisoldif ferences
betweenNGandcontrols
(2)
aSig.AUCgandcortisolmorningresponse
differencesbetweenCG(2-5years)and
CG(<2years)
aSig.higherscoresinICGareassociated
withlower morning cortisol
Good
(Continues)
10 of 24
|
HOPF et al.
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Cortisol:DHEAS ratio
Khanfer
etal(2011)89
Cross-sectional:
within2monthspos t-loss
Bereaved(1)vsnon-bereaved(2) Close family
member
(NR)
Friend(NR)
N =48
n(bereaved)=24
n(non-bereaved)=24
(67)
73(5.3) None Blood cortisol
Cortisol:DHEASratio
aSig.highercortisol:DHEASratioingroup
(1)comparedtogroup(2)
Notsignificant:Differencesincortisol
levelbet weengroup(1)and(2),although
highermeanvaluesingroup(1)
Fair
Vitlicetal(2014)85 Cross-sectional2x2design Youngbereaved(1)vsyoungnon-
bereaved(2)vsoldbereaved(3)vs
oldnon-bereaved(4)
Spouse(65
for(1)and
9.5for(2))
Close
relative(35
for(1)and
91.5for(2))
N = 93
n(1)= 31.8
n(2)= 20
n(3)= 26
n(4)= 26
(58)
(1)= 31. 8
(9. 03)
(2)= 31.7
(8.41)
(3)= 71.3
(5.79)
(4)= 72.6
(5.72)
CoreBereavementItems(CBI)
IES
Venousbloodsamples,
Cortisol,
DHEAS,cortisol:DHEAS-ratio
Sig.lowerDHEAS,highercortisoland
highercortisol:DHEASratioin(3)
comparedto(4)
Nosignificantdifferencesinthese
outcomesbetween(1)and(2)
ThosewithhigherCBI-scoresshowed
highercortisol:DHEASratios
Thosewithhighersocialsupportrepor ted
lowercortsiol:DHEA Sratios
Fair
DST/CRH stimulation test
Royetal(1988)86 Cross-sectional:CRHstimulationtest Bereavement complicated with
depression(1)vsuncomplicated
bereavement(2)vsdepressed
controls(3)vshealthycontrols(4)
Sample(3)and(4)areusedfrom
earlier study
Spouse(25)
1st degree
relative(75)
N = 92
n(1)= 9
n(2)= 19
n(3)= 30
n(4)=34
(41)
(1)=47.6(14)
(2)=41.5
(13.7 )
(3)=42.3
(13.1)
(4)=29.4(5.1)
DSM-IIIassessmentof
complic atedvsnon-complicated
bereavement (with vs without
depression)
TexasInventor yofGrief(128)
PlasmaACTHandcor tisolafter
CRHadministration(1-μg/kg)
Assessmenttimes:
30minutes,50minutes,
60minutes,75minutes,105
min135minand165minutesafter
injection of needle
Sig.higherbasalcortisollevelsingroup(1)
comparedtogroup(2)and(4)
NosignificantdifferencesinACTH-levels
Sig.smallerACTHresponsestoCRHin
group(1)comparedtogroup(2)and(4)
Sig.greatercortisolresponsestoCRHin
group(1)comparedtogroups(2)-(4)
NosignificantdifferencesinACTH
responsestoCRHbetweengroups(1)
and(2)
Fair
Petitto
etal(1992)102
Cross-sectional:assessmentofadults
with affective disorder who had
experiencedlossearlierinlifeDST
Patientswithaffectivedisorder
Early loss (<=19years)vslateloss(>
20years)
Onlyfirstlossexamined
Mother(20)
Fathe r(60)
Sibling(20)
N =45
n(earlyloss)= 22
n(lateloss)= 23
(58)
44.7(14.1) None Blood cortisol
Assessmenttimes:4.0 0pm
and 11.00 p m 1 day after
dexamethasoneapplication
11.0 0 pmdaybefore)
Amongtheaffectivedisorderpatientsof
theearlylossgroup,youngerageatfirst
loss significantlya correlated with higher
4.00pm cortisol levels
FirstlossasstrongestpredictorforHPA
axisfunctioning
Latelosspredic tshighercortisollevelsat
11.0 0 pm
Fair
Gerra
etal(2003)88
Longitudinal:
3time-points:
10days(1),
40days(2)and
6 months
afterstress-fulllifeevent
DSTadministered
Bereaved vs controls Parent(57)
Son(14)
Spouse(29)
N = 28
n(bereaved)=14
n(control)= 14
38(17-75) None
Degree of s tress (Social
AdjustmentScale)
Blood cortisol
bloodACTH
DST
Assessmentofbloodsamples:
between 9.00 and 11.0 0 pm at
times(1),(2)and(3)
Sig.highercortisolplasmalevelsafterDST
intime(1)comparedtotime(2)and(3)
a Sig. higher cortisol plasma levels after
DSTintime(1)inbereavedgroup
compared to control
Sig.highermeanbasalACTH
concentrations in bereaved subjec ts in
time(1)comparedto(2)and(3)
Sig.highermeanbasalACTH
concentrations in bereaved subjec ts
comparedtocontrolsintime(1)
Sig. higher plasma cortisol concentrations
inresponsetodexamethasoneinhigh
responders compared to low responders
in the bereavement group
Sig.correlationsbetweenHRSDand
cortisollevelsattime(1)
Fair
TABLE 1 (Continued)
(Continues)
|
11 of 24
HOPF et al .
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Cortisol:DHEAS ratio
Khanfer
etal(2011)89
Cross-sectional:
within2monthspos t-loss
Bereaved(1)vsnon-bereaved(2) Close family
member
(NR)
Friend(NR)
N =48
n(bereaved)=24
n(non-bereaved)=24
(67)
73(5.3) None Blood cortisol
Cortisol:DHEASratio
aSig.highercortisol:DHEASratioingroup
(1)comparedtogroup(2)
Notsignificant:Differencesincortisol
levelbet weengroup(1)and(2),although
highermeanvaluesingroup(1)
Fair
Vitlicetal(2014)85 Cross-sectional2x2design Youngbereaved(1)vsyoungnon-
bereaved(2)vsoldbereaved(3)vs
oldnon-bereaved(4)
Spouse(65
for(1)and
9.5for(2))
Close
relative(35
for(1)and
91.5for(2))
N = 93
n(1)= 31.8
n(2)= 20
n(3)= 26
n(4)= 26
(58)
(1)= 31. 8
(9. 03)
(2)= 31.7
(8.41)
(3)= 71.3
(5.79)
(4)= 72.6
(5.72)
CoreBereavementItems(CBI)
IES
Venousbloodsamples,
Cortisol,
DHEAS,cortisol:DHEAS-ratio
Sig.lowerDHEAS,highercortisoland
highercortisol:DHEASratioin(3)
comparedto(4)
Nosignificantdifferencesinthese
outcomesbetween(1)and(2)
ThosewithhigherCBI-scoresshowed
highercortisol:DHEASratios
Thosewithhighersocialsupportrepor ted
lowercortsiol:DHEA Sratios
Fair
DST/CRH stimulation test
Royetal(1988)86 Cross-sectional:CRHstimulationtest Bereavement complicated with
depression(1)vsuncomplicated
bereavement(2)vsdepressed
controls(3)vshealthycontrols(4)
Sample(3)and(4)areusedfrom
earlier study
Spouse(25)
1st degree
relative(75)
N = 92
n(1)= 9
n(2)= 19
n(3)= 30
n(4)=34
(41)
(1)=47.6(14)
(2)=41.5
(13.7 )
(3)=42.3
(13.1)
(4)=29.4(5.1)
DSM-IIIassessmentof
complic atedvsnon-complicated
bereavement (with vs without
depression)
TexasInventor yofGrief(128)
PlasmaACTHandcor tisolafter
CRHadministration(1-μg/kg)
Assessmenttimes:
30minutes,50minutes,
60minutes,75minutes,105
min135minand165minutesafter
injection of needle
Sig.higherbasalcortisollevelsingroup(1)
comparedtogroup(2)and(4)
NosignificantdifferencesinACTH-levels
Sig.smallerACTHresponsestoCRHin
group(1)comparedtogroup(2)and(4)
Sig.greatercortisolresponsestoCRHin
group(1)comparedtogroups(2)-(4)
NosignificantdifferencesinACTH
responsestoCRHbetweengroups(1)
and(2)
Fair
Petitto
etal(1992)102
Cross-sectional:assessmentofadults
with affective disorder who had
experiencedlossearlierinlifeDST
Patientswithaffectivedisorder
Early loss (<=19years)vslateloss(>
20years)
Onlyfirstlossexamined
Mother(20)
Fathe r(60)
Sibling(20)
N =45
n(earlyloss)= 22
n(lateloss)= 23
(58)
44.7(14.1) None Blood cortisol
Assessmenttimes:4.0 0pm
and 11.00 p m 1 day after
dexamethasoneapplication
11.0 0 pmdaybefore)
Amongtheaffectivedisorderpatientsof
theearlylossgroup,youngerageatfirst
loss significantlya correlated with higher
4.00pm cortisol levels
FirstlossasstrongestpredictorforHPA
axisfunctioning
Latelosspredic tshighercortisollevelsat
11.0 0 pm
Fair
Gerra
etal(2003)88
Longitudinal:
3time-points:
10days(1),
40days(2)and
6 months
afterstress-fulllifeevent
DSTadministered
Bereaved vs controls Parent(57)
Son(14)
Spouse(29)
N = 28
n(bereaved)=14
n(control)= 14
38(17-75) None
Degree of s tress (Social
AdjustmentScale)
Blood cortisol
bloodACTH
DST
Assessmentofbloodsamples:
between 9.00 and 11.0 0 pm at
times(1),(2)and(3)
Sig.highercortisolplasmalevelsafterDST
intime(1)comparedtotime(2)and(3)
a Sig. higher cortisol plasma levels after
DSTintime(1)inbereavedgroup
compared to control
Sig.highermeanbasalACTH
concentrations in bereaved subjec ts in
time(1)comparedto(2)and(3)
Sig.highermeanbasalACTH
concentrations in bereaved subjec ts
comparedtocontrolsintime(1)
Sig. higher plasma cortisol concentrations
inresponsetodexamethasoneinhigh
responders compared to low responders
in the bereavement group
Sig.correlationsbetweenHRSDand
cortisollevelsattime(1)
Fair
(Continues)
12 of 24
|
HOPF et al.
