Acta Pædiatrica ISSN 0803–5253

REGULAR ARTICLE

Shaken baby syndrome and a baby’s cryInga Talvik (inga.talvik@kliinikum.ee)1, Randell C Alexander3,4,5, Tiina Talvik1,2

1.Children’s Clinic of Tartu University Hospital, Tartu, Estonia2.Department of Paediatrics, Tartu University, Tartu, Estonia3.University of Florida-Jacksonville, Jacksonville, Florida, USA4.Florida Children’s Medical Services, Florida, USA5.Morehouse School of Medicine, Atlanta, Georgia, USA

KeywordsCrying, Inflicted traumatic brain injury, trigger forchild abuse

CorrespondenceInga Talvik, Children’s Clinic of Tartu UniversityHospitals, 6 Lunini Street, Tartu, Estonia.Fax: +327319608 |Email: inga.talvik@kliinikum.ee

Received11 January 2008; revised 20 February 2008;accepted 25 February 2008.

DOI:10.1111/j.1651-2227.2008.00778.x

AbstractThe aim of this study was to investigate the relationship between crying of an infant and inflicted

head injury by shaking and/or impact.

During the period between January 1, 1997 and December 31, 2003, 26 cases of shaken baby

syndrome (SBS) were identified in Estonia.

The incidence of SBS was 28.7 per 100 000 children under 1 year of age during the whole study

period. In this group there were four children from twin pairs: two twin boys and a girl from a twin

pair and a boy from another twin pair. This represents 15.4% of the 26 cases. Twins in Estonia

represent 2.12% of infant births.

The mean age on admission was 3.9 months. According to outpatient records almost all parents

(88.5%) in the study group (23/26) had contacted their family physicians and other specialists

because of excessive crying or irritability of the baby prior to the admission to the hospital with SBS or

death. We found that the time curve of crying was similar to the curve of highest incidence of cases

of SBS except the crying curve began earlier.

Conclusion. Our data confirm that the families with twins are at additional risk for SBS and parent’s complaints of

excessive crying of their infants should be taken as signal that parents need to be carefully counselled.

INTRODUCTIONResearch into the demographics and outcomes of shakenbaby syndrome (SBS) has only just begun.

According to Kochanek et al. (1) shaken baby syndrome/inflicted traumatic brain injury (SBS/ITBI) is the leadingcause of death from traumatic brain injury (TBI) in childrenyounger than 2 years of age. The mortality rate for SBS/ITBIvaries from 13% to 30% (2–8). Specific evidence of previousintracranial injuries (old intracranial haematomas from dif-ferent ages) from shaking episodes is found in about 33% to40% of all cases (9–11). Victims are often under 1 year ofage (2), but children up to 5 years of age may be vulnerablebecause of their smaller size (12). There are case reports ofan 8-year-old victim (13) and an adult (14).

Between 1 in 3000 to 1 in 4000 children younger than1 year of age are victims of significant or fatal inflictedchildhood neurotrauma (ICNT) making SBS/ITBI almostas common as cystic fibrosis and more common than acutelymphocytic leukaemia (1,15). For every child who dies,there are approximately three who survive – usually withmoderate-to-severe deficits (12). There may be a much largernumber of mild cases that are never diagnosed. The esti-mated number of children who are shaken may be over 100times higher than the cases that rise to the level of a formaldiagnosis of SBS (4,16).

In SBS/ITBI it is widely held that brain injury is caused byviolent shaking resulting from caretakers’ tension and frus-

tration generated by crying and irritability (17,18). ‘The actof shaking leading to SBS is so violent that individuals ob-serving it would recognize it as dangerous and likely to killthe child (12). Although it has long been recognized thatsome infants cry more than others, the cause (or causes)remains elusive. Infants who present with persistent cryingare probably a heterogeneous group, requiring a variety ofmanagement approaches (19). Although considered a trivialproblem by many health care professionals, persistent cry-ing in babies has been associated with maternal depression,family stress, family breakdown and child abuse (20,21). Theprovocative effect of infant crying on abuse occurs more of-ten in the first 6 months of life (15,22). The results of a studyby Reijneveld et al. (23) provided a link between data on in-fant crying and infant maltreatment, supported by case stud-ies on fatal abuse that seemed to be induced by infant crying.The actual duration of crying at a given moment seems tobe less relevant than the parents’ perception regarding thecrying of their infant over the long term (24).

