Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. i Treatments for Mild Traumatic Brain Injury: Fish Oil Supplementation and Information Provision in New Zealand Health Care Services A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Clinical Psychology at Massey University, Wellington, New Zealand Brylee Meredith Cresswell 2020 ii iii Abstract Mild Traumatic Brain Injury (mTBI) has the highest incidence of all brain injuries and can lead to symptoms in the physical, cognitive and mood domains. Most symptoms abate within weeks to months, however some individuals experience ongoing symptoms leading to longer- term disruption of social and occupational functioning. Current mTBI management recommendations include providing early injury related education, and addressing symptoms in a multidisciplinary fashion as they arise. As with any injury, it is important to ensure treatments are effective in order to reduce the costs to both the health system and the individual. The studies presented in this thesis aimed to assess the effectiveness of fish oil – a novel treatment for mTBI symptoms, and the current practice for providing information to mTBI patients. Study One was a randomised placebo controlled trial of fish oil as an adjunct treatment for mTBI symptoms. This study was cancelled due to recruitment difficulties, nevertheless the literature review delineates the pre-clinical evidence of its potential to treat both cognitive and mood symptoms via various pathways. In addition, the outlined procedures and the researcher’s reflections highlight the unforeseen difficulties that will need to be addressed should a similar trial be conducted in future. Study Two surveyed New Zealand health practitioners on their current practice of information provision for mTBI patients. It aimed to assess whether practices have changed in the 16 years since similar research was published, and since the introduction of information sheets by the Accident Compensation Corporation (ACC). It also assessed the quality and accessibility of the information provided. The frequency of information provision after mTBI has improved since 2004, though the issues with variability and formatting of presented information remain similar. The majority of respondents provided information verbally and in writing, and had information available only in English and standard print formats, potentially disadvantaging those with visual impairment or whose first language is not English. Time constraints, patient concentration and distress, and a lack of appropriate resources were cited as barriers to providing information. iv v Acknowledgements Conducting clinical research cannot be done singlehandedly. It requires the input and support of many people, perhaps even more so when things do not go according to plan. First, I would like to acknowledge the people suffering mTBI who participated in Study One. They kindly gave their time and energy during a time of illness and uncertainty when these personal resources were already scarce, and I am incredibly grateful for this. I hope that they obtained some benefit from their participation, and made timely and complete recoveries from their injuries. Thank you also to the numerous psychologists, occupational therapists, and physiotherapists who undertook recruitment for Study One. They accommodated me in team meetings to discuss the project, stored materials and took responsibility for majority of the recruitment work. These professionals did their best to recruit participants in often trying conditions, and their enthusiasm for the research provided a much needed boost when our collective efforts struggled to bear fruit. Many of these individuals mourned the termination of Study One alongside me, and provided encouragement for Study Two as well as future fish oil research. For all of this and more, I am very grateful. Thanks must also extend to the healthcare workers who completed the survey for Study Two. Without these people I would have had no data to work with. I appreciate them taking time out of their busy schedules to respond to what was likely one of many requests for research participation. Special thanks also to Mr Harvey Jones from the School of Psychology for providing all the necessary tools to conduct survey research, as well as plenty of guidance for using them. Thanks goes to my partner, Bruce, who has been my calm amidst the chaos of the last four years. To our baby boy Tai, gaining this qualification is even more meaningful now that you are in the world; I hope this one day shows you that you too can achieve your dreams. vi Also to my family, colleagues and friends, particularly those special people I have undertaken the clinical programme with, thank you for listening, encouraging, and believing in me. Finally, infinite gratitude goes to my research supervisors Professor Janet Leathem and Associate Professor Ross Flett. Ross, we may have had minimal direct contact, but having your research expertise to draw on was a great reassurance and I appreciated having you on my team. Janet, you have provided ideas, encouragement, condolences, and stern words when I needed them. You have shown me patience and compassion, and taught me to keep looking forward in times of difficulty. I hope to carry these qualities forward into a long and fruitful career. vii Table of Contents Abstract ......................................................................................................................................... iii Acknowledgements ........................................................................................................................ v List of Figures ............................................................................................................................... xii List of Tables ............................................................................................................................... xiii Overview ..................................................................................................................................... 14 Chapter One: Mild Traumatic Brain Injury and Post-Concussion Syndrome .............................. 19 Terminology ............................................................................................................................ 19 Mild Traumatic Brain Injury .................................................................................................... 20 Diagnosis. ............................................................................................................................ 20 Incidence. ............................................................................................................................ 22 Cost. .................................................................................................................................... 22 Post-Concussion Syndrome .................................................................................................... 23 Diagnostic criteria. .............................................................................................................. 23 Prevalence and cost. ........................................................................................................... 25 Post concussion syndrome as a diagnostic entity. .............................................................. 25 Injury Processes ...................................................................................................................... 27 Initial Injury. ........................................................................................................................ 27 Secondary neuronal injury. ................................................................................................. 28 Secondary glial involvement. .............................................................................................. 28 Symptoms ............................................................................................................................... 30 Acute symptoms. ................................................................................................................ 30 Ongoing Symptoms. ............................................................................................................ 32 Aetiology of ongoing symptoms. ........................................................................................ 36 Assessment of Post-Concussion Symptoms ............................................................................ 41 Treatment of Post-Concussion Symptoms .............................................................................. 42 viii Summary .................................................................................................................................. 44 Chapter Two: Omega-3 Polyunsaturated Fatty Acids ................................................................. 45 A Brief History of Dietary Omega-3. .................................................................................... 47 Omega-3 Fatty Acids and Brain Structure and Function. .................................................... 47 Dietary Omega-3 Fatty Acids and Mortality and Morbidity .................................................... 49 The Mediterranean Diet. ..................................................................................................... 49 Fish Consumption. ............................................................................................................... 51 Omega-3 intake in New Zealand ............................................................................................. 52 Dietary recommendations. .................................................................................................. 52 New Zealand population intake. ......................................................................................... 52 Supplemental intake. ........................................................................................................... 