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Pfeffer
etal(2009)91
Longitudinal:
twotime-points:oneafterstudyentr y
and one within 6 months after entry
(1)Bereaved(asare sultofa
traumaticevent-terrorattackat
09/11/2001 )
vs
(2)non-bereaved
Spouse N =45
n(1)= 23
(96)
(1)=41.79
(6.52)
(2)=41.12
(6.46)
None Basalandpost-dexamethasone
cortisol
Assessmenttimes:30minutes
afterawakening,7.00pm,4.00pm
,9.00pm
on four consecutive days
Dexamethasoneadministration:on
day 3 in the evening
Sig. higher am-cortisolingroup(1)
comparedtogroup(2)
pm-cortisoltendedtobehigheringroup
(1)comparedtogroup(2)
Sig. less afternoon cor tisol suppression in
group(1)comparedtogroup(2)
Notsignificant:groupdifferences
in cortisol suppression during a m
assessment
Sig. higher pm cortisol suppression in
bereavedwithaccompaniedPTSD
compared to bereaved without any
psychiatric disorder
Fair
Catecholamines
Jacobs
etal(1986)83
Cross-sectional:
2 months af ter hospitalisation/death of
the partner
Bereaved(1)vsanticipatory
bereaved(2)
Spouse N =59
n(1)= 39
n(2)= 20
(51)
61.9(NR) Emotional Distress associated
with loss
24-hour
urinary catecholamines
(epinephrineandnorepinephrine)
Assessmenttimes:threesuccessive
day
Higheroutputsofcatechloaminesin
(1)comparedto(2)-notinarange
associated with adrenal medullar y
disease
Nosignificantdifferencein
norepinephrineorepinephrinein(1)
comparedto(2)
Sig. negative correlation between
norepinephrine and depression score
Fair
Jacobs
etal(1997)87
Longitudinal:
sixtime-pointsafterhospitalisationover
thecourseof25-monthsfollow-up:
1sttime-point(1):directlyafterint ake
2ndtime-point(2):1monthafterintake
3rdtime-point(3):2monthsafterintake
4thtime-point(4):between2and
13 months af ter intake
5thtime-point(5):13monthsafterintake
6thtime-point(6):25monthsafterintake
2ndtime-point:baselines ymptom
assessment
3rdtime-point:defensiveand
neuroendocrine assessment
5thand6thtime-point:outcome
assessment
Bereaved/anticipatory bereaved Spouse N = 67
(50)
62(0.9) Unresolved grief/separation
distress(asoutcomevariable)
Aspredictors:
mean24-hoururinaryfreecortisol
attime-point(3)
Mean24-hoururinaryfree
epinephrineattime-point(3)
three samples
Nosignificantdifferencebetween
bereavedandnon-bereavedin
neuroendocrine functioning
Nosig.correlationsbetween
neuroendocrine measures and separation
distress,depression,anxietyand
demoralisation
Highmeancortisolpredictedbetterself-
ratedhealthattime(5)
Highmeanepinephrinepredictedhigher
hopelessnes/helplessness scores at time
(5)
Meancor tisolwasinverselycorrelated
with symptoms of hopelessness and
helplessnessattime(6)
Meanepinephrinewaspositively
correlated with symptoms of
hopelessness/helplessnessattime(6)
Fair
Insulin
Cankaya
etal(2009)98
Cross-sectional(partofalarger
investigationofstress,individual
differences,andhealthinamiddle-aged
andolderprimar ycaresample)
Suddenunexpectedloss(linearor
ordinal)
Naturalvsunnaturaldeath
NR N =75(100) 52.07(9.67) None
TraumaticLifeEventsSc ale
1)Lifetimehistor yofanysudden
unexpectedloss
2)Numberoflifetimesudden
losses
3)Typeofsuddenloss
IG F -1
assessed in blood
assessment times: between late
morning and late af ternoon after
the interview
Sig.lowerIGF-1levelsinwomenwhohad
experiencedasuddenunexpectedloss
compared to women without a history of
sudden loss
Numberofsuddenlossesissignific antly
associatedwithIFG-1levels:thegreatest
decreaseinIGF-1wasshowninthe
group with the most losses (>5sudden
losses)
Nosignificantdifferencesbet weenthose
who lost someone as a result of an
unnatural event vs a natural event
TABLE 1 (Continued)
(Continues)
|
13 of 24
HOPF et al .
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Pfeffer
etal(2009)91
Longitudinal:
twotime-points:oneafterstudyentr y
and one within 6 months after entry
(1)Bereaved(asare sultofa
traumaticevent-terrorattackat
09/11/2001 )
vs
(2)non-bereaved
Spouse N =45
n(1)= 23
(96)
(1)=41.79
(6.52)
(2)=41.12
(6.46)
None Basalandpost-dexamethasone
cortisol
Assessmenttimes:30minutes
afterawakening,7.00pm,4.00pm
,9.00pm
on four consecutive days
Dexamethasoneadministration:on
day 3 in the evening
Sig. higher am-cortisolingroup(1)
comparedtogroup(2)
pm-cortisoltendedtobehigheringroup
(1)comparedtogroup(2)
Sig. less afternoon cor tisol suppression in
group(1)comparedtogroup(2)
Notsignificant:groupdifferences
in cortisol suppression during a m
assessment
Sig. higher pm cortisol suppression in
bereavedwithaccompaniedPTSD
compared to bereaved without any
psychiatric disorder
Fair
Catecholamines
Jacobs
etal(1986)83
Cross-sectional:
2 months af ter hospitalisation/death of
the partner
Bereaved(1)vsanticipatory
bereaved(2)
Spouse N =59
n(1)= 39
n(2)= 20
(51)
61.9(NR) Emotional Distress associated
with loss
24-hour
urinary catecholamines
(epinephrineandnorepinephrine)
Assessmenttimes:threesuccessive
day
Higheroutputsofcatechloaminesin
(1)comparedto(2)-notinarange
associated with adrenal medullar y
disease
Nosignificantdifferencein
norepinephrineorepinephrinein(1)
comparedto(2)
Sig. negative correlation between
norepinephrine and depression score
Fair
Jacobs
etal(1997)87
Longitudinal:
sixtime-pointsafterhospitalisationover
thecourseof25-monthsfollow-up:
1sttime-point(1):directlyafterint ake
2ndtime-p oint(2):1monthafterintake
3rdtime-point(3):2monthsafterintake
4thtime-point(4):between2and
13 months af ter intake
5thtime-point(5):13monthsafterintake
6thtime-point(6):25monthsafterintake
2ndtime-p oint:baselines ymptom
assessment
3rdtime-point:defensiveand
neuroendocrine assessment
5thand6thtime-point:outcome
assessment
Bereaved/anticipatory bereaved Spouse N = 67
(50)
62(0.9) Unresolved grief/separation
distress(asoutcomevariable)
Aspredictors:
mean24-hoururinaryfreecortisol
attime-point(3)
Mean24-hoururinaryfree
epinephrineattime-point(3)
three samples
Nosignificantdifferencebetween
bereavedandnon-bereavedin
neuroendocrine functioning
Nosig.correlationsbetween
neuroendocrine measures and separation
distress,depression,anxietyand
demoralisation
Highmeancortisolpredictedbetterself-
ratedhealthattime(5)
Highmeanepinephrinepredictedhigher
hopelessnes/helplessness scores at time
(5)
Meancor tisolwasinverselycorrelated
with symptoms of hopelessness and
helplessnessattime(6)
Meanepinephrinewaspositively
correlated with symptoms of
hopelessness/helplessnessattime(6)
Fair
Insulin
Cankaya
etal(2009)98
Cross-sectional(partofalarger
investigationofstress,individual
differences,andhealthinamiddle-aged
andolderprimar ycaresample)
Suddenunexpectedloss(linearor
ordinal)
Naturalvsunnaturaldeath
NR N =75(100) 52.07(9.67) None
TraumaticLifeEventsSc ale
1)Lifetimehistor yofanysudden
unexpectedloss
2)Numberoflifetimesudden
losses
3)Typeofsuddenloss
IG F -1
assessed in blood
assessment times: between late
morning and late af ternoon after
the interview
Sig.lowerIGF-1levelsinwomenwhohad
experiencedasuddenunexpectedloss
compared to women without a history of
sudden loss
Numberofsuddenlossesissignific antly
associatedwithIFG-1levels:thegreatest
decreaseinIGF-1wasshowninthe
group with the most losses (>5sudden
losses)
Nosignificantdifferencesbet weenthose
who lost someone as a result of an
unnatural event vs a natural event
(Continues)
14 of 24
|
HOPF et al.
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Oxyto cin
Buietal(2019)99 Cross-sectionalpilotstudy
loss occurred at least 6 months prior to
the study
Bereaved with primary diagnosis of
CG(1)vs
MajorDepressiveDisorder(MDD)
(2)vs
bereavedcontrols(3)
Parent
(25.6in(1),
40in(2),
42.9in(3))
Spouse
(41in(1),12
in(2),8.6
in(3))
Other
(33.3in(1),
48in(2),
48.6in(3))
N = 139
n(1)=47(70.21)
n(2)=46(69.57)
n(3)=46(69.6)
(1)=49.49
(12. 87)
(2)=49.33
(13. 27)
(3)=48.6 4
(12.7)
Inventor yofComplicatedG rief
(ICG)
Structured Clinical Interview for
ComplicatedGr ief(SCI-CG)99
OverallplasmalevelsofOT,
measured through one simple
blood collection
Sig.higherplasmaOTlevelsforgroup(1)
comparedtogroup(2)
NosignificantOTdifferencesbet ween(1)
and(3)
ICGsymptomseverityexplainedonly2%
ofthevariationinplasmaOTlevels
Secondary analysis: a primary or probable
CGdiagnosisispositivelyassociatedwith
plasmaOTlevels
Prolactin
Laneetal
(1987)100
Cross-sectionalbereavedsample,
8 weeks af ter deat h of the spouse
Widows/widowerswithlow
(1),moderate(2),orhigh(3)
developmental level of object
representation(DLOR)
Spouse N =26(46) 58.9(26.4) None
DLOR(highvsmoderatevslow)
Serum prolactin
assessment times: before and after
semistructured interview
pre-topostinterviewprolactin
change
Sig.largermeanPRLchangeinwomen
compared to men
Sig.negativecorrelationbetweenPRL
changeandDLORinwomen
Sig.positivecorrelationbetweenPRL
changeandDLORinmen
Intervention studies
Theorell
etal(1987)101
Interventionstudy(activationprogram) Anticipator ybereavedandbereaved
men and women who are about to
lose/who lost a close relative
(1)Activationprogr ammeGroup
(2)ComparisonGroup
Close female
relatives
(wives,
sibling or
child-ren)
N =72(100)
n(1)= 36
n(2)= 36
(1)=51
(2 4 -7 7 )
(2)=52
(2 1-7 7 )
None
Others:
Depression
Anxiety
MentalExhaustion
Serum prolactin
serum cortisol
assessment times:
during treatment period
before death
1 month after death
2 months af ter death
Increasingdegreeofmentalexhaustion
during the treatment period is associated
with increasing cortisol levels and
decreasing prolactin levels
Sig. increased cortisol levels 1 month after
death compared to the last observation
before death
Sig. lower prolactin levels during
treatment in the activation programme
comparedtogroup(2)
no significant differences in cortisol or
prolac tin levels from 1 to 2 months after
death
Goodkin
etal(1998)94
Randomised,controlle dinter vention
study;
longitudinal:
Baseline(beforeintervention),10weeks
(rightafterinter vention),6months
follow-up
10 months bereavement suppor t group vs
standard care control
about6monthspost-loss
BereavedHIV+homosexualmen
(1)vsbereavedHIV-homosexual
men(2)
Close friend
(NR)
Partner(NR)
N = 119
n(1)=45
(0)
38.3(9.5) None Plasmacortisolatallthree
time-points
sig. decrease of plasma cortisol levels in
the intervention group compared to the
control group
Group(1)interventionsubjectsshowed
a decrease in cortisol levels from time 1
totime3,whereasgroup2intervention
subjects showed an increase in cor tisol
levels from time 1 to time 3
Sig. effect of intervention on cor tisol
levels(time1and3included)when
controlling for baseline cortisol levels
Fair
O'Connor
etal(2013)95
Partofalargerrandomisedclinic altrial.
longitudinal:
Pre-PostComplicatedGriefIntervention
meantimepost-loss= 87 months
ComplicatedGr ief(continuously
measured)
Close friend
(NR)
Spouse(NR)
Parent(NR)
Child(NR)
Sibling(NR)
N =16(88) 64(4.3) InventoryofComplicatedGrief
(ICG)
Blood catecholamines:
epinephrine,norepinephrine,
dopamine
Assessmenttimes:
upto4weeksbeforefirsttherapy
session
between 10.00 am and 3.30 pm
aSig.predictionofpost-treatmentICG
scorebypre-treatmentepinephrine
Notsignificant:pre-treatmentdopamine
andepinephrineinpredictingpost-
treatmentCGscore
Fair
Abbreviations:AUCg,areaunderthecurvewithrespecttothegroundandtheslope;NR ,notreported;QA,QualityA ssessment.