In a series of 591 cases, Lee et al. (25) found that cryingwas the reported stimulus for SBS in 166 (28%) of the cases.They compared prior data about the relationship between in-cidence of crying by age with their incidence of SBS by age.The curves were similar except for a ‘phase shift’ – wherebythe peak of crying was about 6 weeks earlier than the peakof SBS/ITBI. However, their cases were based upon mediareports. Thus the diagnosis of SBS itself was anecdotal, re-ports of crying may have been inaccurate and the selection of

782 C©2008 The Author(s)/Journal Compilation C©2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 782–785

Talvik et al. Shaken baby syndrome and crying

cases biased in that only cases that came to media attentionwere included.

We report a series of consecutive cases for an entire coun-try in which the diagnosis of SBS/ITBI was analyzed ret-rospectively and prospectively, by examination of medicalfindings. Information about crying and the child’s later de-velopment was also obtained for each case.

PATIENT AND METHODSAll head injury cases that were diagnosed in two tertiary cen-tres for children in Estonia (Children’s Clinic of Tartu Uni-versity Hospital and Tallinn Children’s Hospital) between1997 to 1999 were retrospectively reviewed to determinewhether they met study criteria for SBS (brain injury symp-toms, intracranial haemorrhage, retinal haemorrhage). From1997 to 2003, all head injuries in Estonia were reviewed bythe same Paediatric Neurologist (IT) and those that met thecriteria for SBS/ITBI were included.

Whether retrospective or prospective, additional medi-cal information from outpatient records and family sourceswas obtained. The Bioethical Committee of the University ofTartu approved the study. The parents gave informed con-sent for participation in the study.

RESULTSDuring the period from January 1, 1997 to December 31,2003, 26 cases of SBS/ITBI were identified in Estonia(2). There were 5/26 children in the retrospective studygroup (1997–1999) and 21/26 in the prospective study group(2000–2003). All of the children were younger than 1 yearof age. The population at risk during the study period usedin calculation of incidence was 89 002 (26). The incidence ofSBS/ITBI was 28.7 (95%CI: 18.72– 40.68) per 100 000 chil-dren under 1 year of age during the whole study period. Inci-dence rate was 40.5 (95%CI: 25.74–61.15) in the prospectivestudy and 13.5 (95%CI: 4.36–27.52) per 100 000 childrenunder 1 year of age in the retrospective study period. Theincidence rate was 3.5 times higher in the prospective study(p = 0.018).

0

20

40

60

80

100

120

140

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Age in Weeks

Min

ute

s c

ryin

g p

er

da

y

0

1

2

3

4

5

6

7

8

9

Nu

mb

er

of

SB

S c

hil

dre

n

Average crying

SBS curve

Figure 1 Infant crying curve (modified from http://en.citizendium.org/wiki/Infant colic) and age of babies with SBS/ITBI in Estonia.

Among 26 patients there were 2 twin boys, one girl froma twin pair and one boy from another twin pair. These fourrepresented 15.4% of the study population compared withthe prevalence of twins in Estonia of 2.12% during the sametime period. (p = 0.002; 95%CL: 2.1–24.8).

The mean age on admission was 3.9 months (95%CI: 2.7–5.2). Boys were admitted at an earlier age, with the meanages at admission for boys and girls being 3.4 (95%CI: 2.2–4.6) and 5.8 (95%CI: 1.5–10.2) months, respectively, thoughthis difference was not statistically significant (p = 0.2275).

The relationship between the crying curve for in-fants (modified from http://en.citizendium.org/wiki/Infantcolic) and the age on admission in weeks of the infants inthis study are demonstrated in Figure 1. Minutes of cry-ing per day are highest from 4 to 10–12 weeks of age. Themean age in the study group at admission was 3.4 months(∼14 weeks) for boys and 5.8 months (∼24 weeks) for girls.Note that there is a good correlation of the crying curve withthe time of SBS/ITBI admission, although a time lag exists.

