Joby Rowe has spent almost 5 years in jail for supposedly shaking to death his daughter Alanah. Alanah was found to have encephalopathy as well as subdural and retinal haemorrhages, which comprise the “triad” of findings associated with Shaken Baby Syndrome.
I have blogged about this miscarriage of justice and published a peer reviewed article[1] in the Australian Journal of Forensic Science that questioned how this “triad only” case[2] could be prosecuted in Australia, given that there is no scientific basis to the notion that the triad can be used to infer shaking or abuse. Despite acknowledging that the article was refereed by people “of high standing in their respective disciplines”, following “the correct editorial processes”, this peer reviewed article was subsequently unethically retracted by the journal, after the editor succumbed to pressure from senior members of the forensic medicine community, and after legal threats were made against the journal by the experts whose work was scrutinised in the article[3]. One of the experts further notified the police of the article, who illegally used their database to access my personal information.
During the trial there was no evidence that positively indicated that Joby Rowe had shaken Alanah. Instead, the experts used shaking as the presumptive cause and then used a process of “diagnosis by exclusion”: shaking must have occurred because they were unable to find a medical cause for the triad. In other words, they claim to have discounted the wide range of medical causes of the “triad”, many of which are rare and not well researched. Their assumption also discounts the possibility that new causes of these findings will be found, despite numerous new causes being identified in the past decade. The former head of the Victorian Institute of Forensic Medicine, Professor Stephen Cordner, warned back in 2008[4] that “A category we have to be careful of is the diagnosis – or coming to conclusions of any sort – by exclusion”, likening its use in modern criminal cases to its use in 16th century witch hunts.
This “diagnosis by exclusion” also inverts one of the most fundamental principles of our justice system: the presumption of innocence. Shaking (or abuse) is considered the presumptive cause, unless proven otherwise.
It should not be the responsibility of the accused to determine the cause of death in these cases where no evidence of abuse exists. But far too often it is, because “diagnosis by exclusion” presumes that abuse has occurred, unless proven otherwise. That is the reality that Joby Rowe continues to face, and it is for this reason that, along with a colleague who is also interested in these wrongful convictions, I have continued to research the medical files from the case. What we have identified is not only a possible cause that the experts failed to exclude, it is a cause that is better supported by the “constellation” of symptoms and medical findings in the case.
Alanah was born 6 weeks premature after a pregnancy that was complicated by poor intrauterine growth. She did not breath for 57 seconds after being born, and was resuscitated by CPR. She subsequently spent a couple of weeks in Special Care and then remained in hospital for another couple of weeks before going home. She further had an umbilical hernia. Prior to her collapse at just 12 weeks of age, Rowe stated that after he fed her a bottle, she started having trouble breathing. As she collapsed, Alanah was taking big gasping breaths before she stopped breathing. Blood was coming from her nose.
Subsequent findings were:
-cardiac arrest
-hypoxic-ischaemic encephalopathy (HIE, the ultimate cause of death)
-profuse bleeding from the nose
-blood stained fluid in mouth
-extensive retinal haemorrhages
-acute subdural and subarachnoid haemorrhages
-previous subdural haemorrhages (age and cause unexplained)
-blood in the space between the dura and arachnoid mater that surrounds the spinal cord, i.e. possible subdural spinal haemorrhages or possibly blood trickling down from the brain
-lactic acidosis
-prolonged APTT and INR, and high D-Dimer, resulting in diagnosis of Disseminated Intravenous Coagulation (DIC)
-low values of coagulation factors II,V,VII,IX,X,XI
-no skull or spinal fracture
-no ligamentous injury to the neck
-no external sign of trauma such as bruising
Alanah´s body temperature was cooled to 31 degrees in an attempt to minimise the effects of Hypoxia on the brain.