ACTH,adrenocorticotrophichormone;CAR ,cortisolawakeningresponse;CBI,CoreBereavementItems;CG,complicatedgrief;CRH,
corticotrophin-releasinghormone;DHEAS ,dehydroepiandrostheron-sulphate;DSM-III,DiagnosticandSt atisticalManualofMentalDisordersIII;
DLOR,developmentallevelsofthesurvivors’objectrepresentation;DST,dexamethasonesuppressiontest;HRSD,HamiltonRatingScalefor
Depression;ICG,Inventor yofComplicatedGrief;IES,ImpactEventScale;IGF,insulin-likegrowthfactor;MDD,majordepressivedisorder;NG,
normalgrief;OT,oxytocin;PG ,prolongedgrief;PGD,ProlongedGriefDisorder;PRL,prolac tin;PTSD,post-traumaticstressdisorder;SA,separationanxiety.
sig.,significant.
TABLE 1 (Continued)
|
15 of 24
HOPF et al .
Study Study design
Sample characteristics
Grief assessment
Dependent neuroendocrine
measure Results
QA
ratingGrief types/groups
Loss relation
(%)
N/n per group
(% female)
Age (mean/
range or SD)
Oxyto cin
Buietal(2019)99 Cross-sectionalpilotstudy
loss occurred at least 6 months prior to
the study
Bereaved with primary diagnosis of
CG(1)vs
MajorDepressiveDisorder(MDD)
(2)vs
bereavedcontrols(3)
Parent
(25.6in(1),
40in(2),
42.9in(3))
Spouse
(41in(1),12
in(2),8.6
in(3))
Other
(33.3in(1),
48in(2),
48.6in(3))
N = 139
n(1)=47(70.21)
n(2)=46(69.57)
n(3)=46(69.6)
(1)=49.49
(12. 87)
(2)=49.33
(13. 27)
(3)=48.6 4
(12.7)
Inventor yofComplicatedG rief
(ICG)
Structured Clinical Interview for
ComplicatedGr ief(SCI-CG)99
OverallplasmalevelsofOT,
measured through one simple
blood collection
Sig.higherplasmaOTlevelsforgroup(1)
comparedtogroup(2)
NosignificantOTdifferencesbet ween(1)
and(3)
ICGsymptomseverityexplainedonly2%
ofthevariationinplasmaOTlevels
Secondary analysis: a primary or probable
CGdiagnosisispositivelyassociatedwith
plasmaOTlevels
Prolactin
Laneetal
(1987)100
Cross-sectionalbereavedsample,
8 weeks af ter deat h of the spouse
Widows/widowerswithlow
(1),moderate(2),orhigh(3)
developmental level of object
representation(DLOR)
Spouse N =26(46) 58.9(26.4) None
DLOR(highvsmoderatevslow)
Serum prolactin
assessment times: before and after
semistructured interview
pre-topostinterviewprolactin
change
Sig.largermeanPRLchangeinwomen
compared to men
Sig.negativecorrelationbetweenPRL
changeandDLORinwomen
Sig.positivecorrelationbetweenPRL
changeandDLORinmen
Intervention studies
Theorell
etal(1987)101
Interventionstudy(activationprogram) Anticipator ybereavedandbereaved
men and women who are about to
lose/who lost a close relative
(1)Activationprogr ammeGroup
(2)ComparisonGroup
Close female
relatives
(wives,
sibling or
child-ren)
N =72(100)
n(1)= 36
n(2)= 36
(1)=51
(2 4 -7 7 )
(2)=52
(2 1-7 7 )
None
Others:
Depression
Anxiety
MentalExhaustion
Serum prolactin
serum cortisol
assessment times:
during treatment period
before death
1 month after death
2 months af ter death
Increasingdegreeofmentalexhaustion
during the treatment period is associated
with increasing cortisol levels and
decreasing prolactin levels
Sig. increased cortisol levels 1 month after
death compared to the last observation
before death
Sig. lower prolactin levels during
treatment in the activation programme
comparedtogroup(2)
no significant differences in cortisol or
prolac tin levels from 1 to 2 months after
death
Goodkin
etal(1998)94
Randomised,controlle dinter vention
study;
longitudinal:
Baseline(beforeintervention),10weeks
(rightafterinter vention),6months
follow-up
10 months bereavement suppor t group vs
standard care control
about6monthspost-loss
BereavedHIV+homosexualmen
(1)vsbereavedHIV-homosexual
men(2)
Close friend
(NR)
Partner(NR)
N = 119
n(1)=45
(0)
38.3(9.5) None Plasmacortisolatallthree
time-points
sig. decrease of plasma cortisol levels in
the intervention group compared to the
control group
Group(1)interventionsubjectsshowed
a decrease in cortisol levels from time 1
totime3,whereasgroup2intervention
subjects showed an increase in cor tisol
levels from time 1 to time 3
Sig. effect of intervention on cor tisol
levels(time1and3included)when
controlling for baseline cortisol levels
Fair
O'Connor
etal(2013)95
Partofalargerrandomisedclinic altrial.
longitudinal:
Pre-PostComplicatedGriefIntervention
meantimepost-loss= 87 months
ComplicatedGr ief(continuously
measured)
Close friend
(NR)
Spouse(NR)
Parent(NR)
Child(NR)
Sibling(NR)
N =16(88) 64(4.3) InventoryofComplicatedGrief
(ICG)
Blood catecholamines:
epinephrine,norepinephrine,
dopamine
Assessmenttimes:
upto4weeksbeforefirsttherapy
session
between 10.00 am and 3.30 pm
aSig.predictionofpost-treatmentICG
scorebypre-treatmentepinephrine
Notsignificant:pre-treatmentdopamine
andepinephrineinpredictingpost-
treatmentCGscore
Fair
Abbreviations:AUCg,areaunderthecurvewithrespecttothegroundandtheslope;NR ,notreported;QA,QualityA ssessment.
ACTH,adrenocorticotrophichormone;CAR ,cortisolawakeningresponse;CBI,CoreBereavementItems;CG,complicatedgrief;CRH,
corticotrophin-releasinghormone;DHEAS ,dehydroepiandrostheron-sulphate;DSM-III,DiagnosticandSt atisticalManualofMentalDisordersIII;
DLOR,developmentallevelsofthesurvivors’objectrepresentation;DST,dexamethasonesuppressiontest;HRSD,HamiltonRatingScalefor
Depression;ICG,Inventor yofComplicatedGrief;IES,ImpactEventScale;IGF,insulin-likegrowthfactor;MDD,majordepressivedisorder;NG,
normalgrief;OT,oxytocin;PG ,prolongedgrief;PGD,ProlongedGriefDisorder;PRL,prolac tin;PTSD,post-traumaticstressdisorder;SA,separationanxiety.
sig.,significant.
16 of 24
|
HOPF et al.
At time 2, sig nificant hi gher corti sol morning r esponses an d over-
allcortisolresponses were foundin the CGgroupcomparedtothe
same group at time 1.Furthermore,higher scores in grief severity
were associated with lower morning cortisol levels.77
3.4 | Cortisol:DHEAS ratio
Two studies investigated the association between bereavement
and the cortisol:dehydroepiandrostheron-sulphate (DHEAS) ratio.
DHEA S is a sulfated s teroid-hor mone that is as sociated wit h HPA
axis activity. By contrast to cortisol, which has immunosuppres-
siveeffects, DHEAS enhances theimmune response.Studies have
shown that D HEAS can b uffer the su ppressive ef fects of cor tisol
on neutrophil function.89Additionally,anincreasedcortisol:DHEAS
ratio, which represents an imbalance between those biomarkers,
appear s to be a contributi ng factor to the p rocess of age-relate d
immunosenescence.K hanferet al89 hypothesised that ageing and
stress had an additive and deleterious effec t on immunity and that
bereaved older adults should have higher cortisol:DHE AS ratios
thannon-bereavedolderadults.Theyusedthecortisol:DHEASratio
as an indicator of neutrophil function and assessed cortisol levels
inbereavedandnon-bereavedolderadults.Althoughcor tisollevels
wereslightlyhigherinthebereavedgroup,ahighercortisol:DHEAS
ratio was found in the bereaved compared to the non-bereaved
subjects.89Vitlicetal85 compared cortisol:DHEAS ratios between
youngerandolderbereavedvs non-bereavedadultsand foundsig-
nificant lower DHEAS, higher cortisol and higher cortisol:DHEAS
ratios in the older bereaved compared to the older non-bereaved.
Thesedifferenceswerenotshownintheyounggroups.85Although
the younger bereaved showed higher psychological effects of loss
than the o lder subjec ts, the se changes wer e not reflec ted in neu-
roendocrineoutcomes.Finally,those with strongergriefsymptoms
showed higher cortisol:DHEAS ratios, whereas those with higher
levels of social support showed lower ratios.85
3.5 | Dexamethasone suppression test (DST)/CRH
stimulation test
TheDSTisa pplie dtoass essHPAa xisfeedbac kse nsiti vity.103 By ap-
plyingthecorticosteroidedexamethasone,whichmimicstheeffects
ofcortisol,cortisolreleaseshouldbesuppressedinhealthyindividu-
als.Non-suppressionisconsideredanindicatorofhypercortisolism.
TheCRHstimulationtestwasdesignedtotestHPAaxis dysregula-
tionbystimulatingtheACTHresponse.103Af tertheadministration
ofCRH,arapidriseinACTHandcortisolisexpected,followedbya
gradual decrease.
FourstudiesinvestigatedDST/CRHresultsinbereavedindividu-
als. Roy et al86appliedtheCRHstimulationtestinbereavedmenand
women with or without depression and hypothesised that depressed
bereaved wo uld show similar r eactions to C RH stimulati on as de-
pressednon-bereaved. The non-depressed women appear to have
“normally”bluntedresponsestoCRHstimulation,whichmayreflect
their normal reac tion to the negative feedback of hypercortisolism
that is often found in depressive patient s.86ACTHandcor tisolwere
assessedin 92 participantsafter receivingtheDST.Highercortisol
andlowerACTHlevelswerefoundinthedepressedbereavedcom-
pared to th e non-depress ed bereaved and t he healthy contr ols.86
Petitto et al102examinedthe relationshipbetweenlossexperience
and HPAaxis function insubjects with anaffective disorder.They
comparedcor tisol levelsafterDSTin45 men and women who had
aloss experience atthe age of 17 years or earlier withthose who
had a lossexperience at the age of 18years orlater and included
major depressive disorder as a control variable. Depressed men
and women s howed lower cor tisol level s than the non- depresse d.