53 Supplemental Omega-3 PUFA Intervention Studies in Human populations ........................... 54 Omega-3 PUFAs and cognitive function. ............................................................................. 60 Omega-3 PUFAs and mood. ................................................................................................. 62 Omega-3 PUFAs after brain injury. ...................................................................................... 63 Supplemental Omega-3 PUFA Studies in Animal Models of mTBI .......................................... 65 Summary .................................................................................................................................. 68 Chapter Three: Study One Methodology .................................................................................... 70 Design ...................................................................................................................................... 70 Participants .............................................................................................................................. 71 Power analysis. .................................................................................................................... 71 Eligibility criteria. ................................................................................................................. 72 Exclusion criteria. ................................................................................................................. 72 Measures ................................................................................................................................. 73 Questionnaires. ................................................................................................................... 74 Cognitive tests. .................................................................................................................... 75 ix Scoring. ................................................................................................................................ 79 Materials ................................................................................................................................. 79 Localities ................................................................................................................................. 80 Procedure ................................................................................................................................ 80 Randomisation. ................................................................................................................... 80 Recruitment. ....................................................................................................................... 81 Testing and Communications. ............................................................................................. 82 Ethics ....................................................................................................................................... 83 Statistical Analyses .................................................................................................................. 83 Dissemination of Results to Participants ................................................................................ 84 Chapter Four: Study One Cessation ............................................................................................ 85 Study One Processes ............................................................................................................... 85 Who was involved in recruitment. ...................................................................................... 85 Alterations to Recruitment Processes .................................................................................... 87 Flow of Participants Through the Study.................................................................................. 88 Justification of Cancellation .................................................................................................... 90 Limitations .............................................................................................................................. 90 Recommendations .................................................................................................................. 91 Conclusion ............................................................................................................................... 93 Chapter Five: Information Provision after Mild Traumatic brain Injury ..................................... 94 Information Provision as Treatment for MTBI ........................................................................ 94 Information provision for adults. ........................................................................................ 94 Information provision for young persons. .......................................................................... 97 Recommendations for Information Provision ........................................................................ 99 Issues with Information Provided Following MTBI ............................................................... 103 Information Provision in the New Zealand Context ............................................................. 106 x Summary ................................................................................................................................ 107 Chapter Six: Study Two Methodology ....................................................................................... 109 Design .................................................................................................................................... 109 Hypotheses ............................................................................................................................ 109 Participants ............................................................................................................................ 110 Measures ............................................................................................................................... 110 Survey questions. .............................................................................................................. 110 Assessment of information sheets. ................................................................................... 111 Ethics ..................................................................................................................................... 113 Procedure .............................................................................................................................. 113 Dissemination of Results to Participants ............................................................................... 114 Chapter Seven: Study Two Results ............................................................................................ 115 Respondent Characteristics ................................................................................................... 115 Provision of Written Information Sheets .............................................................................. 117 Format of Information ........................................................................................................... 119 Content and Accessibility of Written Information ................................................................ 119 General Advice Provision ....................................................................................................... 121 Specific areas of concern. .................................................................................................. 121 Medications & Supplements. ............................................................................................ 123 Referral to concussion service. .......................................................................................... 124 Barriers to Providing Information .......................................................................................... 125 Further Information ............................................................................................................... 125 Summary ................................................................................................................................ 126 Chapter Eight: Information Provision Discussion ...................................................................... 128 Usefulness of Information Provided ...................................................................................... 128 Content guidelines. ............................................................................................................ 128 xi Accessibility guidelines. .................................................................................................... 129 Advice for Different Populations .......................................................................................... 131 Barriers to Providing Information ......................................................................................... 132 Flow of Patient Information Across Services ........................................................................ 133 Conclusion ............................................................................................................................. 134 Recommendations ................................................................................................................ 135 Limitations ............................................................................................................................ 135 Chapter Nine: Overall Discussion .............................................................................................. 137 Terminology .......................................................................................................................... 137 Diagnostic Difficulties ........................................................................................................... 