Novel Applications and Refinements of Current Imaging Techniques in Perinatal and Childhood Death Investigation

Background:
It is estimated that 1 in 80 pregnancies in the UK will result in a termination or stillbirth due to an abnormality of the baby. An autopsy can provide information on the cause of death and help change management in future pregnancies (1), however many parents refuse this examination as they feel their child has suffered through enough. This means they miss out on valuable information about why their child died and whether future children will suffer a similar fate.

'Less invasive' ways for investigating the cause of death using imaging have been pioneered at our institution and shown that by using magnetic resonance imaging (MRI) together with 'key-hole' surgery methods to obtain tissue samples, we get the same answer as autopsy in >90% (2) cases. This method is however expensive and unfortunately not widely available. We believe we can improve on this technique by using ultrasound imaging with needle biopsies which makes fewer and smaller marks to the skin and would be easier to roll-out nationally if it can achieve a similar level of accuracy as a full autopsy.

Aims:
This study will test how good ultrasound guided needle tissue biopsies are compared to autopsy. We will also investigate how good different imaging tests are when used alone and in combination with tissue sampling

Methods:
Parents who consent for childhood autopsy will be offered a series of imaging tests and less invasive needle biopsies. We will then compare these results with our current autopsy results and test for any differences in findings.

Applications and benefits:
Parents tell us that they want the 'minimum necessary' approach to autopsy during this very difficult and emotional time for them. By showing that a less invasive, imaging-led technique is achievable and can give accurate results, we intend to offer parents an alternative method for discovering the cause for their child's death. This means not only offering needle biopsies, but also understanding when imaging alone would be sufficient in coming to a diagnosis.

The information gathered by using an alternative, more acceptable method of investigation may help parents in their grieving process by providing closure and understanding for reasons behind their child's death, and help to understand whether future children are at increased risk of similar diseases.

Increasing the rate of uptake of this investigation may also mean we can better understand and explore reasons behind why children die. This can help inform public health policies and initiatives in improving maternal and child health. At present not all centres in the UK can offer a less invasive method of investigation. If this research can provide evidence for better practice, we hope this will set the foundations of a new NHS autopsy service.

(1) Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: When? Where? Why? Lancet 2005; 365:891-900
(2) Thayyil S, Sebire NJ, Chitty LS et al Post-mortem MRI versus conventional autopsy in fetuses and children: a prospective validation study. Lancet 2013; 382: 223-233

Objective: To develop a less invasive alternative to standard autopsy, thereby increasing parental acceptance rates of the post-mortem examination and information for causes of perinatal death.

Outcomes: Technique optimisation and establishment of diagnostic accuracy rates in post-mortem imaging and ultrasound guided organ biopsies for perinatal death investigation.

Methods: All fetuses referred for perinatal autopsy with tissue sampling will undergo a post-mortem MRI (if >500g) or micro-computed tomography (if <500g), and a comprehensive ultrasound study. The studies will be reported by radiologists specialising in post-mortem imaging, blinded to the results from the other modalities. After the initial radiology review, consensus imaging will take place and an ultrasound biopsy of target organs (liver, spleen, lungs, heart, and kidneys) will be performed. This will then be followed by the standard conventional autopsy (gold standard).

Diagnostic accuracy rates for MRI, micro-computed tomography and ultrasound imaging, both alone and in conjunction, will be established. Results from the ultrasound guided biopsy relating to both tissue yield and pathological diagnosis will be compared to autopsy findings. Where tissue yield is insufficient, technique will be adapted with either differing organ approach or biopsy needle calibre to ensure best possible tissue samples.

Scientific opportunities: If diagnostic accuracy and parental acceptance rates are high, medical opportunities will be created that allow increased monitoring of trends and explanation for causes of perinatal death. This raw data is currently lacking given the poor autopsy uptake rates.

The long term aim is for image led perinatal autopsy to become the 'best practice' and promote widespread NHS adoption for service delivery. Our results could then contribute to epidemiological studies and inform future public health initiatives.

The main objective of this research is to address whether and how less invasive, imaging-led methods for perinatal death investigation should be integrated into existing care pathways. Conventional autopsy uptake rates are currently low, meaning raw data regarding perinatal death diagnoses are lacking.

The main beneficiaries of this research would be parents and clinicians wishing to make a fully informed, evidence-based choice regarding alternative methods to the conventional perinatal autopsy. Parents would benefit as they will be offered a method that is non-disfiguring to their child, whilst still obtaining useful clinical information to inform future family planning. Clinicians would benefit by having clear guidelines (informed by the evidence produced in this research) regarding appropriate clinical indications for referral and diagnostic accuracy rates of different imaging modalities (including image-led tissue sampling options). Both these benefits hand-in-hand would represent an improvement in patient centred, individualised care with tailored counselling at a very emotional and difficult time.

Further advantages for clinicians would include opportunities for subspecialty training in 'image-led' post-mortem techniques. This would mainly benefit radiologists and pathologists with an interest in post-mortem medicine. With potential increases in demand and wider usage of the imaging techniques, there would be more focus on refining and developing these skills, in turn again benefitting future patients.

On a wider societal level, other beneficiaries would include the National Health Service, global policy makers and public health doctors. Our plan is for the results to eventually feed into an economic evaluation. If image-led post-mortem examinations are found to be clinically and economically advantageous, the introduction of such procedures as a 'new standard' may offer cost-saving benefits to our healthcare system. A more widely available and acceptable method of capturing causes of deaths could also allow better allocation of resources to inform public health initiatives both locally and world-wide, especially in socioeconomically deprived countries where the perinatal death rates are far higher.

Detailed measures of success and proposed milestones are fully outlined in our 'Pathways to Impact' statement. In the short term (within the first year), we shall present our preliminary work regarding imaging and image led biopsy techniques at international conferences and peer reviewed journals. By the end of the study (year 3), the final results regarding diagnostic accuracy rates and appropriate clinical indications for imaging-led procedures will be converted into peer-reviewed publications, technical reports to the Royal Colleges regarding best practice for 'image-led' post-mortem examinations and lay summaries for stakeholders and interested charity groups (GOSH charity, ARC - antenatal results and choices, SANDS - stillbirth and neonatal death charity). The lay documentation will be created in corroboration with our LIA -EAP (Less Invasive Autopsy External Advisory Panel).

In the longer term, we will identify a health economics team to devise a cost saving calculation on benefits to the NHS, should our techniques be integrated into patient care. We also hope to explore the feasibility of using our techniques in less developed countries to inform and enable ways of identifying and reducing perinatal death rates worldwide. This will be achieved via contact and discussion with current global health medicine project leaders (such as CaDMIA, World Federation of Paediatric Imaging (WFPI) and the global health taskforce of the British Society of Paediatric Radiology (BSPR)).