All children had serious intracranial injuries on admission:subdural haematomas (SDH) were confirmed in 20/26 cases(76.9), subarachnoid haemorrhages in 3/26 cases (11.5%),intracerebral haemorrhages in 2/26 cases (7.7%) and epidu-ral haemorrhage in one case (3.9%). In 9/20 cases theSDH were chronic and in 8/20 cases bilateral. In 9/26cases(34.6%) there were signs of repeated injuries: SDH ofdifferent ages, old fractures (prior extracranial abuse).

According to outpatient records almost all parents(88.5%) in the study group (23/26) had contacted their fam-ily physicians and other specialists because of excessive cry-ing or irritability of the baby prior to the admission to thehospital with SBS or death.

DISCUSSIONThe correlation of crying and age has been studied by Brazel-ton (27) and others (28,29). The association between cryingand SBS/ITBI has also been suspected for decades (17,18).

From an evolutionary perspective, crying is attachmentbehaviour, promoting proximity to the infant’s primary

C©2008 The Author(s)/Journal Compilation C©2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 782–785 783

Shaken baby syndrome and crying Talvik et al.

caregiver and ensuring survival and the development of so-cial bonds (30). Crying begins in the first few weeks of life,and typically the duration peaks at 2.4 hours per day at theage of 6 weeks (27). Episodes of crying tend to cluster in theevening, but can be throughout the day (31). For most ba-bies, crying and irritability decrease substantially by the ageof 3–4 months (31,32).

Lee et al. (25) contrasted the incidence curve of crying inthe first months of life with the incidence curve of SBS/ITBI.While they found the two incidence curves to be similar inshape with a phase shift of several weeks, the reliance onmedia reports necessarily introduced uncertainties about thedetermination as to whether the cases were truly SBS/ITBI,whether the ages accurately represented the actual distri-bution of ages of SBS/ITBI victims and whether the mediareports accurately portrayed the concerns about crying inthis population. In contrast, this study (Fig. 1) relied uponreview of the actual medical reports to determine the va-lidity of the diagnosis of SBS/ITBI, compared both retro-spective and prospective detection of cases, viewed an entirepopulation of cases and obtained more detailed informationabout development and crying. Interestingly, the results ofthe study by Lee et al. (25) and this one are very similar inshowing an apparent relationship between shaking and cry-ing incidences. The shape and peak of the SBS/ITBI curvesare very similar between these data arrived at in somewhatdifferent ways, and between two different countries. Coupledwith long-standing clinical observations (12), these studiesappear to support the theory of crying as a trigger for infantSBS/ITBI cases.

Our data show that a vast majority of the parents com-plained to their doctors about excessive crying by the child.Lee et al. (25) reported only 28% of cases were confirmedto be related to infant crying via media reports; our datashowed that 88.5% of parents had reported the child beingexcessively irritated or crying too much. Unfortunately nei-ther effective advice nor help was provided to these families,likely explained by underestimation of the situation by thedoctors.

The outcome for children with SBS/ITBI is usually poor.The results of our prospective population-based outcomestudy confirmed that SBS/ITBI is a serious medical and so-cial problem (33). Keenan et al. (34) stressed that youngchildren with SBS/ITBI have a much worse outcome thanthe older ones. Longer follow-up periods were reported byKarandikar et al. (35), Bonnier et al. (36) and Barlow et al.(37) with very different outcome results: the number of ap-parently healthy children being variable from 4% to 55%.Our study demonstrated that 91% of survivors had signifi-cant problems (33). Because actual school performance wasnot measured at later ages, it is possible that even higherrates of disability exist. This study differs from other studiesreviewed by Makaroff and Putnam (38), where 22% of pa-tients had no impairments. Seventy-eight percent of the sur-viving children with SBS/ITBI in their study were left withimpairments in their motor and cognitive abilities, language,vision and behaviour (38). Despite being in a more or lessstable clinical condition immediately after the abuse (10/26

children had the GCS = >13), children with SBS/ITBI ex-perience considerable developmental problems later (2). Itshould also be noted that 34.6% of the children in our studyhave signs of repeated intracranial injuries, which is similarto the numbers described before (9,11).