138 Flow of Information Through Services .................................................................................. 138 Treatment ............................................................................................................................. 139 Recommendations for Future Research ............................................................................... 141 Reflections ............................................................................................................................ 142 Bibliography .......................................................................................................................... 145 Appendix A: Study One Contact Log ......................................................................................... 171 Appendix B: Study One Self-Report Eligibility Questionnaire ................................................... 182 Appendix C: Study Two Information Sheet and Survey ............................................................ 184 Appendix D: DHB Research Approval Communications ........................................................... 189 Appendix E: Case Study ............................................................................................................. 201 xii List of Figures Figure 1.1. The neurochemical cascade of concussion. .............................................................. 30 Figure 2.1. The chemical structure of EPA and DHA. .................................................................. 46 Figure 2.2. Potential points of DHA intervention on the concussion cascade. ........................... 68 Figure 3.1. Recruitment process undertaken by the concussion services. ................................. 81 Figure 3.2. Individual participants’ flow through the study. ....................................................... 82 Figure 4.1. Flow of participants through the study. .................................................................... 89 Figure 7.1. Percentage of respondents from each service area providing written information to mTBI patients. ........................................................................................................................... 118 Figure 7.2. Percentage of respondents from each service area providing written information for specific groups of mTBI patients. ......................................................................................... 118 Figure 7.3. Percentage of practitioners in each service area providing information on specific areas of concern. ...................................................................................................................... 122 Figure 7.4. Symptom reported by patients as most concerning in each service area………….…123 Figure 7.5. Medications and supplements advised by each service area. ................................ 124 Figure 7.6 . Barriers to providing information reported by each service group. ...................... 125 xiii List of Tables Table 2.1 Randomised Placebo Controlled Trials of the Cognitive and Mood Effects of LC n-3 PUFA Supplementation ............................................................................................................... 55 Table 3.1. Test Selection and Associated Cognitive Domains in Order of Administration ......... 76 Table 3.2 List of replacement terms used in the MR-IV .............................................................. 77 Table 4.1. Concussion Service Involvement in Trial Recruitment ............................................... 86 Table 7.1 Number of Health Professionals and Regional Distributions of Respondents in each Service Area .............................................................................................................................. 116 Table 7.2 Average EQIP, FRE and Clarity Scores for each Service area ..................................... 121 14 Overview Mild Traumatic Brain Injury (mTBI) represents 70-90 percent of all TBI and a significant public health problem. Typical symptoms in the acute phase of mTBI include headache; dizziness; nausea; confusion; focal neurological signs such as tinnitus; and cognitive symptoms such as memory, attention, and executive function difficulties. These symptoms usually abate within days or weeks post-injury, however 20-30% of sufferers develop ongoing symptoms that can persist for many months, or even years (Barker-Collo et al., 2015; Heitger et al., 2009). Ongoing symptoms can delay an individual’s return to work, school, or other premorbid activities, and are associated with increased rates of anxiety and depression (Macleod, 2010; McCauley et al., 2008; Mickeviciene et al., 2002; Styrke, Sojka, Björnstig, Bylund, & Stålnacke, 2013). Post-concussion syndrome (PCS) is the common term for the experience of ongoing symptoms after mTBI, but is a contentious issue. Debate exists about both the classification and causation of post-concussion symptoms. Rates of diagnosis may differ greatly depending on the criteria used, and some studies have found persisting cognitive deficits in individuals who do not meet diagnostic criteria for any specific disorder (Barker-Collo et al., 2015; Oldenburg, Lundin, Edman, Nygren-De Boussard, & Bartfai, 2016). Regardless of diagnosis, symptoms may persist and cause distress for some individuals. Standard practice for treating mTBI in New Zealand involves taking a symptom-based approach within a bio-psycho-social framework, including offering reassurance and information regarding symptoms and their management (The New Zealand Guidelines Group, 2006). For those ongoing, more severe or complex post-concussion symptoms, treatment such as occupational therapy or cognitive-behavioural therapy may be offered alongside medical intervention, through a concussion service (Accident Compensation Corporation [ACC], 2019). For the majority of individuals who suffer injury, returning to work and normal function as 15 soon as possible is desirable, and having a safe, non-invasive and readily available treatment option would be welcomed. Omega 3 polyunsaturated fatty acids (n-3 PUFA) have been found to have many health benefits and may have a role to play in the recovery of cognitive function following brain injury. Studies using rodent models of mTBI have demonstrated multiple possible pathways of injury attenuation. These include lowered neuroinflammation, axonal and neuronal damage, and apoptosis after injury (Barrett, McBurney, & Ciappio, 2014); lowered levels of beta amyloid precursor protein (APP; Trojian & Jackson, 2011); lowered expression of pro- inflammatory cytokines, and prevention of microglial activation to their pro-apoptotic form, significantly lessening both neuroinflammation and the associated behavioural deficits (Zendedel et al., 2015). The few studies that have included n-3 PUFA as a treatment for TBI sequelae have utilised multiple supplements and medications thus precluding a causal inference of n-3 PUFA’s role in participant improvement. As n-3 PUFAs have shown experimental efficacy in treating mTBI symptoms, they may have an additive effect in treating these symptoms within the current standard practice. Due to the dearth of research with human mTBI participants, high quality human trials are necessary to determine whether n-3 PUFA supplementation could be an effective treatment. The opportunity to conduct the present research arose as I entered the Doctor of Clinical Psychology Programme. A PhD student under the same lead supervisor had a significant quantity of a high-quality fish oil supplement, and placebo, left over from a randomised control trial (RCT) with a different population, and after reading the research cited above, sufferers of mTBI were seen as a population who may benefit from further RCT research. Despite the significant efforts of all involved, Study One did not attract enough participants to be a feasible trial. Due to this and the impending expiry date of the 16 supplements the timeframe could not be extended, and so the trial was ceased. Factors contributing to lower than expected participation are discussed in Chapter Four. Throughout the recruitment period and during practical placements, I noted that many mTBI sufferers expressed confusion and concern regarding their symptoms and recovery, and often did not possess written materials to refer to. These individuals were distressed by their uncertainty, which was concerning given the potential for such distress to maintain mTBI symptoms and for quality information to lower distress. It was clear that the provision of information to mTBI patients warranted investigation. As Study One drew to a close, information provision became the focus of a subsequent study, which replicated and extend previous survey research conducted by the lead supervisor and a former master’s student approximately sixteen years prior. Moore and Leathem (2004) found that only 45.9% of the New Zealand Emergency Departments and General Practitioners that responded to their postal survey provided written information to patients who had a confirmed or suspected mTBI. These ranged from one to ten pages in length, and just over half met the criteria for being able to be read by 70% of the population. Since this study took place, at least two mTBI-specific patient leaflets have been produced by the ACC. It is not known however, whether these are routinely provided to patients by their healthcare practitioners, how accessible they are, or which leaflets are most used. International studies have shown substantial variability in the written information provided to patients after mTBI (Baker, Unsworth, & Lannin, 2015; Kempe, Sullivan, & Edmed, 2014; Macdonald, McMillan, & Kerr, 2010; Peachey, Hawley, Cooke, Mason, & Morris, 2011). Patients should be provided with written discharge information that advises about what to expect and what to do in the first few days post-injury, a description of which symptoms signal a medical emergency or require a reassessment, and a description