Among the affective disorder patients of the early loss group,
younger age at first loss significantly correlated with higher after-
nooncortisollevels.Furthermore,intheafternoon,menintheearly
loss group showed significantly higher cortisol levels than women.
Late losssignificantlypredic ted higher cortisol levels in themorn-
ing.102Gerraetal
88 compared ACTH, cortisol levels and immune
markersafterDSTin28bereavedvsnon-bereavedmenandwomen
10days,40daysand6monthsafterloss.Theyfoundhighercortisol
levels in the bereaved, compared to the non-bereaved. ACTH lev-
elswere significantly higher inthe bereavedgroup at time-point 1
only.Interestingly,cortisolandACTHlevelswerehighestintheearly
stageofbereavement.Furthermore,theeffectoftemperamentwas
investigated:theyfoundnon-suppressionofdexamethasoneinsub-
jects with high depression and harm avoidance compared to subject s
with low depression and harm avoidance 6 months after bereave-
ment.88Pfefferetal91examinedbasal and post-DST cortisol in 23
traumaticallybereaved participantsovertwotime-points following
the 9/11 terror attacks. Bereaved spouses showed higher morning
basalcortisol and lessafternoonpost-dexamethasonesuppression
than non-b ereaved subject s. Additionall y, bereaved sub jects with
PTSDshowed significantlygreater afternoon post-dexamethasone
suppressionthanbereavedsubjectswithoutPTSD,indicatinghigher
glucocor ticoidreceptorsensitivityinthebereavedwithPTSD.91
3.6 | Catecholamines
Two studies exam ined the associa tion between b ereavement and
catecholamines as outcomes of sympathetic adrenal medullary
function (SAM). Jacobs et al83 investigated 24-hour urinary free
epinephrine and norepinephrine on three successive days in 59
bereaved and anticipatory-bereaved subject s and found higher
catecholamine outputs in the bereaved compared to the anticipa-
tory bereaved; however, these differences were not significant.
Norepinephrine was inversely correlated with depression scores
andpositively correlated with age. Thelatter finding is in line with
past research showingthat the SAM system in olderadultsadapts
more slowly to stress.83Jacobsetal87 examined the predictiveef-
fec tofadrenalfunctio nondep ression,anx iety,hope lessness ,orun-
resolvedgrief.Theyassessed24-hoururinarycortisol,epinephrine
|
17 of 24
HOPF et al .
and norepinephrine in bereaved and anticipatory-bereaved indi-
viduals.Theneuroendocrinemarkerswereassessedthreetimesat
time-poi nt 3 (2 months af ter hospit alisation), at w hich 63% of the
subjectswerewidowed.Thepsychological variableswereassessed
at time-points 2, 3, 5 and 6 (1, 2, 13 and 25 months after intake.)
Theneuroendocrinemarkersdidnotdifferbetweenthetwogroups.
Epinephrineandcortisolonlypredictedhopelessnessattime-point
5inthebereaved subjects, althoughtheydidnot predictanyother
psychological outcomes. Additionally, higher mean cortisol levels
(average of the t hree assessm ents) at time-poi nt 3 predicted b et-
ter self-rated health at time-point5.Mean cortisolmeasureswere
inversely correlated and mean epinephrine levels were positively
correlatedwithhopelessnessscoresattime-point6.Theresultsin-
dicate that adrenal function may ser ve as a mediator between social
lossandhealth-relatedoutcomes.87
3.7 | Insulin
Cankaya et al98 investigated a ssociations of int erleukin (IL-)-6 and
insulin-likegrowth factor (IGF )-1withthesuddendeath of aloved
onein75femalesin anurbanprimarycareset ting. IGF-1isposited
as a protec tive factor in a geing-related dise ases and is negat ively
correlated with immune markers such as IL-6. It was hypothesised
thataprolongedexposuretostressandasuddendeathwouldresult
ingreater insulinchanges than shorter exposure and aless sudden
death. Si gnificant ly lower IGF-1 levels were fou nd in women who
had experienced a sudden unexpected loss compared to women
without ahistory ofsuddenloss.Thenumberofsuddenlosseswas
significantlyassociatedwithIGF-1levels,meaningthatthegreatest
decreaseinIGF-1wasshowninthegroupwiththemostlosses.
3.8 | Oxytocin
Bui et al99 investigated peripheral plasma OT levels in men and
women with CG. They compared a single assessment ofOT levels
of participants with a primary CG diagnosis to participants suf-
fering from depression as primary diagnosis and bereaved control
parti cipants with no c omorbid diagno sis. They found sign ificantly
higherOTlevelsintheCGgroupcomparedtothedepressedgroup.
TherewerenosignificantdifferencesbetweentheCGgroupandthe
groupofnon-pathologicalgrief.99 Secondar y analyses revealed that
aprimar y or probable CG diagnosis waspositively associated with
plasmaOTlevels.
3.9 | Prolactin
Laneetal104investigatedsexdifferencesinprolactin(PRL)changes
duringmourningin26spouses.Amongstothers,PRLplaysarolein
the stimulation of maternal care, acts as an endogenous anxioly tic
agentandregulatesoxytocinneurones.104TheyassessedserumPRL
beforeandafterasemi-structuredinterview.Theaimwastoexam-
inesexdifferencesin theassociationbetween the developmental
levels of the survivors’ object representation (DLOR). The DLOR
represents the verbal description of a person and the level of cogni-
tivecomplexityofthatdescription.100Theresultsshowasignificant
larger me an PRL change in wome n compared to men . A negative
correlat ion betwee n PRL change an d DLOR was found in wo men,
whereas a positive correlation was found in men.100
3.10 | Effects of bereavement interventions on
neuroendocrine stress markers
Three st udies exami ned the effe cts of bere avement inter ventions
on stress-related neuroendocrine markers. In the first study, the
effect of an activation programme on plasma cortisol and prolac-
tin levels wa s examined in 72 c lose female r elatives of c ancer pa-
tients.101P la sma co r tis oland pro la c tin ,a swell as anx iet y,de pre ssi on
andmentalexhaustion,wereassessedduringtheintervention,right
beforethedeathoftherelativeand1 and 2monthsafter loss. The
resultsshowthatanincreasingdegreeofmentalexhaustionduring
the treatment period is significantly associated with increasing corti-
sollevelsanddecreasingprolactinlevels.Furthermore,significantly
higher cor tisol levels were found 1 month after death compared to
the last assessment before death. Also, lower prolactin levels dur-
ing treatment were found in the activation group compared to the
control group.101 In the secon d study, the effe cts of a shor t-term
bereavement support group intervention with 119 widowed men
infectedwithHIVon immunevariablesandcor tisol levelswere as-
sessed.94Rece ntlyb ereavedHIVserop ositive(HI V+)andHI Vsero n-
egative(HIV–)menwererandomlyassignedtoeitherabereavement
supportgroupinterventionorastandardcaregroup.Plasmacortisol
wasassessedpre,postandat6-monthfollow-up.Significantlylower
cortisol levels were found in the intervention group compared to the
controlgroup6monthsaftertheinter vention.HIV+ men in the in-
tervention group showed significant decreases in cortisol levels from
pre-assessmenttofollow-up,whereasHIV–menintheintervention
groupshowedincreasedlevelsofcortisolwithinthesametime-pe-
riod.94Thethirdstudyassessedpredictiveeffectsofcatecholamines
asmoderatorsofabereavementinterventionandCGtreatmentout-
comes after bereavement.95Sixteenbereavedindividuals provided
informat ion on the Inventor y of Complic ated Grief (I CG) pre- and
post-psychotherapyandbloodepinephrine,norepinephrineanddo-
pamine were assessed 4 weeks beforethe intervention. Thepost-
treatment ICG-score was significantly predicted by pre-treatment
epineph rine levels. SAM a ctivity an d autonomous fu nction in the
participantsshowedimpairedCGoutcomesaftertherapy.95
3.11 | Summarized results of good-quality studies
Insummar y,theresultsof“goodquality ”studie ssug gestthefollow-
ing neuroendocrine changes after the loss of a loved one:
18 of 24
|
HOPF et al.
• Themoredeathsoflovedonessomeoneexperiences,thehigher
his/her the cortisol levels.77
• Morning cortisol levels are significantly higher in bereaved
compared to non-bereaved 2 weeks and 6 months after
bereavement.80
• The longe r the forewar ning of someo ne's death, th e higher the
cortisol levels after bereavement.80
• Bereaved men suf fering from emotional numbness 6 months after
loss show higher cortisol levels 12 months after death compared
to bereaved men who do not suffer from emotional numbness.78
• Compared to non-pathologically grieving subjects, people with
CG show flat tened diurn al cortisol s lopes, sug gesting tha t HPA
axis dysregulation is more pronounced in prolonged grief.79
PeoplewithCGshowsignificantlylowermorningandoverallcor-
tisollevelscompared to non-pathological grievers 2 years after
loss.77
• People with CG show significant higher cortisol morning re-
sponses5yearsafterlosscomparedtotwoyearsaf terloss.77
• Higher sco res in grief seve rity 5 year s after loss a re associated
with lower morning cortisol levels.77
4 | DISCUSSION
Thelos sofalovedonecanbeassoc iat edwit hne uroendo cr inealter-
ations and dysfunction both in the early and late stage of bereave-
ment.Thissystematicreviewsummarisesoriginalarticlesexamining
neuroendocrine correlates of social loss. Of the original studies
included in this review, most focused on HPAaxis (eg,cortisol) or
SAM-related hormones (epinephrine, norepinephrine) as primary
outcomes . These stud ies not only sugge st elevated me an cortisol
levels and flattened diurnal cortisol slopes, but also increased epi-
nephrineandnorepinephrinelevelsaftersocialloss.Ingeneral,flat-
tened diurnal cortisol slopes have been associated with negative
health outcomes in different study populations.104 Individuals suf-
feringfromCGshowflatteneddiurnalcortisolslopes70,77 as well as
lowermorningandmeancortisollevels,77thanthoseshowingnon-
complicatedgrief. Furthermore, both closenessto the deceased82
and grief severity79 play an important role in the development of
neuroendocrinedysregulations.Theclosertherelationshipandthe
moreorenduringthesubjectiveimpairmentisarticulated,themore
endocrine dysregulation is pronounced. Particularly, higher grief
levels and lower social support are associated with higher cor tisol
levels.85 Inaddition,specificstressors, aswellaspsychologicaland
demographicfactors,partiallyaccountforHPAaxisalterations.For
example,increases in separation anxietyinthe course of bereave-
ment were associated with higher levels of cortisol84 and having a
longer forewarning before death lead to higher cor tisol levels than
experiencinganunexpected loss.78 Sudden, unexpected losses, as
well as a risingnumber of losses, areassociated with lowerinsulin
levels,showingthatthosecontextvariablesinfluencehealth-reduc-
ing neuroendocrine alterations after bereavement.98Apositiveaf-
fect was i nversely cor related with co rtisol leve ls,90 whereas rising
emotional numbness in men during the course of bereavement en-
hanced cortisol levels,78 suggesting again that psychological vari-
ablesareimportantwhenexaminingneuroendocrine changesafter
loss. Regarding gender differences, one study revealed that men
showeddecreasingcortisollevels,whereaswomenshowedincreas-
ing cortisol levels during the course of bereavement.78 Older men and
women showed stronger alterations in their neuroendocrine stress
responsesthanyoungercohorts,85,89 indicating that high age may
haveanadditiveeffectonlossconsequences.Furthermore,changes
instress-relatedalterationswereshown,especiallyintheearlystage
ofbereavement,althoughthereisinter-individualvariability.88 In the
latter study, however, almost no direct correlations between psy-
chologicalandbiochemical reactions were found. Neuroendocrine
alterationswerenotonlyfounddirectlyafterbereavement,butalso
months afterlossexperience.97 Interesting results evolved with re-
gard to psychiatric diseases: especially depressive symptoms were
associated with higher cortisol levels86,102 andhigher cortisol non-
suppression88inbereavedsubjects.Furthermore,individualssuffer-
ingfrom PTSDafter a traumaticloss showedhigher cor tisol levels
thanthosewithnotrauma-relatedpsychiatricdiagnosis.80Thesame
study su ggests t hat trauma-re lated psycho patholog y may foster a
prolonged neuroendocrine response to social loss up to 8 years after
the event . One study i nvestigated OT as a b iomarker of gri ef and
fo und hig her OT level sin peo p le suf f eri n gf r om CG.99 Regarding pro-
lactin changes, women have higherprolac tin levelsthan men after
havingbeeninterviewedaboutthedeceasedpartner.Interestingly,
womenwhohaveamore complexinsightintothedeceasedperson
alsoshowhigherprolactinlevels, whereastheopposite association
is observed in men.100
In summar y,t hese studie s suggest that n ot only bereaveme nt
byitself, but alsobereavement-associated psychopathologyinpar-
ticularisassociatedwithstress-relatedneuroendocrinealterations.