The time lag between the peak of the infant crying andSBS/ITBI admissions is important. This was shown by Barret al. (39) where the difference between the peak cryingcurve (5–6 weeks) and the curve of cases of SBS/ITBI (10–13 weeks) showed a lag. This time lag gives medical profes-sionals time to counsel with and intervene with families inneed. Such counselling should take into account the fact thatthe actual duration of crying is less relevant than the care-giver’s perception regarding the crying of their infant overthe long term (24). Our data also confirm prior research thatthe families with twins are at additional risk (40). Medicalprofessionals need to be more determined to help the care-givers with complaints of their infant’s crying, especially inthe case of twins. Prevention is the only way of reducing thenumber of children with abusive head injury.

ACKNOWLEDGEMENTThe study was supported by the GARLA 6627 and TARLA2695.

References

1. Kochanek PM, Berger RP, Margulies SS, Jenkins LW. Inflictedchildhood neurotrauma: New insight into the detection,pathobiology, prevention, and treatment of our youngestpatients with traumatic brain injury. J Neurotrauma 2007 Jan;24: 1–4.

2. Talvik I, Metsvaht T, Leito K, Poder H, Kool P, Vali M, et al.Inflicted traumatic brain injury (ITBI) or shaken babysyndrome (SBS) in Estonia. Acta Paediatrica 2006; 95:799–804.

3. King WJ, MacKay M, Sirnick A, Canadian Shaken Baby StudyGroup. Shaken baby syndrome in Canada: clinicalcharacteristics and outcomes of hospital cases. CMAJ 2003;168: 155–9.

4. Keenan HT, Runyan DK, Marshall SW, Nocera MA, MertenDF, Sinal SH. A population based study of inflicted traumaticbrain injury in young children. JAMA 2003; 290: 621–6.

5. Dias MS, Backstrom J, Falk M, Li V. Serial radiography in theinfant shaken impact syndrome. Pediatr Neurosurg 1998; 29:77–85.

6. Hadley MN, Sonntag VKH, Rekate HL, Murphy A. The infantwhiplash-shake injury syndrome. A clinical and pathologicalstudy. Neurosurgery 1989; 24: 536–40.

7. Sinal SH, Ball MR. Head trauma due to child abuse: serialcomputerized tomography in diagnoses and management.South Med J 1987; 80: 1505–12.

8. Zimmerman RA, Bilaniuk LT, Bruce D, Schut L, Uzzell B,Goldberg HI. Computed tomography of craniocerebral injuryin the abused child. Radiology 1979; 130: 687–90.

9. Alexander R, Crabbe L, Sato Y, Smith W, Bennett T. Serialabuse in children who are shaken. AJDC 1990; 144: 58–60.

10. Jenny C, Hymel KP, Ritzen A, Reinert SE, Hay TC. Analysis ofmissed cases of abusive head trauma. Pediatrics 1997; 99:371–5.

11. Ewing-Cobbs L, Kramer L, Prasad M, Canales DN, Louis PT,Fletcher JM, et al. Neuroimaging, physical, and developmental

784 C©2008 The Author(s)/Journal Compilation C©2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 782–785

Talvik et al. Shaken baby syndrome and crying

findings after inflicted and noninflicted traumatic brain injuryin young children. Pediatrics 1998; 102: 300–7.

12. American Academy of Pediatrics Committee on Child Abuseand Neglect. Shaken baby syndrome: rotational cranialinjuries—Technical report. Pediatrics 2001; 108: 206–10.

13. Mierisch RF, Frasier LD, Braddock SR, Giangiacomo J,Berkenbosch JW. Retinal hemorrhages in an 8-year-old child:an uncommon presentation of abusive injury. PediatricEmergency Care 2004; 20: 118–20.

14. Pounder DJ. Shaken adult syndrome. Am J Forensic MedPathol 1997 Dec; 18: 321–4.

15. Barlow KM, Minns RA. Annual incidence of shaken impactsyndrome in young children. Lancet 2000 Nov; 356: 1571–2.

16. Theodore AD, Chang JJ, Runyan DK, Hunter WM,Bangdiwala SI, Agans R. Epidemiologic features of thephysical and sexual maltreatment of children in the Carolinas.Pediatrics 2005 Mar; 115: e331–7.