of the symptoms that may be experienced in the post-acute period and how to manage these (Kempe et al., 2014; 17 Peachey et al., 2011). Given the New Zealand Guidelines Group’s (2006) recommendations for information and reassurance as a first-line treatment for mTBI, the lack of recent New Zealand research into mTBI information provision and the frequent uncertainty noted from mTBI sufferers, a replication of Moore and Leathem’s 2004 work was warranted. In addition, this would provide an indication of if and how practice has changed since 2004. Chapter One provides a background for the two studies by presenting the definitions, diagnostic criteria, and epidemiology of both mTBI and PCS. It also includes a discussion of the contention surrounding PCS as a diagnostic entity. Evidence for biological and psychological aetiologies of post-concussion symptoms is presented, as are the current treatment guidelines for New Zealand. As background to Study One, Chapter Two presents a history of dietary n-3 PUFA consumption and evidence of its benefits, before discussing several randomised controlled trials of n-3 PUFA supplementation in human populations with various cognitive difficulties. As there are currently no published trials of the specific effect on human mTBI, a discussion of rodent models of n-3 PUFA supplementation following mTBI is also presented, along with an outline of the possible mechanisms of effect. Chapter Three draws together the evidence of n-3 PUFAs potential for effect on post- concussion symptoms and presents justification for Study One as well as specific aims, hypotheses, and methodology. Chapter Four describes alterations made to Study One’s methodology in response to operational challenges, and its ultimate cessation. It also outlines the key limitations and recommendations for future research. Chapter Five reviews literature relating to the importance of providing information to those who have suffered mTBI. The measures for assessing the quality of information are presented and common issues with post-mTBI information provision are discussed. Chapter Six contains the hypotheses and methodology for Study Two. Chapter Seven presents the results 18 from the Study Two survey, and Chapter Eight discusses these results in relation to prior research and future recommendations. Finally, Chapter Nine provides a discussion of the research as a whole and sets out limitations and recommendations for further research in the area of treatment following mTBI. The thesis concludes with researcher reflections on the process, difficulties, and highlights of conducting these studies, from both professional and personal perspectives. 19 Chapter One: Mild Traumatic Brain Injury and Post-Concussion Syndrome This chapter reviews the current literature on mild traumatic brain injury (mTBI) and post-concussion syndrome (PCS). It includes an examination of the differing terminology within this field of research before presenting incidence rates and diagnostic criteria for both mTBI and PCS. The review then proceeds from a more integrated standpoint, considering the symptoms experienced after suffering mTBI rather than specific diagnoses. Research relating to the acute and chronic symptoms suffered after mTBI is presented, as is the aetiology of symptoms, and their assessment and treatment in current practice. The purpose of this review is to provide a background for both Study One and Study Two. For Study One it will address the rationale for the recruitment criteria adopted, and the possible mechanisms of effect of a fish oil supplement on mTBI related symptoms. For Study Two, it will provide an overview of symptoms and their potential duration, which comprises part of the information patients may benefit from receiving. Terminology Many different terms have been used to describe mTBI and the symptoms experienced afterwards. Some alternative terms include mild head injury, minor head injury, concussion, and post-concussion disorder. New terms are also being suggested, such as post- inflammatory brain syndrome (Rathbone, Tharmaradinam, Jiang, Rathbone, & Kumbhare, 2015), as evidence for similar symptoms across differing trauma aetiologies is discovered. The New Zealand Guidelines Group (2006) used ‘head trauma’ or ‘head injury’ to describe the initial injury to the head, rather than the brain, because head injuries do not always cause brain injuries and individuals who have had an injury to the head may not always consider ‘brain injury’ information applicable to them. This group also noted that while classification of the initial severity of the injury can be useful for predicting some short- and long-term outcomes, some terms (e.g., mild) may not be acceptable to the injured individual, as the impact on their functioning may be far from mild. 20 The term concussion is common in research as well as everyday language, however most recent research instead uses the term mild traumatic brain injury or mTBI. Concussion and mTBI are considered synonymous, though as outlined in the next section, mTBI is sometimes considered more severe than concussion. The term concussion is frequently used in the context of sport related head injury, and is often considered to be at the milder end of the mTBI spectrum as while the acute symptoms are the same as any other injury mechanism, recovery reportedly occurs within a shorter timeframe (Carroll, Cassidy, Holm, Kraus, & Coronado, 2004; Karr, Grindstaff, & Alexander, 2012; Reuben, Sampson, Harris, Williams, & Yates, 2014). In line with previous research, the present study will use the terms concussion and mTBI synonymously. The differing diagnostic criteria used to categorise PCS have led to some confusion within the literature. The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition – Text Revised (DSM-IV-TR; American Psychiatric Association [APA], 2000) stipulates that symptoms must have been present for a minimum of three months after sustaining mTBI, while other classification systems do not clearly state a required timeframe. In line with this, many studies use the term post-concussion syndrome to describe any symptoms occurring after mTBI, without considering a specific timeframe, thus there may be overlap between mTBI and PCS. Definitions and classifications of both mTBI and PCS are explained in more detail in the following two sections. Mild Traumatic Brain Injury Diagnosis. Traumatic brain injury (TBI) is sustained when a direct external mechanical force, or acceleration and deceleration forces, are applied to the head and result in a disruption of brain function. There are countless scenarios in which such injuries can occur, though some of the most common are motor vehicle accidents, falls, and assaults (Feigin et al., 2013). The severity of TBI is usually classified using the Glasgow Coma Scale (GCS) score from the initial emergency room presentation, the duration of loss of consciousness (LoC), and/or 21 the duration of post-traumatic amnesia (PTA), and may be considered mild, moderate, or severe. Seventy to 90 percent of TBI fall into the ‘mild’ category (Cassidy et al., 2004); this end of the spectrum is the focus of the present studies. The World Health Organisation (WHO) Collaborating Centre for Neurotrauma Task Force on Mild Traumatic Brain Injury recommend the following criteria as an operational definition of mTBI, based on the American Congress of Rehabilitation Medicine (ACRM) and their own literature review: MTBI is an acute brain injury resulting from mechanical energy to the head from external physical forces. Operational criteria for clinical identification include: (i) 1 or more of the following: confusion or disorientation, loss of consciousness for 30 minutes or less, post-traumatic amnesia for less than 24 hours, and/or other transient neurological abnormalities such as focal signs, seizure, and intracranial lesion not requiring surgery; (ii) Glasgow Coma Scale score of 13–15 after 30 minutes post-injury or later upon presentation for healthcare. These manifestations of MTBI must not be due to drugs, alcohol, medications, caused by other injuries or treatment for other injuries (e.g., systemic injuries, facial injuries or intubation), caused by other problems (e.g., psychological trauma, language barrier or coexisting medical conditions) or caused by penetrating craniocerebral injury. (Carroll, Cassidy, Holm, Kraus, & Coronado, 2004, p. 115). Variations of these criteria, including one or more of the listed signs and symptoms but excluding others, can be found throughout the literature. It is because of such discrepancy in the previous research that the above-listed criteria were published, however as Levin and Diaz- Arrastia (2015) note, neither WHO nor ACRM specified any minimum duration for LoC or PTA, nor did they specify how practitioners should differentiate between extreme stress caused by a traumatic event, and confusion caused by trauma to the head. The possibility of misclassification remains. 