The medical report of the child abuse paediatrician, the central expert who testified for the prosecution, concluded that “in summary Alanah had the following abnormal findings:
1. Subdural and subarachnoid brain haemorrhages,
2. Subdural spinal haemorrhages.
3. Extensive widespread retinal haemorrhages.
4. A severe hypoxic-ischaemic brain injury.”
These summarised injuries are the classic signs of the hypothesised Shaken Baby Syndrome. The expert had cherry picked the subset of findings that fit the SBS narrative in her summary of findings.
The prosecutor then stated that “each of subdural haemorrhages in isolation, retinal haemorrhages in isolation and HIE in isolation could have arisen by alternative or non-traumatic causes. That's a concession we make at the outset. However, it's the constellation, the grouping of those three key clinical findings, which I expect each of the experts will identify as providing the bases for their opinion as to the mechanism which brought about these injuries.. together they are the smoking gun… that shaking occurred to cause these injuries in Alanah.”
Was the full constellation of injuries really considered? Other findings, such as DIC, low coagulation factors, liver and blood chemistry abnormalities, the previous subdural bleeding, the bleeding from the nose, were then either ignored or assumed to be secondary to trauma. The medical report stated that “Alanah's blood chemistry results, blood gas results and liver function results showed some abnormalities. In the setting of critical illness and following a cardiac arrest these patterns of abnormalities are commonly seen and in Alanah's case these abnormal results are almost certainly secondary to her cardiac arrest and resuscitation” and that the “coagulation factor levels were probably low because the factors had been consumed during the process of disseminated intravascular coagulation.”
Let us look at disseminated intravascular coagulation (DIC) in particular. DIC is “a blood clotting disorder that can turn into uncontrollable bleeding”[5]. This includes intracranial bleeding[6],[7], bleeding of the retina[8] and “external bleeding… from the mucosa”[9]. It is well established that coagulation problems can cause the triad of findings that are associated with SBS.[10],[11] So the findings of subdural haemorrhages, extensive retinal haemorrhages, and blood around the spinal cord, could all be secondary to DIC.
So in this case one needs to determine the cause of the DIC.
The central prosecution expert concluded that the DIC was secondary to trauma. This seems to be based on the fact that trauma was established by the presence of a particular manifestation of the “triad” (including extensive retinal haemorrhages). But these haemorrhagic aspects of the triad may be secondary to DIC whatever the cause, whilst hypoxic-ischaemic encephalopathy (HIE) is caused by a lack of oxygen to the brain, not by trauma. The expert´s reasoning was circular: she presumed trauma because of the existence of a particular manifestation of the triad, including extensive retinal haemorrhages, and from there it followed that DIC must be secondary to the trauma, and therefore the haemorrhages were caused by trauma rather than the DIC. But what if something else, not trauma, caused the DIC, and then the DIC caused the haemorrhages? In that case the haemorrhages could not be used to infer trauma.
The many causes of DIC include complications of obstetrics, asphyxia, infection or sepsis, respiratory distress syndrome, acidosis, cancer, and “shock due to any condition that causes ischaemic tissue injury and exposure or release of tissue factor.”[12] Trauma is a relatively uncommon cause of DIC.[13],[14]
It may not ever be possible to know what caused the DIC in Alanah, because it is a non-specific condition that can follow from a range of other conditions. Nevertheless, we propose two possible causes of DIC in this case, noting that many other possibilities exist.
Congenital heart disease is the most common type of congenital defect, and accounts for more deaths in the first year of life than any other condition not involving infection[[15]]. Some heart defects do not cause noticeable symptoms for several months or even years, making it hard to diagnose in infants. If a congenital heart condition caused Alanah´s cardiac arrest, this could have led to systemic inflammatory response syndrome (SIRS). DIC is a frequent complication of SIRS[16]. So a heart condition can account for hypoxic-ischaemic encephalopathy by restricting oxygen supply to the brain, whilst the onset of DIC would then result in subdural and retinal haemorrhages, as well as the bleeding from the nose, and the low coagulation factors. Heart failure can also explain the lactic acidosis[17] and is consistent with the lack of external signs of trauma.