This is in lin e with research o n trauma,53, 54 loneliness26-28 ,10 5 and
disrupted attachment,44 which can all be individual psychosocial
aspects of bereavement and separately serve as causal factors of
stress-relatedneuroendocrinedysregulation.Bereavementcanhave
health-impairing and fatal consequences for the surviving individ-
ual4,6,9,106and,asaconsequence,inter ventionshavebeendesigned
andevaluatedto buf fer theseeffects. The available studiesonthe
effec ts of these inter ventions found the i ntervention to red uce corti-
sol levels.94,95Furthermore,epinephrinelevelspredictedpsychopa-
thology-relatedtreatmentoutcomessuggesting thatpre-treatment
stress levels moderated the effectiveness of the intervention.95 In
linewiththis,catecholamineswerecorrelatedwithhelplessnessand
hopelessnessinthecourseofbereavement,87 and thus can ser ve as
endocrine markersofsubjective burdenand treatment effects.An
interventionbeforethepartners’deathwasfoundtoelevatecortisol
levels andreduce prolactin levels, especially right before death.101
Thelatter finding is consistentwith the hypothesis thatgriefisac-
tivated by an intervention and that the ac tive mourning may have a
prophylacticvaluetotherelative’sgriefreaction.101
The results indicate that, even years after loss, bereavement
mightbeassociatedwithneuroendocrinechanges.Thesechangesare
|
19 of 24
HOPF et al .
moderatedbygriefseverity,psychiatricstateandpsychologicalreac-
tions to loss, as well as ageand gender.On a psychobiologicallevel,
neuroendocrine responses may serve as moderators between the
los s-e ventandl ong-te rmpsychologi ca loutcomes(Figur e2).Ho wev er,
because of the methodological difficulties and contradictory results of
thestudies,theseconclusionsmustbetreatedwithcaution.
For exampl e, Ong et al90 found that prolonged forewarning of
death was a ssociated with hig her cortisol le vels. They argue t hat
alonger duration of care is associated with more stressful experi-
ences, an d thus leads to s tronger physiol ogical stre ss reaction s.90
ResearchonthedevelopmentofPGD/PCBDshowsthatsuddenness
ofdeathisa riskfactor,107 which initially appears to contradict the
findings of Ong et al90. Inthis context, it is important to note that
PGD/PCBD sy mptoms are not o nly charac terised by phy siologica l
distor tions such as elev ated cortisol l evels, but also by e motional
andbehaviouralsymptomssuchasintenseyearning,longingoremo-
tional pain.108Thecontradictoryfindingsmayindicatethatthereare
moderator variables between instant physiological reactions and the
development of PGD/PCBDthat fosterthe maintenance of an ab-
normallyhighcortisollevel.Itisnotclear,yet,whetherhighcortisol
levels or flattened diurnal slopes are risk factors of the development
of PGD/PCBD. So f ar, only corre lative conclusions c an be drawn
aboutHPAaxisdysfunctionandPGD/PCBD.Toobtainabetterun-
derstandingofwhatroleHPAaxisfunctionmayplayinPGD/PCBD,
it would be essential to measure cortisol levels in regular intervals
over a longer time span after bereavement at the same time as mea-
suring moderator variables.
Therearefurther inconsistent study results regardingmorning
and mean cortisol. Buckley et al80 found significantly higher morning
cortis ol levels in the b ereaved, wh ereas Perez et a l77 found lower
morning c ortisol le vels in people w ith CG. This di screpanc y might
be because of the dif ferent methods and timeframes investigated.
Cortisol levels might be differentially affected depending on the
time since loss. Furthermore, Buckley et al80 compared grieving
with non-g rieving peo ple, where as Perez et al77 compared people
with CG and non-CG. Additionally,two studies found significantly
elevated cor tisol levelsinbereavedandanticipatory bereaved,86,92
whereas two other studies did not.79,9 3 One reason for the conflic t-
ingresults could be their methodological diversity,using different
measurementsofcortisol,differenttime-framesanddifferentmea-
surementfoci.Furthermore, thestudypopulationsweresomehow
different. For example, Spratt a nd Denney93 only examined sud-
denlybereavedparents,whereasJacobsetal.87 and Irwin92 had par-
ticipant swithalongerhistoryofend-of-lifecare.
4.1 | Limitations
Thestudiesincludedinthissystematicreviewrevealsomelimita-
tions – one of themistheoverall smallsamplesize,which limits
the statisticalpower of the results.Although21studieswere of
fair-qualityandonlyfivestudies wererated ashigh-quality stud-
ies, the results must be interpreted with caution. For one thing,
most of the s tudies consider ed the loss event as the ex posure
FIGURE 2 Modelwithsummarisedresult s(includingpotentialmoderators/mediators)ofthestudiesinvestigatingneuroendocrine
mechanismsofgrief.DHE AS,dehydroepiandrostheron-sulphate;DST,dexamethasonesuppressiontest;IGF,insulin-likegrow thfac tor;OT,
oxytocin;PTSD,post-traumaticstressdisorder
Critical life-event
Social Loss
Higher mean cortisol levels
Flattened diurnal cortisol
slopes
Higher morning cortisol levels
Blunted cortisol suppression
after DST
Higher cortisol: DHEAS ratio
Higher OT levels
LowerIGF levels
Altered Prolactin Levels
Relationship/Closeness to the
deceased
Psychiatric symptoms
(depression, PTSD)
Time after bereavement
Emotional reactions to loss
Suddenness of death
(Subjective) Grief severity
Age
Gender
Seperation Anxiety
Congnitive Representation of
Bereaved
Moderators/Mediators Neuroendocrine changes
20 of 24
|
HOPF et al.
variable without assessing continuous levels of subjective grief.
According tosomeresults,griefseverity andsubjectiveappraisal
of loss have a stronger influence on neuroendocrine reactions
thanthelossexperienceitself.77,78,84,85,90Basedonthis,grieflev-
els should be measured continuously to enhance the validit y of the
study.Furthermore,manystudiestookplace yearsaf ter the loss
event. Potential confounding factorscould have occurred within
thistimeframe,whichmakesitdifficulttodisentangletheeffect
of bereavementfrom other factors influencing long-term endo-
crinechanges.Anotherimportantfactisthatmanystudiesdonot
report whether the survivors made use of social support or psy-
chosocialbereavementinterventions,althoughsocialsupportcan
bufferthelo ss re act io n.Inthest udiesun de rly in gt hisreview, gr ief
severities, as well as neuroendocrine outcomes, were assessed
differently, which hampers their comparability or calculation of
effect sizes using meta-analytic methods.Moreimportantly and
asaresultoftheunpredictabilityofdeath,neuroendocrinemark-
erswerenotassessedbeforethelossofalovedperson,andthus
provide limited information on stable predictors only.
Despite the limitations mentioned, thefindingsof elevatedcor-
tisol and flattened diurnalslopesare relatively reliable, whichsug-
gests that they seem to be robust.
Ithastobenoted that,sofar,onlycortisol has beenexamined
moreextensivelyandresearchonotherneuroendocrinemeasures,
suchasOT,AC THorcatecholamines ,iss tillsc arce.99Theonlystudy
examining OT in the context ofbereavement97 has some method-
ologica l flaws, as the aut hors measure d OT in the periphe ry. The
assessment of peripheral OT levels iscriticised because ofits lack
ofassociationwithcentral-nervousOTlevels,109 and ongoing meth-
odological discussion aboutthe reliablemeasurement of OTin the
periphery even.110Thesepoints make it is difficult to draw reliable
conclusio ns, especia lly with respe ct to neuroen docrine gri ef reac-
tions in the human central ner vous system. Based on the findings of
animal studies on social loss as well as human studies with healthy
couples,however,itcanbeassumedthatthepainfulexperienceofa
closeperson'sdeathmightalsoinvolvetheOTsystem.Thereward-
ing,stimulatingeffectofawell-functioningrelationshipiseliminated
and the OT system remains under-stimulated.111 This under-stim-
ulation could in the long run even be related to the symptoms of
PGD/PCBD. Additionally,studies investigating neural correlates of
sociallossindicategrief-relatedalteredactivationinbrainareassuch
as the NAcc112 andtheACC,
113 which are associated with the re-
ward-systemandhighOTreceptordensities.However,endogenous
OTmechanismsinthecentralnervoussystemc annot bemeasured
inthehumanliving brain sofar,whichlimits thepossibilitiestotest
fordirectinvolvementofOTinthegrievingprocess.Therefore,44 an-
imal models can be helpful to better underst and those mechanisms.
Moreover,humanandanimalstudiescancomplementeachotherin
ameaningfulwaybecausetheirmethodsleadtocontext-dependent
results:experimentalsettingsarear tificialandmayleadtodif ferent
reactionsthanreal-lifeevents.45Abovethis, animalmodelscannot
give us sufficient insights into the psychological reactions to loss.
Ontheotherhand,sofar,thehighlyindividualhumangrief-reaction
cannot be investigated in a standard procedure or related to spe-
cific neuroendocrine changes in the living brain. Despite the lack
oftransferability,animalresearch cangive us importanthintsasto
where to st art in human research and what hypotheses to establish.
Therefore,the combination of knowledgefromanimalandhuman
researchcanprovideabroaderpictureonthiscomplextopic.