22 Incidence. Cassidy and colleagues (2004) reviewed 121 studies from around the world and found that the annual incidence of mTBI was approximately 100-300 cases per 100,000 population. These researchers acknowledge that while most research comes from hospital settings, many cases of mTBI are not treated in hospitals, thus the true incidence is likely closer to 600 per 100,000 population. In a population based study in the Waikato region of New Zealand Feigin et al. (2013) reported 790 traumatic brain injuries per 100,000 person years. Of these 790 injuries, 749 - almost 95% - were considered mild, classified by a GCS score of 13 – 15 or a period PTA less than 24 hours in duration. This incidence rate is likely more accurate than the vast majority reported in the literature, owing to sampling techniques that involved review of hospital and general practitioner records, contact with sports clubs and schools, and offered individuals the chance to self-refer through responding to advertisements. Cost. Traumatic Brain Injury (TBI) is a major public health concern both abroad and in New Zealand. The Accident Compensation Corporation1 (ACC; 2016) recorded 13,519 new claims for head injuries/concussion in the year leading up to June 2015, and the 17,416 active claims during this time cost a total of $75,616,367. These costs represent rehabilitation, compensation for being incapacitated for work, and ancillary services such as transport to treatment. Although these cost statistics do not account for severity, with 95% of TBI in New Zealand being mild it is likely that a large proportion of the total cost represents recovery from mTBI. Indeed in a study of the cost of TBI in New Zealand from a societal perspective, Te Ao and colleagues (2014) estimated the one year cost of mTBI in 2010 to be US $3395 (NZ $4719) per case, and the lifetime cost per case as US $4636 (NZ $6444). While the per case cost for mild injury was considerably lower than the cost per moderate/severe case, the substantially higher incidence of mild cases meant that the total cost of mTBI was approximately three times higher than moderate and severe TBI. 1 ACC is a state-sector organisation providing no-fault accident insurance and injury prevention initiatives. It is funded by levies on income tax, business, fuel, vehicle licencing, and government funding, and is available to every person who suffers an injury caused by an accident in New Zealand. 23 Post-Concussion Syndrome Diagnostic criteria. The two most widely used classification systems in the field of psychology are the DSM and the International Classification of Diseases (ICD). Both of these systems provide diagnostic labels for ongoing symptoms after mTBI and the different iterations of each system list different criteria. The DSM-5 (American Psychiatric Association [APA], 2013) label for such ongoing symptoms is ‘mild neurocognitive disorder due to traumatic brain injury’ and criteria include: • Evidence of modest cognitive decline compared to premorbid function in one or more cognitive domains. This may be based on the concern of the injured individual, the clinician, or a knowledgeable third party. The impairment must be documented by an appropriate quantified clinical assessment. • Cognitive deficits do not prevent the individual from completing instrumental activities of daily living, though compensatory strategies or extra effort may be necessary. • The symptoms do not occur solely during an episode of delirium and are not better attributed to another mental disorder. • There must be evidence of TBI such as loss of consciousness, post-traumatic amnesia, disorientation and confusion, or neurological signs. • The problems present immediately after injury, or upon regaining consciousness, and “persists past the acute post-injury period” (APA, 2013, p. 624). While criteria for the duration of LoC and PTA, as well as a GCS score of 13-15 are outlined as necessary for evidencing a mTBI, the DSM-5 does not provide guidelines for the duration of the acute post-injury period, other than a rather vague statement that mTBI usually resolves within a few weeks to months (APA, 2013). The DSM-5 criteria represent a change from the DSM-IV-TR research criteria for postconcussional disorder. The DSM-IV-TR criteria focused on assessment of attention and memory rather than other cognitive domains, and required the individual to be experiencing 24 three or more symptoms such as: becoming fatigued easily; disordered sleep; headache; vertigo or dizziness; irritability or aggression on little or no provocation; anxiety, depression, or affective lability; apathy or lack of spontaneity; and other changes in personality. These symptoms must have developed following a head trauma of sufficient severity to cause significant concussion, be present for more than three months, and if pre-existing symptoms were present, they must have been significantly worsened by the trauma. Contrary to the DMS-5’s neurocognitive disorder, the symptoms of postconcussional disorder must lead to a significant impairment in social or occupational functioning (APA, 2000). In line with changes to the DSM, the ICD criteria evolved from a ‘postconcussional syndrome’ to a ‘mild neurocognitive disorder’ when moving to the latest version. The ICD 10 postconcussional syndrome simply required the presence of three or more symptoms similar to those from the DSM-IV-TR postconcussional disorder criteria, and stated that depression or anxiety may be present, attributing this to loss of self-esteem and fear of permanent brain injury (World Health Organisation [WHO], 2004) thereby highlighting the psycho-social aspects of PCS while to some extent discounting the possible neuro-chemical causes of mood dysfunction. Like the DSM-5, neither the ICD-10 or -11 provides guidance for the timeframe of symptoms or recovery. Mild neurocognitive disorder in the ICD-11, as with the DSM-5 disorder of the same name, requires a subjective decline from the previous level of functioning and objective evidence of such a decline in one or more cognitive domains that does not significantly interfere with activities of daily living. While this can be specified as due to concussion, no limits are placed on the severity of the initial injury (WHO, 2019). The lack of consensus in the diagnostic criteria both reflects and perpetuates disparities in the PCS literature. High quality research is required to inform diagnostic criteria, and standardised criteria are necessary for comparing conclusions between studies and collecting valid epidemiological data. It is not yet clear when the acute phase of mTBI ends and 25 PCS begins, nor what degree of dysfunction the sufferer must experience in order to qualify for diagnosis. Prevalence and cost. Diagnostic criteria chosen may have serious repercussions on prevalence rates, cost to the health system, and compensation claims. In a study comparing the ICD-10 and DSM-IV-TR criteria for postconcussional syndrome (PCS) and postconcussional disorder (PCD) at 6 months post-injury, it was found that 14.4% of the participants met PCD criteria, and 44.6% met criteria for PCS (McCauley et al., 2008). Additionally, higher proportions of participants who had the potential to access compensation for their injury through workplace compensation schemes or litigation met criteria for each disorder. Of those who had the potential to access compensation, 28.3% met criteria for PCD and 60.9 % met criteria for PCS, compared with 7.8% and 37.8% respectively for those who did not have the potential for compensation. While the effect sizes of these findings were considered small, the costs to healthcare systems could potentially be large. In the 2015 calendar year, ACC received 1168 claims for PCS, which included diagnoses termed ‘postconcussional syndrome’ and ‘post-traumatic brain syndrome’ (ACC, 2016). The 1927 active (requiring some form of payment from ACC) claims for these diagnoses during 2015 cost $31,159,993 (ACC, 2016). Individuals initially covered for TBI may have their diagnosis changed (or added to) and therefore their injury costs initially attributed to TBI then later covered under PCS, thus the PCS cost differs from the TBI costs discussed previously. It may be that if PCD criteria were required for cover, the above figure would be considerably lower. Post concussion syndrome as a diagnostic entity. Post-concussion syndrome is a contentious issue. According to several studies (e.g., Heitger et al., 2009; Barker-Collo, et al., 2015) the symptoms of mTBI usually abate within days or weeks post-injury, however 20-30% of sufferers develop ongoing symptoms that can persist for many months, or even years. Several researchers (e.g., Levin & Diaz-Arrastia, 2015; Meares et al., 2011; Mounce et al., 2012; 26 Rathbone et al., 2015) argue that residual symptoms such as fatigue and headaches are non- specific, occur frequently in the general population, and may be misattributed to mTBI. Levin and Diaz-Arrastia (2015) state that the evidence for persisting cognitive deficits is weak, as there are few prospective longitudinal studies longer than six months. Many authors however, seem to consider chronic symptoms to be those that persist for longer than one to three months (Carroll et al., 2014). This is sometimes stated explicitly, but is more often implicit in the assertion that symptoms usually remit within this time for the majority of individuals. In a three month prospective study of 62 adults with mTBI and 58 non-brain trauma controls, Meares et al. (2011) found that mTBI did not predict PCS using the ICD-10 criteria; PCS rates and symptoms did not differ between groups. Pain however, was associated with PCS and the authors posited this may be due to the increased attention given to physical sensations when people experience chronic pain, leading to greater endorsement of PCS like symtoms. In contrast, Webb and colleagues’ (2015) study of US airmen with mTBI compared to those with other injuries, reported hazard ratios for cognitive disorder NOS (not otherwise specified) and memory loss at 180 days or longer since injury as 10.75 and 4 respectively. This study excluded individuals who had suffered mTBI in the two years prior to the study entry, or mTBI during the study period, indicating that a single mTBI event can have ongoing effects. Oldenburg, Lundin, Edman, Nygren-De Boussard, and Bartfai (2016) compared a PCS group, recovered mTBI group and a community control group on measures of post-concussion symptoms and various cognitive domains. The PCS group was defined as those who endorsed three or more symptoms of the Rivermead Postconcussion Symptoms Questionnaire (RPQ) at three months post-injury, thus meeting the ICD-10 criteria for PCS, while the recovered mTBI group had fewer or no symptoms three months post injury. There was limited evidence of the PCS group performing lower on cognitive measures than the recovered group, and the authors attributed the slight variation to the pre-morbid characteristics of the PCS group as they were found to have a lower ‘cognitive reserve’ based on IQ, education and occupation information. 