17. Dykes LJ. The whiplash shaken infant syndrome: what hasbeen learned? Child Abuse Negl 1986; 10: 211–21.

18. Levitt C, Smith WL, Alexander R. Abusive head trauma.In: Reece R editor. Child abuse: medical diagnosis andmanagement. Philadelphia: Lea & Febiger, 1994.

19. Hiscock H, Jordan B. Problem crying in infancy. MJA 2004;181: 507–12.

20. Barr RG, Kramer M, Pless IB, Boisjoly C, Leduc D. Feedingand temperament as determinants of early infant crying/fussing behavior. Pediatrics 1989; 84: 514–21.

21. McMahon C, Barnett B, Kowalenko N, Tennant C, Don N.Postnatal depression, anxiety and unsettled infant behaviour.Aust N Z J Psychiatry 2001; 35: 581–88.

22. Agran PF, Anderson C, Winn D, Trent R, Walton-Haynes L,Thayer S. Rates of pediatric injuries by 3-month intervals forchildren 0 to 3 years of age. Pediatrics 2003 Jun; 111(6 Pt. 1):e683–92.

23. Reijneveld SA, Brugman E, Hirasing RA. Excessive infantcrying: the impact of varying definitions. Pediatrics 2001 Oct;108: 893–7.

24. Reijneveld SA, Van Der Wal MF, Brugman E, Sing RA,Verloove-Vanhorick SP. Infant crying and abuse. Lancet 2004Oct; 364: 1340–2.

25. Lee C, Barr RG, Catherine N, Wicks A. Age-related incidenceof publicly reported shaken baby syndrome cases: is crying atrigger for shaking? J Dev Behav Pediatr 2007 Aug; 28:288–93.

26. Statistics of Estonia www.stat.ee.27. Brazelton TB. Crying in infancy. Pediatrics 1962; 29:

579–88.28. Barr RG. The normal Crying curve: what do we really know?

Deve Med Child Neurol 1990; 32: 356–62.29. Barr RG, Paterson JA, MacMartin LM, Lehtonen L,

Young SN. Prolonged and unsoothable crying bouts ininfants with and without colic. J Dev Behav Pediatri 2005; 26:14–23.

30. Bowlby J. The nature of the child’s tie to his mother. Int JPsychoanal 1958; 39: 350–73.

31. St James-Roberts I, Halil T. Infant crying patterns in the firstyear: normal community and clinical findings. J Child PsycholPsychiatry 1991; 32: 951–68.

32. Barr RG. Crying in the first year of life: good news in the midstof distress. Child Care Health Dev 1998; 24: 425–39.

33. Talvik I, Mannamaa M, Juri P, Leito K, Poder H, Kool P, et al.Outcome of infants with inflicted traumatic brain injury(shaken baby syndrome) in Estonia. Acta Paediatr 2007; 96:1164–68.

34. Keenan HT, Runyan DK, Nocera M. Child outcomes andfamily characteristics 1 year after severe inflicted ornoninflicted traumatic brain injury. Pediatrics 2006; 117:317–24.

35. Karandikar S, Coles L, Jayawant S, Kemp AM. Theneurodevelopmental outcome in infants who have sustained asubdural haemorrhage from non-accidental head injury. ChildAbuse Rev 2004; 13: 178–87.

36. Bonnier C, Nassogne M-C, Saint-Martin C, Mesples B,Kadhim H, Sebire G. Neuroimaging of intraparenchymallesions predicts outcome in shaking baby syndrome. Pediatrics2003; 112: 808–14.

37. Barlow KM, Thomson E, Johnson D, Minns RA. Lateneurologic and cognitive sequelae of inflicted traumatic braininjury in infancy. Pediatrics 2005; 116: e174–85.

38. Makaroff KL, Putnam F. Outcomes of infants and childrenwith inflicted traumatic brain injury. Dev Med Child Neurol2003, 45: 497–502.

39. Barr RG, Trent RB, Cross J. Age-related incidence curve ofhospitalized Shaken Baby Syndrome cases: Convergentevidence for crying as a trigger to shaking. Child Abuse &Neglect 2006; 30: 7–16.

40. Alexander RC, Smith WL. Investigating abuse in asymptomatictwin. Arch Pediatr Adolec Med 1996; 150: 444–45.

C©2008 The Author(s)/Journal Compilation C©2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 782–785 785