Finally,somelimitations shouldbe mentionedregardingthe re-
cruitmentofbereaved individuals. First, bereavedpeopleareinan
altered stateofmind, with somedescribingnumbnessand agenu-
inelossofinterestindailymat ters.Theymightbedifficulttoreach
withbroadrecruitmenttools.Forthosewhoare,thereasonforpar-
ticipati ng might be the h ope for psyc hologica l suppor t, meaning a
rather vulnerable and selective subgroup might agree to participate.
Astoworkingwithabereavedsample,itcanbechallengingtomain-
tainthemotivationof the participantstoremaininthestudy.Also,
becausethetimeofdeathmostoftenoccursunforeseen,thereusu-
ally are no individual baseline measuresbefore the loss.Toobtain
pre-bereavement measurements, participants should be recruited
before the deathof theircloseone,treatingthemwith the highest
sensibility and psychological supervision.
4.2 | Future research and implications for
psychosocial interventions
Tohelpestablishacomprehensivemodeloftheneuroendocrinefac-
tors underlying thepsychobiologicalreactions to socialloss, in ad-
dition to th e neuroend ocrine st ress respo nse, future r esearch c an
benefit from a focus on further and interacting neuroendocrine sys-
tems.Animalresearchonsociallosssuggests that,forexample, the
OTsystem interactswith the HPA axis andmight be involveddur-
inggriefreactions. BothCRHandOT have been shownto interact
withthedopamine) system,whichregulatesrewardandisinvolved
indepressive disorders and addic tion. In both animalsandhumans,
dopamine appears to play a role in the formation of a romantic re-
lationship;for example, reward-associated brain regions are highly
activatedinassociationwithpositiveattachmentinteractions,114-116
and it is assumed that this system could also be affected after loss
by remaining under-stimulated.111 Indeed, human studies already
indicateactivationsinbrain-regionswithhighOTanddopaminere-
ceptor density.91,92Inlinewiththis,withdrawalfromdrugabusehas
been associated with similar activation patterns compared to sepa-
ration from a partner.46,111
Inaddition,longitudinal studiesassessingsubjec tiveandneuro-
endocrine markers before and after loss could minimise confounding
inter-individualvariationsandtherebyimprovestatisticalpowerand
long-term pre dictive power. Alth ough necessari ly in such studie s,
loss wouldalways bepredicted by a lethal illness, thereby limiting
the range of different possible events to trigger grief.
Initial studies investigating neuroendocrine alterations after a
bereavementintervention show promising effects and suggest that,
beside subjective measures, neuroendocrineand stress-related out-
comes can serve as meaningful indicators of treatment success.94,95,101
|
21 of 24
HOPF et al .
Inthiscontext,itisimportanttokeepinmindthatsubjectiveandob-
jectivemeasuresoftendivergeinresearchonstress,andthusitisim-
portantto revealthe differencesbetweenthesemeasurements.This
leadstothenextstepofexploringthereasonswhythesedifferences
occurandwhattheymeanfortreatmentsuccess.However,tobetter
interpret the meaning and importance of neuroendocrine measures
fortherapeuticsuccess,moreresearchisnecessary.Furthermore,the
assessment of neuroendocrine measures is associated with some hur-
dles.Forexample,theassessmentofblood,urineorsalivasamplesis
time-consumingandmaybeareasonforgrievingparticipants notto
take part in a study. One possibility to address this issue and to enable
dataon aggregated cortisol levels to be collected overan extended
time period of weeks to months is the use of hair samples to quan-
tify,forexample,cortisolsecretion.Ecologicalmomentaryassessment
could be used to reduce the participants’burden of being torn out
of theirevery-day life. Regardlessof discussing in what way neuro-
endocrinemeasures serve as an indicatorfor treatmentsuccess,the
ResearchDomainCriteria(RDoC)initiativeoftheNationalInstituteof
MentalHealthexplicitlyrecommendstheinclusion ofobjectivemea-
sures into interventional studies.117All in all, more research is nec-
essary,investigatingpotentialfactors thatmayinfluencethe efficacy
ofbereavementinterventionswithdifferentpopulations,varyingage
groupsandsocialbackgroundSofar,onlyoneearlystudyhasinvesti-
gated the effects of a psychosocial intervention before the separation
experiencewith the aim of preventing grief reactions and associated
neuroendocrinealterations after loss.As faraswe know, the antici-
pation of losing someone close may already lead to neuroendocrine
changes,84 which highlights the need for early strategies preventing
neuroendocrine dysfunction and buffering the negative effects of so-
cial loss. It is known fromstudies on healthy couples’ interventions
that positive psychosocial interventions together with the partner or
afamilymember activatesthe rewardsystemand exertsstress-buff-
ering effects.118,119This leads to the assumption that the described
stressandunder-stimulationreactionto sociallosscanbeinfluenced
by appropriate interventions and specific death and bereavement
management programmes.6,46 Psychosocialinterventions before the
loss of the partnermight strengthen the bond, reduce stress levels,
affectthe endogenous OTrelease and buffergrief-relatedstress-re-
actions,preventinglong-termnegativehealtheffectssuchasthede-
velopmentofCGorotherpsychiatricproblems.Accordingtothedual
processmodelof coping,120 oscillations between thoughts about the
lost attachment figure on the one hand and evaluating a future with-
out the lost loved one on the other hand are considered important
factorsofanadaptivegriefcoping process.In thiscontext,interven-
tionsthathelp to strengthen the bond, and which make unresolved
issuesa subject of discussion, might fostera healthy coping process
and therefore affect neuroendocrine as well as psychological health
changesaftertheloss.Althoughthereisnostudyinvestigatingtheef-
fectsofpre-deathinterventionsonneuroendocrinereactionssuchas
OTsignalling,initialstudies showthatpsychosocialinterventionsbe-
forelossareabletoimprovethewell-beingoftheparticipants.121,122
However, this hypothesis needs further investigationand additional
researchisnecessarytounderstandwhethermechanismssuchasOT
signallingcontributetothe efficacyofsuchtreatments.Furthermore,
itis importanttoconsider inter-individualdifferenceswhen deciding
on whether to implement an intervention or not.
Insummary,neuroendocrinecorrelatesofanticipatorygriefand
grief after social loss could help us identify individual needs and
serve as tools to evaluatenot only impairment, but also treatment
success. I n the long run , this knowle dge might allow t he develop-
mentofspecificinterventionsthatimprovestress-relatedresponses
in the survivors and thereby their health.
ACKNOWLEDGEMENTS
We thank the G erman FAZIT-Stiftung and t he Marsilius Koll eg at
HeidelbergUniversit yformakingthis review possiblebyfinancially
supportingDHandBD.CARissupportedfromtheOlympiaMorata
Program atHeidelberg Universit y.Furthermore,we thankHannah
Melles (HM), whohelped us findrelevant articlesand Star Dubber
for proofreading the manuscript submitted for publication. Open
accessfundingenabledandorganizedbyProjektDEAL.
CONFLICT OF INTERESTS
Theauthorsdeclarethattheyhavenoconflictsofinterest .
AUTHOR CONTRIBUTIONS
DH,ME, CARandBD definedthe literaturesearch criteria.DHcon-
ductedtheliteraturesearchandsummarisedthefindings.DH,MEand
CARratedthe internal validity ofthestudiesbythe National Heart,
Lung,and Blood InstituteStudyAssessmentTool.DH, CAR,MEand
MWwrotethepaper.BD reviewedthemanuscriptandgavecritical
advice.
PEER REVIEW
Thepeer reviewhistoryforthis article isavailableathttps://publo
ns. com/publo n /10.1111/j ne.12887.
ORCID
Dora Hopf https://orcid.org/0000-0002-9476-0478
Monika Eckstein https://orcid.org/0000-0002-1846-4992
Corina Aguilar-Raab https://orcid.org/0000-0001-9956-7047
Marco Warth https://orcid.org/0000-0003-3277-5516
Beate Ditzen https://orcid.org/0000-0001-5853-4572
REFERENCES
1. ZisookS,ShearK.Griefandbereavement :whatpsychiatristsneed
to know. World Psychiatry.2009;8:67-74.
2. GoodkinK,BaldewiczTT,AsthanaD,etal.Abereavementsuppor t
group intervention affects plasma burden of human immunodefi-
ciency virus type 1. Report of a randomized controlled trial. J Hum
Virol.2001;4:4 4-54.
3. Kristensen P, Weisaeth L, Heir T. Bereavement and mental
health after sudden and violent losses: a review. Psychiatry.
2012;75:76-97.
4. Ramsay S , Ebrahim S, Wh incup P, et al. Social en gagement and
the risk of c ardiovascular disease mortality: result s of a pro-
spective population-based study of older men. Ann Epidemiol.
20 0 8;18: 476 -48 3.
22 of 24
|
HOPF et al.
5. BiondiM,PicardiA.Clinicalandbiologicalaspectsofbereavement
andloss-induceddepression:areappraisal.Psychother Psychosom.
1996;65:229-245.
6. Assar eh AA, S harpley CF, McFarla ne JR, Sach dev PS. Biol ogical
determinants of depression following bereavement. Neurosci
Biobehav Rev.2015;49:171-181.
7. Clayton PJ. Bereavement and depression. J Clin Psychiatry.
1990;51(Suppl):34-40.
8. ManzoliL,VillariP,PironeG,BocciaA.Maritalstatusandmortal-
ity in the elderly: a systematic review and meta-analysis. So c Sci
Med.2007;64:77-94.
9. Moon JR,Kondo N, Glymour MM, SubramanianSV.Widowhood
andmor talit y:amet a-analysis.PLoS One.2011;6:e23465.
10. Carey IM, Shah SM, DeWilde S, Harris T,Victor CR , Cook DG .
Increased risk of acute cardiovascular event s after part-
ner bereavement: a matched cohort study. JAMA Int Med.
2014;174:598-605.
11. Dawson DK. Acute stress-induced (takotsubo) cardiomyopathy.
Heart.2018;104(2) :96-102.
12. Templin C, Ghadri JR, Diekmann J, et al. Features and out-
comes of Takotsubo (Stress) cardiomyopathy. N Engl J Med.
2015;373:929-938.
13. Rosner R, Wagner B, Komplizier te T. In: Maercker A ed.
Posttraumatische Belastungsstörungen. Berlin, Heidelberg:
Springer;2009:441-456.
14. Killikelly C,Maercker A.ProlongedgriefdisorderforICD-11:the
primacy of clinical utility and international applicability. Euro J
Psychotraumatol.2018;8(Suppl6):1476441.
15. PrigersonHG ,Horowitz MJ,JacobsSC,etal. Prolongedgriefdis-
order:psychometricvalidationofcriteriaproposedforDSM-Vand
IC D -11. PLoS Medicine. 2009;6:e1000121.
16. Steinig J, Kersting A. Anhaltende komplexe Trauerreaktion – ein
neuesKrankheitsbild?Psych Update.2015;9:281-295.
17. Shear K, Sh air H. Attachm ent, loss, and co mplicated grie f. Dev
Psychobiol.2005;47:253-267.
18. PrigersonHG,Vander werkerLC,MaciejewskiPK.Acaseforinclu-
sionofprolongedgriefdisorderinDSM-V.In:StroebeMS,Hansson
RO, Schut H, Stroebe W, eds.Handbook of bereavement re search
and practice: Advances in theory and intervention.Washington, DC:
AmericanPsychologicalAssociation;200 8:165-186.