27 There was however, a small but significant difference between the combined PCS and recovered group when compared with the community control group on the total recall, consistent long-term recall, and cued recall trials of the Selective Reminding Test (SRT). This led the authors to conclude that PCS sufferers are not a subgroup of individuals who have suffered mTBI, but that mTBI is associated with ongoing subtle executive memory deficits. In this study 23.5% of the recovered group (15.6% of the total sample) were experiencing one or two ongoing symptoms from the RPQ. Had the DSM-5 criteria requiring only evidence of modest cognitive decline in one domain been employed, the PCS group may have been substantially larger and between group results may have differed. This study highlights not only the problems of inconsistent diagnostic criteria, but also a common question within mTBI research: Should PCS be a diagnostic entity? Most researchers agree that a small percentage of individuals do suffer ongoing symptoms after mTBI, however some such as Oldenburg and colleagues (2016) believe that many of those suffering mTBI experience ongoing but very subtle effects rather than a clinical syndrome or disorder. Similarly, in a 12 month prospective cohort study of adult mTBI sufferers, Booker, Sinha, Choudhari, Dawson, and Singh (2019) considered ‘persistent post- concussion symptoms’ (PPCS) rather than a specific diagnosis due to the lack of consensus on PCS as a diagnosis. Because participants in Study One may have been given various diagnoses for similar symptoms, and such different labels may lead to confusion within both academic and patient literature, the review will consider the experience of symptoms after mTBI rather than specific diagnoses from this point forward. Injury Processes Initial Injury. As the mechanisms of injury that lead to mTBI are varied, different areas of the brain may be affected depending on the site of the force to the head. Frontal and temporal areas are most frequently injured (Broshek, De Marco, & Freeman, 2015; Mathias, 28 Beall, & Bigler, 2004). These injuries are usually diffuse, though focal symptoms and space occupying lesions may occur in a minority of cases (Borgaro, Prigatano, Kwasnica, & Rexer, 2003). Secondary neuronal injury. Secondary to the initial mechanical trauma, a complex cascade of reactions occurs in the damaged area of the brain. Membrane disruption and stretching leads to a loss of Na+/K+/Ca2+ flux, neural membrane depolarisation, and indiscriminate release of excitatory neurotransmitters (MacFarlane & Glenn, 2015). These neurotransmitters bind to N-methyl-D-aspartate (NMDA) receptors, leading to further neuronal depolarisation and contributing to the excitotoxicity (Giza & Hovda, 2001). In an effort to restore chemical balance to the damaged area of the brain, sodium/potassium pumps work excessively, a process which requires a large amount of energy, or adenosine triphosphate (ATP); this hyper-metabolic state occurs during a time of reduced cerebral blood flow, resulting in a disparity between energy supply and demand, and thus leads to energy crisis (Barkhoudarian, Hovda, & Giza, 2011). The influx of calcium during the excitotoxicity phase can further damage axonal structures and exacerbate the energy crisis (Barret, McBurney & Ciappo, 2014); as the brain attempts to provide more ATP during a time of decreased oxidative metabolism, it increases rates of glycolysis, leading to the production of lactate, which due to the decreased rate of oxidative metabolism cannot be metabolised effectively. Lactate accumulation may lead to neuron dysfunction by way of acidosis, membrane damage, altered blood brain barrier permeability, and oedema (Giza & Hovda, 2001). ATP stores become diminished after the period of hyper-metabolism, leading to a hypo- metabolic state where insufficient supplies of energy are available; this may last days to weeks (MacFarlane & Glenn, 2015). Secondary glial involvement. Damaged neural cells signal the activation of microglia. This is initially a protective factor but may become a positive feedback loop where an overactive inflammatory response can damage neuron cell membranes further (Barret et al., 29 2014). Inflammation and other brain injury phenomena are frequently studied in animals, to allow for controlled injury and treatment conditions as well as close study of the brain post- mortem. A study of diffuse mTBI in pigs observed neuroinflammation 6 hours post injury (Lafrenaye, Todani, Walker, & Povlishock, 2015). It found pervasive microglial activation in thalamic areas of injured pigs, the degree of which was correlated with the amount of observed diffuse axonal injury (DAI). Activated microglial processes exhibited increased contact with neurons positive for amyloid precursor protein (APP) – a marker of the axon damage that occurs due to the initial injury, exacerbated by the resultant excitotoxicity and oxidative stress (Barrett et al., 2014) – when compared with microglial process contact with myelinated axons in sham injured animals. The majority of contacts in injured animals were bulbous end processes, rather than the passing by/crossing over of microglial processes and axons in sham pigs, who were operated on but not injured. The authors speculate that the influx of calcium that occurs in neural injury may alter the extracellular calcium concentration, signalling the observed convergence of microglial processes. Activated microglia perform a number of functions that may be helpful or harmful in the injured brain, such as phagocytosis, cytokine secretion and/or neutrophil secretion, thus this study was among the first to show the relationship between DAI and acute neuroinflammation (Lafrenaye et al., 2015). In a rat study using closed head injury as an mTBI model, Singh, Trivedi, Devi, Tripathi, and Khushu (2016) discovered evidence of an inflammatory cascade. This began with increases in the pro-inflammatory cytokine TNF-α four hours post-injury, proceeding to increases in the anti-inflammatory cytokine IL-10 one day post-injury, and at three- and five-days post-injury increased amounts of hypertrophied astrocytes and glial fibrillary acidic proteins (GFAP) were found. GFAP is a brain specific protein expressed predominantly by astroglia and has been found to differentiate mTBI sufferers from others (Singh et al., 2016). Injury cascades occur in both neurons and glia after the initial mechanical insult to the brain. These processes lead to energy crisis, inflammation, and impaired neurotransmission, 30 which in turn lead to the experience of symptoms such as fatigue, headache, and cognitive difficulties that characterise mTBI (MacFarlane & Glenn, 2015); acute and ongoing symptoms are discussed further in the following sections. See Figure 1.1 below for a pictographic representation of the neurochemical processes that occur after mTBI. This provides a background for the possible points of intervention by n-3 PUFA, which will be discussed in depth in the following chapter. Figure 1.1. The neurochemical cascade of concussion. Adapted from ‘ω-3 fatty acid supplementation as a potential therapeutic aid for the recovery from mild traumatic brain injury/concussion’ by E. Barret, M. McBurney and E. Ciappio, 2014, Advances in Nutrition: An International Review Journal, 5, p. 269. Symptoms Acute symptoms. Upon regaining consciousness or recovering from post-traumatic amnesia, most individuals who have sustained mTBI experience symptoms, though the duration of the symptoms varies widely (Karr, Areshenkoff, & Garcia-Barrera, 2014). Nausea, tinnitus, difficulty concentrating, and memory loss are common symptoms that Macleod Hypermetabolism Hypometabolism Injury Oxidative stress Membrane Disruption/Excitotoxicity Axon Damage Neural Cell Damage Mitochondrial Dysfunction Altered Neurotransmission Altered Gene Expression Apoptosis Inflammatory Response Endoplasmic Reticulum Stress 31 (2010) states wear off within days to weeks, as the cellular physiology heals. However, according to Donovan, Cancelliere, & Cassidy (2014), self-reported symptoms of fatigue, dizziness, and headache are the most common reasons for seeking treatment after mTBI. With regard to cognitive symptoms, Kay, Newman, Cavallo, Ezrachi and Resnick (1992) found that speed and ease of processing were the primary cognitive deficits after mTBI. Using the Rey Auditory Verbal Learning Test (RAVLT), they found that what appeared to be difficulty with memory related more to the encoding and registration of new information, rather than a deficit in retrieval. Similarly, Shanmukhi and Panigrahi (2003) found significant differences in learning and remembering a non-verbal pattern when comparing 40 sufferers of mTBI approximately two weeks post-injury with 40 uninjured controls. This was considered a deficit in procedural learning. A common criticism of research into the symptoms of mTBI is the comparison to uninjured control groups, as the pain of the injury and shock of the injury event may confound findings. To combat this, Landre, Poppe, Davis, Schmaus, and Hobbs (2006) administered the adult vigilance and distractibility subtests of the Gordon Diagnostic System, the Wechsler Memory Scales logical memory 1 and 2 subtests, the Trail Making Test A and B, as well as measures of body pain, mental health, and post-concussive symptoms to 37 hospitalised patients with mTBI and 39 patients hospitalised for other traumas, approximately 5 days after injury. The mTBI patients scored significantly lower on all the cognitive measures with moderate to large effect sizes, and scores on these measures were not related to pain or stress. Interestingly, no differences were found in self-reported post-concussive symptoms, as measured by a modified version of the Postconcussive Symptom Checklist, between groups. It may be that such checklists measure symptoms common to multiple traumas, while the cognitive tests are more specific to mTBI. In a retrospective