In the medical report, it states that “The aetiology of the diffuse cerebral oedema, subdural blood and subarachnoid blood is most likely related to trauma. A diffuse hypoxic injury such as with non-trauma related cardiac arrest could result in the cerebral parenchymal changes, but does not account for the subdural haemorrhages identified.” Yet cardiac arrest that elicits SIRS, and leads to DIC, does account for the subdural haemorrhages. It also accounts for the extensive retinal haemorrhages and all other findings in the case.
In fact, Joby Rowe´s brother died at 3 months of a congenital heart defect. No test was made to see if Alanah had a genetic congenital heart condition.
Birth asphyxia is another well recognised cause of DIC[18]. Whilst acute (also referred to as “overt phase”[19]) DIC develops “over hours or days”[20], DIC can pass through a “latent and compensated”[21] phase which “evolves slowly, over weeks or months”.[22] DIC in this phase can cause “venous thrombotic and embolic manifestations”.[23] What if Alanah had DIC since her well documented birth asphyxia? The latent, compensated DIC may have progressed after several weeks to the overt, bleeding phase of the DIC[24]. This can in turn lead to the haemorrhages, to the lack of oxygen supply to the brain and HIE, and to cardiac arrest.
If DIC was triggered during Alanah´s birth asphyxia and evolved over Alanah´s 12 week lifespan, then a range of event sequences and combinations of conditions are possible that all could have led to the final collapse, and can explain all the findings in the case. For example, in one study[25] of 54 patients with sepsis related DIC, 6 had congenital cardiac disease whilst 3 had dilated cardiomyopathy.
So whilst is true that trauma can cause DIC, there was no outward sign of trauma. DIC is not listed as a symptom or outcome of shaking in any literature, and there is no scientific evidence that shaking can cause DIC. It should be enough for Joby Rowe that non traumatic causes of DIC cannot be excluded. But we can go further than this. The full constellation of findings in the case, including findings disregarded by the “experts”, show that it is shaking that can be excluded.
Let us look at the significance of the blood nose. As she collapsed, Alanah “had blood pouring out of her nose”[26], and was “losing a lot of blood out of her nose”.[27] Indeed the call to which the paramedics were responding was coded as signal 1, an emergency, with “a child with ineffective breathing and a blood nose”. So the blood nose was present at the very start.
Nowhere in the long list of findings that are supposedly[28] associated with shaken baby syndrome, including irritability, poor feeding, vomiting, lethargy, respiratory compromise, seizures, apnea and delayed development[29], is any mention of bleeding from the nose. However, DIC is known to cause bleeding from mucosa, including the nose.[30]
Even acute DIC takes “hours to days” to evolve. Yet Alanah was bleeding profusely from the nose during her collapse. The bleeding nose, a finding that was totally ignored by the prosecution “experts”, is the smoking gun that DIC was involved in the collapse, and that whatever caused the DIC had been evolving for “hours” at the very least. Joby had not shaken Alanah in the moments prior to her collapse, as alleged by the experts. This was a medical collapse. The intra-cranial and retinal haemorrhages followed from the DIC. Not the other way around.
Blood stains in regurgitated baby formula that was found in the mouth of Alanah[31] provide further corroborating evidence of this scenario, as they are indicative of bleeding in the mouth, stomach and/or throat, all of which are associated with DIC[32].
Explanations for Alanah´s collapse where medical conditions were developing and evolving over days or weeks, can also explain some of the other behaviour of Alanah in the days leading up to her collapse. Alanah´s high pitched, persistent screams, and difficulty settling were likely caused by the pain[33] she was suffering from the previous subdural haemorrhages that were found. Yes, I am aware that all infants cry, but in the days leading up to her collapse, Alanah´s crying had “escalated to screaming”[34] that was “uncontrollable” [35] and “unrelenting”[36]. The assertion that Joby Rowe shook Alana moments prior to the collapse cannot explain this behaviour of Alanah, nor the previous bleeding in her brain. These findings were simply disregarded by the “experts”, and not included in the “constellation” of findings upon which their conclusions were based.