19. DiagnostischesundStatistischesManualPsychischerStörungen-
DSM-5®.Göttingen;Bern;Wien[u.a.]:Hogrefe2015.
20. Shear MK ,Simon N, Wall M, et al. Complicatedgrief and related
bereavementissuesforDSM-5.Depress Anxiety.2011;28:10 3-117.
21. Wagner B. Wann ist Trauer eine psychische Erkrankung?
Psychotherapeutenjournal.2016;3:250-255.
2 2. Maciejew ski PK, Mae rcker A, Bo elen PA, et al. " Prolonge d grief
disorder" and "persistent complex bereavement disorder", but
not "complicated grief", are one and the same diagnostic en-
tity: an analysis of data from the Yale Bereavement Study. World
Psychiatry.2016;15:266-275.
23. Maciejewsk i PK, Priger son HG. Pr olonged, bu t not complic ated,
grief is a mental disorder. Br J Psychiatry.2017;211:189-191.
24. LuoY,HawkleyLC,WaiteLJ,CacioppoJT.Loneliness,health,and
mortality in old age: a national longitudinal study. Soc Sci Med.
2012;74:907-914.
25. Cacioppo JT, Hawkley LC, Crawford LE, et al. Loneliness and
health: p otential mechanisms. Psychosom Med.2002;64:407-417.
26. Hackett RA , Hamer M , Endrighi R, Br ydon L , Steptoe A.
Loneliness and stress-related inflammatory and neuroendocrine
responses in older men and women. Psychoneuroendocrinology.
2012;37:1801-1809.
27. Schutter N , Holwerda TJ, Stek ML, Dekker JJM, Rhebe rgen D,
ComijsHC.Lonelinessinolderadultsisassociatedwithdiminished
cortisol output. J Psychosom Res.2017;95:19-25.
28. BrownEG,GallagherS,CreavenA-M.Lonelinessandacutestress
reactivity: a systematic review of psychophysiological studies.
Psychophysiolog y.2018;55:e13031.
29. Gerst-Emer son K, Jayaw ardhana J. L onelines s as a public he alth
issue: the impact of loneliness on health care utilization among
older adults. Am J Public Health.2015;105:1013-1019.
30. del-Pino-Casado R, Espinosa-Medina A, López-Martínez C,
Orgeta V.Senseof coherence, burdenandmental healthin care-
giving: a systematic review and met a-analysis. J Affect Disord.
2019;242:14-21.
31. Backhaus J, Junghanns K , Hohagen F. Sleep disturbances are
correlated with decreased morning awakening salivary cortisol.
Psychoneuroendocrinology.200 4; 29:1184 -1191.
32. SbarraDA,Hazan C.Coregulation,dysregulation, self-regulation:
an integrative analysis and empirical agenda for understanding
adultattachment,separation,loss,andrecovery.Pers Soc Psychol
Rev.2008;12:141-167.
33. Doerr JM , Nater UM , Ehler t U, et al. Co-variation of fa-
tigue and psychobiological stress in couples’ everyday life.
Psychoneuroendocrinology.2018;92:135-141.
34. Aguilera G .Chapter 8. The hypothalamic–pituitary–adrenal axis
and neuroendocrine responses to stress. In Fink G, Pfaff DW,
Levine G eds. Handbook of Neuroendocrinology. London, UK:
Elsevier ;2012:175-196.
35. StephensMAC ,WandG.StressandtheHPAaxis:roleofglucocor-
ticoids in alcohol dependence. Alcohol Res.2012;34:468-483.
36. Stalder T, Kirs chbaum C , Kudielka BM , et al. Ass essment of th e
cortisol awakening response: expert consensus guidelines.
Psychoneuroendocrinology.2016;63:414-432.
37. Adam EK, Kumari M. Assessing salivary cor tisol in large-
scale, epidemiological research. Psychoneuroendocrinology.
2009;34:1423-1436.
3 8. Chatterj ee O, Patil K, Sa hu A, et al. A n overview of th e oxyto-
cin-oxytocin receptor signaling network. J Cell Comm un Signal.
2016;10:355-36 0.
39. Scheele D, WilleA ,KendrickKM,etal.Oxy tocin enhances brain
reward system responses in men viewing the face of their female
partner. Proc Natl Acad Sci USA.2013;110:20308-20313.
40. Uvnas-MobergK,PeterssonM.Oxytocin,amediatorofanti-stress,
well-bei ng, social inte raction, g rowth and hea ling. Zeitschrift fur
Psychosomatische Medizin und Psychotherapie.2005;51:57-80.
41. DitzenB,SchaerM,GabrielB,BodenmannG,EhlertU,HeinrichsM.
Intran asal oxytoc in increases pos itive communic ation and reduce s
corti sol levels duri ng couple confl ict. Biol Psychiatry.2009;65:728-731.
42. NeumannID.Involvementof thebrain oxytocinsystem in stress
coping:interactionswiththe hypothalamo-pituitary-adrenalaxis.
Prog Brain Re s.2002;139:147-162.
43. Hurlemann R, Scheele D. Dissecting the role of oxytocin in
the formation and loss of social relationships. Biol Psychiatry.
2016;79:185-193.
44. Pohl TT,YoungL J,Bosch OJ.Lostconnections: oxytocin and the
neural , physiologic al, and beh avioral conse quences of dis rupted
relationships. Int J Psychophysiol.2019;136:54-63.
45. BoschOJ,YoungL J.Ox ytocinandsocialrelationships:fromattach-
ment to bon d disruption. Curr Topics Behav Neurosci20 1 8;3 5 : 97 -11 7.
4 6. Bosch OJ, Nai r HP,Ahe rn TH, Neu mann ID, Young LJ . The CRF
system mediates increased passive stress-coping behavior fol-
lowing the loss of a bon ded par tner in a monogamous rodent.
Neuropsychopharmacology.2009; 34 :140 6-1415.
47. McNealN,ScottiMA ,WardwellJ,etal.Disruptionofsocialbonds
induces behavioral and physiological dysregulation in male and fe-
male prairie voles. Auton Neurosci.2014;180:9-16.
48. SunP,SmithAS,LeiK,LiuY,WangZ.Breakingbondsinmaleprai-
rie vole:Lo ng-te rmef fe ctsone mo ti on alandso ci albe havior,p hys-
iology,andneurochemistry.Behav Brain Res.2014;265:22-31.
|
23 of 24
HOPF et al .
49. BoschOJ,DabrowskaJ, ModiME, et al. Oxytocinin thenucleus
accumbens shell reverses CRFR2-evoked passive stress-cop-
ing after partner loss in monogamous male prairie voles.
Psychoneuroendocrinology.2016;64:66-78.
50. GrippoA J,TrahanasDM,Zimmerman RR, PorgesSW, Car terCS.
Ox y t oc i np rot ect s ag a ins tn e gat ivebe h av ior a la nda uto nom iccon -
sequencesoflong-termsocialisolation.Psychoneuroendocrinology.
2009;34:1542-1553.
51. GrippoAJ,Pournajafi-NazarlooH,SanzenbacherL,etal.Peripheral
oxytocinadministrationbuffersautonomicbutnotbehavioralre-
sponses to environmental stressors in isolated prairie voles. Stress.
20 1 2;1 5: 14 9-1 61 .
52. Meyer-Lindenberg A , DomesG, Kirsch P,HeinrichsM. Oxy tocin
and vasopressin in the human brain: social neuropeptides for
translational medicine. Nat Rev Neurosci.2011;12:524-538.
53. Daskalaki s NP, McGill MA , Lehrner A , Yehuda R. Endo crine as-
pects of PTSD: hypothalamic-pituitar y-adrenal (HPA) axis and
beyond. In:Martin CR, Preedy VR , Patel VB, eds. Comprehensive
Guide to Post-Traumatic Stress Disorders. Heidelberg: Springer;
2016:245-260.
54. MorrisMC,HellmanN,AbelsonJL,RaoU.Cortisol,heartrate,and
bloodpressureasearlymarkersofPTSDrisk:asystematicreview
andmeta-analysis.Clin Ps ychol Rev.2016;49:79-91.
55. OlffM,vanZuidenM.Neuroendocrineandneuroimmunemarkers
inPTSD:pre-,peri-andpost-traumaglucocorticoidandinflamma-
tory dysregulation. Curr Opin Psychol.2017;14:132-137.
56. MasonTM ,DuffyAR.Complicatedgriefandcor tisolresponse:an
integrative review of the literature. J Am Psychiatr Nurses Assoc .
2019;25:181-188.
57. Moher D, Libe rati A, Tetzl aff J, Altm an DG. Prefe rred repo rting
items for systematic reviews and met a-analyses: the PRISMA
statement. J Clin Epidemiol.2009;62:10 06-1012.
58. Knowles LM, Ruiz JM ,O'Connor MF.A sys tematicreview of the
association between bereavement and biomarkers of immune
function. Psychosom Med.2018;80:415-433.
59. KimK,JacobsS.Neuroendocrinechangesfollowingbereavement.
In:StroebeMS,StroebeW,HanssonRO,eds.Handbook of bereave-
ment: Theory, research, and intervention.NewYork,NY:Cambridge
UniversityPress;1993:143-159.
6 0. Archer J, Fish er H. Bereaveme nt and reaction s to romantic re-
jection:A psychobiological per spective. In:StroebeMS ,Hansson
RO,Schut H, Stroebe W,eds.Handbook of Bereavement Research
and Practice: Advances in Theory and I ntervention.Washing ton,DC:
AmericanPsychologicalAssociation;200 8:349-371.
61. HallM, Irwin M. Physiological indices of functioning in bereave-
ment. In: Stroebe MS, Hansson RO, Stroebe W, Schut H, eds.
Handbook of Bereavement Research: Consequences, Coping, and
Care. Washington, DC: American Psychological Asso ciation;
2001:473-492.
62. Lieberw irth C , Wang ZX. T he neurobi ology of pai r bond form a-
tion, bo nd disruption , and social buf fering. Curr Opin Neurobiol.
2016;40:8-13.
6 3. N ational Heart L , and Blood Institute. Qualit y assessment tool
forobservational cohort andcross-sectional studies2017[serial
online];Availablefrom:https://www.nhlbi.nih.gov/health-topics/
s t u d y - q u a l i t y - a s s e s s m e n t - t o o l s
64. KostenTR,JacobsS,MasonJW.Thedexamethasonesuppression
test during bereavement. J Nerv Mental Dis.1984;172:359-360.
65. ShuchterSR,ZisookS,KirkorowiczC,RischC.Thedexamethasone
suppression test in acute grief. Am J Psyc hiatry.1986;143:879-881.
66. DasM,BerriosGE.Dexamethasonesuppressiontestinacutegrief
reaction. Acta Psychiatr Scand.1984;70:278-281.
67. HoferMA .Relationshipsasregulators:apsychobiologicperspec-
tive on bereavement. Psychosom Med.1984;4 6:18 3-197.
6 8. Jacobs S, Maso n J, Kosten T, et al. Acute be reavement, t hreat-
enedloss, ego defenses andadrenocorticalfunction. Psychother
Psychosom.1985;44:151-159.
69. JacobsS,BrownSA ,MasonJ,etal.Psychologicaldistress,depres-
sion and prolactin response in stressed persons. J Human Stress.
1986;12:113 -118.