cohort study using electronic data for active duty US Air Force airmen, Webb and colleagues compared mTBI, other injured, and non-injured groups at less 32 than 30, 31-179, and 180 days or longer post-injury. In accordance with Landre and colleagues (2006), many cognitive symptoms were more prevalent in the mTBI group. In the ‘acute period’ (<30 days post-injury) hazard ratios for the mTBI group compared to the other-injured group were 55.88 for memory loss/amnesia and 85.17 for ‘cognitive disorder NOS’. However, the risk of seizures, headaches, migraines, pain, dizziness, and sleep disorders were also significantly higher in the mTBI group, compared with both the other-injured group and the full cohort for the acute injury period. Results were adjusted for post-traumatic stress disorder (PTSD) and depression, thus the symptoms reported are likely to be accounted for by the brain injury. The research presented above provides evidence of both cognitive and physical symptoms occurring within the first week to first month of injury. Donovan and colleagues (2014) formed the International Collaboration on mTBI prognosis (ICoMP) to update the WHO task-force meta analytic results from 2004 and stated that substantial evidence exists for the presence of cognitive deficits in the first two weeks after injury. However, according to both the ICoMP authors and Karr et al. (2014) the cognitive deficits reported are very heterogeneous, and effect sizes range greatly between studies. Less evidence exists for ongoing cognitive difficulties following mTBI. Ongoing Symptoms. As Carroll and colleagues (2014) and Levin and Diaz-Arrastia (2015) point out, longitudinal studies of cognitive deficits following mTBI with a follow up period greater than six months are lacking. The few that have been published yield varied results and contain methodological biases that make generalisation questionable. In a prospective cohort study, Røe, Sveen, Alvsåker, and Bautz-Holter (2009) followed up with 52 adult mTBI sufferers at 3, 6, and 12 months post-injury. More than half (55.8%) of the sample met the ICD-10 criteria for PCS at 3 months post injury, and 42.3% met criteria at 6- and 12-months post-injury. Cognitive symptoms were more prominent at each time point than physical or behavioural symptoms. Two participants who met criteria at 6 months no longer 33 met criteria at 12 months; interestingly, two participants developed PCS between the 6 and 12 month follow up. The authors did not theorise the reasoning for this, however they did highlight the high degree of between participant variability in the progression of symptoms. The authors analysed results for differences between males and females in the development of PCS; no sex differences were found. However, this study included a range of ICD-10 diagnostic codes covering diagnoses including coma; haemorrhage; focal, diffuse, and unspecified injury. While participants were required to have a GCS of 13-15, they were recruited after being admitted to hospital care up to 48 hours post-injury, so at least some of the injuries suffered may have been on the moderate to severe end of the TBI scale. In addition, individuals who dropped out of the study tended to report fewer symptoms than those who completed all follow ups, and the sample size of 52 participants completing all follow ups was small. In a Swedish population based three year follow up study, it was found that 50% of females and 30% of males who had presented to an emergency department after mTBI met the ICD-10 criteria for PCS according to their RPQ scores (Styrke et al., 2013). In contrast to Røe et al. (2009), these results represented a significant difference between sexes. In this study almost all RPQ symptoms, with the exception of double vision in females, were significantly higher in the mTBI population than the control group of blood donors. The Rivermead Head Injury Follow Up Questionnaire (RHFUQ) assessed disability, finding 52% of females and 37% of males experienced some level of disability after mTBI. The most frequently endorsed items were tiredness at work, difficulty sustaining previous workload, and difficulty in ability to enjoy previous leisure activities. Unsurprisingly, endorsement of a high frequency of symptoms on the RPQ predicted disability in both sexes (Styrke et al., 2013). This study may represent a higher injury severity than other research on mTBI, as although each individual in the mTBI group had a GCS of 13-15 upon presentation to the ED, it included any degree of LOC or disorientation, PTA, nausea, vomiting, severe headache, & neurological deficit. It may also 34 have been subject to non-response bias; while alcohol intoxication at the time of the injury – which has the potential to lower GCS scores – was more common among individuals who declined to participate in the study, loss of consciousness was more prevalent among respondents. This may indicate that those who responded to the postal survey were more severely injured. In a six month prospective cohort study Roy et al. (2019) found that injury severity factors predicted PCS at both one and six months post-injury. A combination of loss of consciousness and altered mental state (feeling dazed, confused, or disoriented) within 24 hours of mTBI predicted both depression and PCS symptoms, as per the RPQ, one and six months later. Similarly, a six month Dutch cohort study found a higher injury severity scores on the Injury Severity Scale (ISS) and Abbreviated Injury Scale – Head (AISH) were associated with PCS six months post mTBI (Voormolen et al., 2019). Additionally, Voormolen and colleagues found that female sex, lower education, assault as the injury mechanism, hospitalisation and a higher RPQ score predicted lower health-related quality of life on the SF-36 health survey. The participants meeting PCS criteria had on average a 20% lower health-related quality of life compared with mTBI sufferers who were not experiencing three or more PCS symptoms, and had lower scores on all SF-36 domains (physical functioning, bodily pain, role limitations due to physical health problems, role limitations due to personal or emotional problems, emotional well-being, social functioning, energy/fatigue, and general health perceptions) than the Dutch population norm. In line with Voormolen and colleagues, a 12 month English adult mTBI cohort study found female sex, assault as the injury mechanism, and injury severity factors (GCS of less than 15) as well as psychiatric illness history and alcohol intoxication at the time of injury was associated with poorer outcomes on the RPQ and RHFUQ (Booker et al., 2019). While these studies found similar causes of poor outcomes including injury severity, they all represent samples collected from hospital emergency departments and thus likely more ‘severe’ mTBI. 35 In contrast to the previously discussed studies, Barker-Collo and colleagues (2015) recruited participants from numerous settings including self-referrals (participants responded to advertisements), and utilised cognitive testing rather than solely self-report measures for a 12 month follow up study. They found that symptoms endorsed on the RPQ declined steadily over the 12 months. At baseline, the majority of participants performed within the average ranges of the CNS Vital Signs neurocognitive assessment, but a substantial proportion of individuals performed in the very low range for executive ability (21.5%), complex attention (26.0%) and cognitive flexibility (27.4%). In addition, over 20% of the sample produced below average scores across indices of processing speed, executive ability, psychomotor speed, reaction time and cognitive flexibility. The mean performance on all cognitive domains except for memory improved from baseline to 12 months, with the majority of this improvement from baseline to six months; improvements from six to 12 months were non-significant for all domains except for processing speed – one of the most common deficits following mTBI (Barker-Collo et al., 2015; Borgaro et al., 2003; Diwakar et al., 2015; Killgore et al., 2016). At 12 months follow up 16.3% of participants remained in the very low range on the test of complex attention, where only 0.13% of the general population would be expected to score (Barker- Collo et al., 2015). Additionally, over 20% of participants continued to have some level of difficulty with complex attention and memory at the 12 month follow up period. While a substantial subset of Barker-Collo and colleagues’ participants experienced ongoing symptoms, the authors noted that a high proportion of participants scored in the above average ranges on tests of visual memory and executive ability from one-month follow- up onwards, and on cognitive flexibility and complex attention at six- and 12-months follow up. Much like Oldenburg and colleagues (2016) these authors theorise that this may be due to the relatively high education level of their sample, as those with higher education tend to have a larger cognitive reserve and may be less prone to difficulties after mild brain injury (Satz et al., 1993; Stern, 2002). 