The medical report stated that “Alanah was a previously well 12 week-old baby” prior to collapse, with no regard taken of her prematurity which is a known risk factor in conditions associated with SBS[37], no regard taken of her birth asphyxia, no regard taken of her extended stay in Special Care, no regard taken of the previous episodes of bleeding on her brain, no regard taken of her unrelenting screaming over the previous days. None of these findings fit the narrative of shaking, which was created by only considering a compatible subset of findings rather than the full constellation.
A case study[38] exists in the literature where DIC was found to have caused the “triad” including “extensive intraretinal hemorrhages from the posterior pole to the ora serrata” that were described as “severe”. The autopsy determined the cause of death to be a “myocardial infarct in the distribution of an anomalous coronary artery”. DIC, diagnosed on admission, was found to be secondary to cardiogenic or hypotensive shock. The medical examiner had considered concerns for abuse raised by the Child Protection team, based on the existence of the triad, but after considering the autopsy findings, laboratory results, and investigation of the circumstances preceding the death, ruled the death to be of natural causes.
Joby Rowe is an innocent man incarcerated with the label “child killer”, who should not have to spend one more day in jail. His conviction was based on the erroneous medical testimony of three expert witnesses who testified for the prosecution. What is left to be seen is whether any of these experts have the intellectual and moral fortitude to inform the Court of the errors they made, as is their obligation under the Practice Direction for Expert Evidence in Criminal Trials.[39]
[1] Brook, C. Is there an evidentiary basis for shaken baby syndrome? The conviction of Joby Rowe Aust. J. Forensic Sci., 53 (2019), p. 1 [2] see J. Tully Oral Presentation “Protego Maxima” at Don't Forget the Bubbles August (2018) [3] Brook C., Lynøe N., Eriksson A., Balding D.,Retraction of a peer reviewed article suggests ongoing problems with Australian forensic science, Forensic Science International: Synergy, Volume 3, 2021 [4] Cordner, S. R v Klamo: an example of miscommunication and misunderstanding of expert evidence where the conviction was overturned, Australian Journal of Forensic Sciences 2012, 44:4, 323-331 [5] Cleveland Clinic [6] Monagle, P. & Andrew, M., Acquired Disorders of Hemostasis, in Nathan And Oski’s Hematology Of Infancy And Childhood 1631 (Stuart H. Orkin et al. eds., 6th ed., 2003). [7] Levi, M & Ten Cate, H Disseminated Intravascular Coagulation, 341 NEW ENGLAND J. OF MEDICINE..586, 586 (1999) [8] Levinson JD, Pasquale MA, Lambert SR. Diffuse bilateral retinal hemorrhages in an infant with a coagulopathy and prolonged cardiopulmonary resuscitation. J AAPOS. 2016;20(2):166-168. [9] What are the Signs and Symptoms of Disseminated Intravascular Coagulation? Haemotology-Oncology Associated of CNY [10] Narang, S A Daubert Analysis of Abusive Head Trauma/Shaken Baby Syndrome, 11 HOUS. J. HEALTH L. & POL’Y 505, 571 (2011) [11] Christian CW, Levin AV, AAP Council On Child Abuse And Neglect, AAP Section On Ophthalmology, American Association Of Certified Orthoptists, American Association For Pediatric Ophthalmology And Strabismus, American Academy Of Ophthalmology. The Eye Exam ination in the Evaluation of Child Abuse. Pediatrics. 