70. SchieleMA,CostaB,AbelliM,etal.Oxytocinreceptorgene vari-
ation,behaviouralinhibition,andadultseparationanxiety:rolein
complic ated grief. World J Biol Ps ychiatr y.2018;19:471-479.
71. Fagundes CP, Murdock KW, LeRoy A, Baameur F, Thayer JF,
Heijnen C . Spousal bereavement is associated with more
pronounced ex vivo cytokine production and lower heart
rate variability: Mechanisms underlying cardiovascular risk?
Psychoneuroendocrinology.2018;93:65-71.
72. Luecken LJ , Appelh ans BM. Ear ly parent al loss an d salivar y cor-
tisol in young adulthood: the moderating role of family environ-
ment. Dev Psychopathol.2006;18:295-308.
73. Meinlschmidt G, Heim C. Decreased cortisol awakening re-
sponse af ter early loss experience. Psychoneuroendocrinology.
2005;30:568-576.
74. Nicolson NA .Childhood parental lossand cortisol levelsin adult
men. Psychoneuroendocrinology.2004;29:1012-1018.
75. Tyrka AR , Wier L, Price LH, et al. C hildhood parental los s and
adult hypothalamic-pituitary-adrenal function. Biol Psychiatry.
2008;63 :1147-1154.
76. Luecken LJ , Hagan MJ, S andler IN , et al. Lon gitudina l mediator s
of a randomized prevention program ef fect on cortisol for youth
from parentally bereaved families. Prev Sci.2014;15:224-232.
77. Perez HCS , Direk N, Mil ic J, Ikram MA , Hofman A , Tiemeie r H.
The Impact of complicated grief on diurnal cortisol levels t wo
years after loss: a population-based study. Psychosom Med.
2017;79:426-433.
78. RichardsonVE,Bennet tKM,CarrD,Gallagher S,KimJ,FieldsN .
Howdoesbereavementgetundertheskin?Theeffect soflate-life
spousal loss on cortisol levels. J G erontol Ser B-Psychol Sci Soc Sci.
2015;70:341-347.
79. Cohen M, Granger S, Fuller-Thomson E. The association be-
tween bereavement and biomarkers of inflammation. Behav Med.
2015;41:49-59.
80. BuckleyT,BartropR,McKinleyS,etal.Prospectivestudyofearly
bereavement on psychological and behavioural cardiac risk fac-
tors. Intern Med J.20 09;39:370-378.
81. O'Connor M F,Wellis ch DK, Stan ton AL, Olm stead R, I rwin MR .
Diurnal cortisol in complicated and non-complicated grief:
slope dif ferences across the day. Psychoneuroendocrinology.
2012;37:725-728 .
82. Holland JM , Rozalski V, Thomp son KL, et a l. The uniqu e impact
oflate-lifebereavementandprolonged grief ondiurnal cortisol. J
Gerontol Ser B-Psychol Sci Soc Sci.2014;69:4-11.
83. Jacobs SC , Mason JW, Kosten TR, Wahby V, Kasl SV, Ostfeld
AM. Bereavement and catecholamines. J Psychosom Res.
1986;30:489-496.
84. Jacobs SC , Mason J, Koste n TR, Kasl S V,Os tfeld AM, Wa hby V.
Urinaryfreecortisolandseparationanxiety earlyinthecourseof
bereavement and threatened loss. Biol Psychiatry.1987;22:148-152.
85. VitlicA,KhanferR,LordJM,CarrollD,PhillipsAC.Bereavement
reduces neutrophil oxidative burst only in older adults: role
of the HPA axis and immunesenescence. Immunit y & Ageing.
2014;11:13.
86. RoyA,GallucciW,AvgerinosP,LinnoilaM,GoldP.TheCRHstim-
ulation test in bereaved subjects with and without accompanying
depression. Psychiatry Res.1988;25:145-156.
87. Jacobs S, BruceM,KimK. Adrenalfunctionpredicts demor aliza-
tion after losses. Psychosom.1997;38:529-534.
24 of 24
|
HOPF et al.
88. GerraG,MontiD,PaneraiAE,etal.Long-termimmune-endocrine
effec ts of bereavement: relationships with anxiety levels and
mood. Psychiatry Res.20 03;121:145-158.
89. Khanfer R , Lord JM, Phill ips AC. Neutr ophil funct ion and cort i-
sol:DHE AS ratio in bereaved older adult s. Brain Behav Immun.
2011; 25:1182-1186 .
90. OngAD,Fuller-RowellTE,BonannoGA ,AlmeidaDM.Spousalloss
predicts alterations in diurnal cortisol activity through prospective
changes in positive emotion. Health Psychol.2011;30:220-227.
91. PfefferCR,AltemusM,HeoM,JiangH.Salivarycortisolandpsy-
chopathologyin adultsbereaved bytheSeptember11,2001ter-
ror attacks. Int J Psychiatry Med.2009;39:215-226.
92. IrwinM,DanielsM,RischSC,BloomE,WeinerH.Plasmacor tisol
and natural killer cell activity during bereavement. Biol Psychiatry.
1988;24:173 -217.
93. Sprat tML,DenneyDR.Immunevariables,depression,andplasma
cortisol over time in suddenly bereaved parents. J Neuropsychiatry
Clin Neurosci.1991;3:299-306 .
94. GoodkinK ,Feaster DJ, Asthana D,etal.Abereavementsuppor t
group intervention is longitudinally associated with salutar y ef-
fects on the CD4 cell count and number of physician visits. Clin
Diagn Lab Immunol.1998;5:382-391.
95. O'Connor MF, Shear MK, Fox R , et al. Catecholamine predic-
tors of complicated grief treatment outcomes. Int J Psychophyiol.
2013;88:349-352.
96. Minton M, Her tzog M , Barron C, et al. The first anniversar y:
stress,well-being,andoptimisminolderwidows.West J Nurs Res.
200 9;31:1035-1056.
97. Andersen JP,Silver RC, Stewart B, et al. Psychological and phys-
iologic al responses following repeated peer death. PLoS One.
2013 ;8:e758 81.
98. CankayaB, Chapman BP,Talbot NL,etal.Histor y of suddenun-
expected loss is asso ciated with elevated interleukin-6 and de-
creasedinsulin-likegrowthfactor-1inwomeninanurbanprimary
care set ting. Psychosom Med.2009;71:914-919.
99. BuiE,Hellberg SN, Hoeppner SS,etal.Circulating levels of oxy-
tocin may be elevated in complic ated grief: a pilot study. Eur J
Psychotraumatol.2019;10:1646603.
100. LaneRD,JacobsSC,MasonJW,etal.Sexdif ferencesinprolac tin
change during mourning. J Psychosom Res.1987;31:375-383.
101. Theorell T, Häggmark C, Eneroth P. Psycho-Endocrinological
ReactionsinFemaleRelativesofCancerPatients:Effectsofanac-
tivation programme. Acta Oncol.1987;26:419- 424.
102. Petitto J M, Quade D, Evans D L. Relatio nship of objec t loss during
developmenttohypothalamic-pituitary-adernalaxisfunctionduring
majoraffective-illnesslaterinlife.Psychiatry Res.1992;44:227-236.
103. Ditzen B, Nater UM, Heim C . Pharmacological stress tests. In:
GellmanMD,TurnerJR, eds. Encyclopedia of Behavioral Medicine.
NewYork,NY:Springer;2013:1468-1471.
104. Adam EK, Quinn ME , Tavernier R, McQuillan MT, Dahlke K A,
Gilber t KE. Diurnal cortisol slopes and ment al and physi-
cal health outcomes: a sys tematic review and meta-analysis.
Psychoneuroendocrinology.2017;83:25-41.
105. SteptoeA, Owen N, Kunz-Ebrecht SR, Br ydon L. Loneliness and
neuroendocrine,cardiovascular,andinflammatorystressresponses
in middle-aged men and women. Psychoneuroendocrinology.
2004;29:593-611.
106. StroebeM,SchutH,StroebeW.Healthoutcomesofbereavement.
Lancet.2 007;370:196 0-1973.
107. Burke LA , Neimeyer RA . Prospective risk factors for compli-
catedgrief.In:StroebeM,SchutH, Bout Jeds.Complicated Grief:
Scientific Foundations for Health Care Professionals.Lon don, New
York:Routledge;2013:145-161.
108. Shear MK. Clinical pr actice. Complicated grief. N Engl J Med.
2015;372:153-160.
109. ValstadM, AlvaresGA ,Egknud M,etal.Thecorrelationbetween
centraland peripheral oxytocin concentrations:a systematic re-
viewandmeta-analysis.Neurosc Biobehav Rev.2017;78:117-124.
110. MacLean EL, Wilson SR, Martin WL, Davis JM, Nazarloo HP,
CarterC S.Challengesformeasuringox ytocin:theblindmenand
the elephant? Psychoneuroendocrinology.2019;107:225-231.
111. Burkett JP,Young L J. The behavioral, anatomical and pharmaco-
logica l parallels be tween social a ttachment , love and addic tion.
Psychopharmacology.2012;224:1-26.
112. O'ConnorMF,WellischDK,StantonAL,EisenbergerNI,IrwinMR,
Lieberman MD. Craving love?Enduringgrief activatesbrain's re-
ward center. NeuroImage.2008;42:969-972.
113. Mc ConnellMH, Killgore WDS, O'Connor MF. Yearningpredicts
subgenual anterior cingulate activity in bereaved individuals.
Heliyon.2018;4:e00 852.
114. CarterSC.Neuroendocrineperspectivesonsocialattachmentand
love. Psychoneuroendocrinology.1998;23:779-818.
115. YoungLJ,WangZ.Theneurobiologyofpairbonding.Nat Neurosci.
2004;7:104 8-105 4.
116. EcksteinM, ZietlowA-L, GerchenMF,et al. The NeMoreal-time
fMRI neu rofeedbac k study: p rotocol of a ra ndomised co ntrolled
clinicalintervention trial in the neural foundationsof mother–in-
fant bonding. BMJ Open.2019;9:e027747.
117. CuthbertBN,InselTR .Towardthefutureofpsychiatricdiagnosis:
the seven pillars of RDoC. BMC Med. 2013 ;11:12 6.
118. Ditze n B, Eckstein M , Fischer M, A guilar-Raab C . Partner schaft
undGesundheit.eingereicht.
119. Frisch J, Aguilar-Raab C , Eckstein M, Dit zen B. Einfluss
von Paarinterak tion auf die Gesundheit . Psychotherapeut.
20 17; 62: 5 9-7 6.
120. SchutM.The dual processmodel ofcopingwith berevement: ra-
tionale and description. Death Studies.199 9;2 3:197-224.
121. McLeanLM,WaltonT,RodinG,etal.Acouple-basedintervention
forpatient sandcaregiversfacingend-stagecancer:outcomesofa
randomized controlled trial. Psychooncology.2013;22:28-38.
122. McClementS,ChochinovHM,HackT,etal.Dignitytherapy:fam-
ily member perspectives. J Palliat Med.2007;10:1076-1082 .
How to cite this article:HopfD,EcksteinM,Aguilar-RaabC,
WarthM,DitzenB.Neuroendocrinemechanismsofgriefand
bereavement:Asystematicreviewandimplicationsfor
future interventions. J Neuroendocrinol. 2020;32:e12887.
https://doi .org /10.1111/j ne.12887