2018 [12] Moake, J. L. MSD Manual Disseminated Intravascular Coagulation (DIC), 2021 [13] Moake, J. L. MSD Manual Disseminated Intravascular Coagulation (DIC), 2021 [14] H. Ören, I. Cingöz, M. Duman, S. Yılmaz & G. Irken (2005) Disseminated Intravascular Coagulation In Pediatric Patients: Clinical and Laboratory Features and Prognostic Factors Influencing the Survival, Pediatric Hematology and Oncology, 22:8, 679-688 [15] Lopes, Selma Alves Valente do Amaral et al. “Mortality for Critical Congenital Heart Diseases and Associated Risk Factors in Newborns. A Cohort Study.” Arquivos brasileiros de cardiologia vol. 111,5 (2018): 666-673, doi:10.5935/abc.20180175 [16] Gando S, Kameue T, Nanzaki S, Nakanishi Y. Disseminated intravascular coagulation is a frequent complication of systemic inflammatory response syndrome. Thromb Haemost. 1996;75:224–8 [17] Whelan, C., reviewed by Biggers, A. MD Lactic Acidosis: What You Need to Know, Healthline Nov. 1 2018 [18] Rajagopal, R Thachil, J & Monagle, P Disseminated intravascular coagulation in paediatrics, Arch Dis Child 2016, doi:10.1136/archdischild-2016-311053 [19] Papageorgiou C, Jourdi G, Adjambri E, et al. Disseminated Intravascular Coagulation: An Update on Pathogenesis, Diagnosis, and Therapeutic Strategies. Clin Appl Thromb Hemost. 2018;24(9_suppl):8S-28S. [20] What are the Signs and Symptoms of Disseminated Intravascular Coagulation? Haemotology-Oncology Associated of CNY [21] Papageorgiou C, Jourdi G, Adjambri E, et al. Disseminated Intravascular Coagulation: An Update on Pathogenesis, Diagnosis, and Therapeutic Strategies. Clin Appl Thromb Hemost. 2018;24(9_suppl):8S-28S. [22] What are the Signs and Symptoms of Disseminated Intravascular Coagulation? Haemotology-Oncology Associated of CNY [23] What are the Signs and Symptoms of Disseminated Intravascular Coagulation? Haemotology-Oncology Associated of CNY [24] Papageorgiou C, Jourdi G, Adjambri E, et al. Disseminated Intravascular Coagulation: An Update on Pathogenesis, Diagnosis, and Therapeutic Strategies. Clin Appl Thromb Hemost. 2018;24(9_suppl):8S-28S. [25] H. Ören, I. Cingöz, M. Duman, S. Yılmaz & G. Irken (2005) Disseminated Intravascular Coagulation In Pediatric Patients: Clinical and Laboratory Features and Prognostic Factors Influencing the Survival, Pediatric Hematology and Oncology, 22:8, 679-688 [26] R v Rowe trial transcripts @p242 [27] R v Rowe trial transcripts @p242 [28] There is no evidence base for the association of these findings with shaking. [29] e.g. Tully, J. Abusive head trauma – mechanisms, myths and mysteries , VFPMS seminar 2019 [30] e.g. What are the Signs and Symptoms of Disseminated Intravascular Coagulation? Haemotology-Oncology Associated of CNY [31] Statement of Michael Holden, ALS paramedic team manager, @p3 dated 03/09/2015 [32] What are the Signs and Symptoms of Disseminated Intravascular Coagulation? Haemotology-Oncology Associated of CNY [33] e.g. The Cleveland Clinic Subdural Hematoma [34] R v Rowe trial transcripts @p787 [35] R v Rowe trial transcripts @p787 [36] R v Rowe trial transcripts @p787 [37] e.g. Squier W, Mack J, Jansen AC. Infants dying suddenly and unexpectedly share demographic features with infants who die with retinal and dural bleeding: a review of neural mechanisms. Dev Med Child Neurol 2016;58(12):1223–34. [38] Levinson JD, Pasquale MA, Lambert SR. Diffuse bilateral retinal hemorrhages in an infant with a coagulopathy and prolonged cardiopulmonary resuscitation. J AAPOS. 2016;20(2):166-168. doi:10.1016/j.jaapos.2015.11.003 [39] see section 4.3 of the Practice Direction for Expert Evidence in Criminal Trials