36 It seems that regardless of whether or not PCS is diagnosed, a proportion of individuals who have suffered a mTBI may benefit from treatment of cognitive symptoms, and the assessment of some factors, such as cognitive reserve and injury severity features, may help to prospectively identify who will fall into this group. Aetiology of ongoing symptoms. As mentioned earlier, post-concussion symptoms are a complex problem of unclear aetiology. The differing classification systems used to diagnose PCS and PCD allude to divergence in the literature regarding the cause of ongoing symptoms. The ICD-10 (WHO, 2004) states that PCS aetiology is unclear and includes both neurological and psychological factors, but seems to consider the ongoing symptoms from a psychosocial standpoint with its assertion that the depression and anxiety commonly found in PCS sufferers is due to fear of permanent disability and an injury to self-esteem, rather than the physical injury (McCauley et al., 2008). Some sufferers may become hypochondriacal, search for diagnosis and cure, and adopt a permanent sick role (WHO, 2004). The DSM-5 (APA, 2013) acknowledges both psychological and biological factors. Traditionally, initial symptoms such as headache and dizziness have been attributed to biological processes, while ongoing symptoms have been attributed to psychosocial problems. Psychosocial issues typically include anxiety and depression that develop following injury, as well as pre-injury morbidity and personality styles. There is however, a growing body of evidence highlighting potential physiological causes of persisting symptoms after mTBI. Psychogenic symptoms. Acute symptoms usually remit within days or weeks, however for some individuals psychological symptoms evolve during this time and complicate the clinical picture. According to Kay et al. (1992) persisting cognitive symptoms may alter an individual’s sense of self, leading to anxiety and low mood which exacerbate cognitive symptoms, causing further frustration and distress, maintaining the disorder. In this positive feedback loop psychological sequelae may become more disabling than the initial injury. There are many factors that may contribute to emotional upset following mTBI, for example, a 37 distressing injury event such as an assault, or cognitive difficulties preventing normal functioning. Mickeviciene et al. (2002) found few differences between mTBI sufferers and age and sex matched controls in their postal survey 22-35 months post-injury. The only significant differences between groups were sensitivity to alcohol, worry about brain injury, and feelings of depression which were higher in the mTBI group. Symptoms may have remitted by the time of follow up. McCauley et al (2008) found an increased incidence of major depressive episodes and post-traumatic stress disorder six months post-injury. Rather than post-injury psychological symptoms, Meares’ and colleagues’ (2011) found that pre-injury psychological symptoms predicted PCS. Their study of acute PCS symptoms in mTBI and other trauma patients found that pre-injury depression or anxiety and pre-injury post-traumatic stress disorder increased the odds of developing PCS after both brain and non- brain injury (OR for depression or anxiety: 2.99, OR for PTSD: 1.05). Pre-injury psychological difficulties and personality traits may also affect the duration of symptoms. Hou et al. (2012) conducted a prospective cohort study with 107 adult mTBI sufferers recruited from a hospital. After phone-based follow ups three and six months post injury, they discovered that 22% of participants met the ICD-10 criteria for PCS at three months, and 21% met this criteria at 6 months. Regression analyses showed that all-or-nothing behaviour, as measured by the Behavioural Response to Illness Questionnaire (BRIQ), was the only significant predictor of PCS at three months post-injury, and head injury perceptions as measured by the Brief Illness Perception Questionnaire (BIPQ) were the only significant predictor of PCS six months post-injury, with all-or-nothing behaviour trending towards significance (Hou et al., 2012). Maladaptive cognitive responses to injury were present early in the recovery process and may have been reinforced over time, as individuals attempted too much and experienced cognitive and physical failures. The authors suggest early mTBI 38 intervention targeting coping processes as a way to prevent the development or lessen the severity of PCS. Like Mickeviciene et al. (2002), Styrke and colleagues (2013) discovered a higher rate of depression three years post-injury. However, Barker-Collo and colleagues (2015) found that the incidence of depression declined over their 12-month study period, while anxiety increased until 6 months post-injury, then declined until the 12-month end point. Similar to Hou et al. (2012) Snell, Hay-Smith, Surgenor & Siegert (2013) interviewed 147 mTBI sufferers within 3 months of their injuries, then again 6 months later, and found that participants with stronger injury identity beliefs, expectations of lasting severe consequences, and distress at the initial meeting had higher odds of poor outcomes at the second meeting. The authors applied Leventhal’s Common Sense Model, whereby individuals construct their own representations of a health condition to help make sense of the symptoms experienced and provide a basis for coping. In a 2015 cluster analysis study Snell and colleagues found that participants could be grouped into high, medium, and low adaptation clusters based on injury beliefs. This presents a possibility for identifying individuals who may be at risk of developing PCS, and thus may benefit from higher intensity intervention to prevent the development or shorten the course of the syndrome. In a study of 171 individuals with TBI of all severities, Lee, Jayasinghe, Swenson, and Dams-O’Connor (2019) found that the personality trait of dispositional optimism was positively correlated with cognitive function at least one year post-injury. This association was significant after controlling for age, race, positive and negative affect, health status, and injury severity. The authors state that optimism is a teachable skill and recommend further research into its utility as a TBI symptom reduction tool. Considering the above studies, it seems that psychological symptoms may be both a risk factor for, as well as an outcome of, mTBI. Pre-existing psychological difficulties and maladaptive injury related beliefs may affect the development and duration of other mTBI 39 symptoms, and psychological symptoms may develop after mTBI complicating recovery from cognitive symptoms. Neurogenic symptoms. Broshek and colleagues (2015) suggest that emotional difficulties experienced after mTBI may be due to the initial structural and secondary metabolic injury, noting that injury often involves frontal-temporal damage or diffuse axonal injury (DAI) to cortical areas connecting to the limbic system. Indeed in a neuroimaging study of 56 young male athletes, Chen, Johnston, Petrides, and Ptito (2008) found that athletes suffering both PCS and depressive symptoms exhibited altered neural responses during a working memory task. There were no significant differences in performance on the task between the four groups which consisted of: three groups of concussed athletes with PCS symptoms five to seven months post-injury – without depression, with mild depression, or with moderate depression as measured by the BDI-II; and a healthy control group. The PCS with depression groups showed reduced activation in the prefrontal cortex and striatum, less task related reduction in activity in several other brain regions, and reduced grey matter volume in the medial frontal and temporal cortical regions when compared to healthy controls and concussed athletes without depressive symptoms. As these athletes were young and had no history of mood disorders, the authors reason that it is unlikely that the lower grey matter volumes were premorbid and acted as a diathesis for the development of depression, but were probably due to injury processes. Heitger and colleagues (2009) compared 36 PCS sufferers to matched controls who had recovered well after mTBI, on measures of eye movements and cognition. Eye movement control relates closely to the functional integrity of the brain, as it requires the coordination of neural circuitry in both cortical and subcortical structures, as well as the cerebellum; several parameters of eye movements are beyond cognitive control (Cifu et al., 2015; Heitger et al., 2009). The PCS group in the Heitger et al. (2009) study performed lower than controls on anti and self-paced saccades, memory guided sequences, and smooth pursuit eye movements, 40 suggesting problems in response inhibition, short-term spatial memory, motor-sequence programming, visuospatial processing, and visual attention. The PCS group also performed lower on neuropsychological measures of memory, complex attention, and executive function, however the oculomotor measures were more strongly related to symptom load as measured by the RPQ. The control group in this study were found to have a higher predicted IQ determined by the Wechsler Test of Adult Reading (WTAR), which affected some of the neuropsychological measures; the oculomotor deficits however, were unrelated to depression, intellectual functioning, or malingering, indicating ongoing cerebral impairment in the PCS group. Su et al. (2014) examined the C-reactive protein (CRP) levels of 213 patients with mTBI in a three month follow up study. CRP is a biomarker of systemic inflammation that is sensitive, but not specific, to trauma. Thus, participants who had suffered other injuries, recent infection, previous head trauma, were cognitively impaired or had psychiatric diagnoses were excluded from the study. Su and colleagues (2014) found that higher CRP levels at baseline were associated with persistent post-concussion symptoms, psychological problems (anxiety and depression), and cognitive impairment, with odds ratios of 2.72, 1.54, and 1.69 respectively. This indicates that systemic inflammation, as well as neuro inflammation, may play a role in post concussive symptoms. A small Diffusion Tensor Imaging (DTI) study of 15 mild to moderate TBI sufferers and 15 matched controls found that alterations in the microstructural integrity of white matter was associated with poor balance and cognitive symptoms after injury (Kim et al., 2019). This suggests glial involvement in ongoing symptoms as well as the systemic, neural, and psychological involvement found in other studies. There is consistent evidence of acute cognitive symptoms such as memory, attention, and processing speed deficits, after mTBI. There also exists some evidence for ongoing symptoms, however methodological issues as well as the heterogeneity of both symptoms 41 experienced as well as their progression, prevent a consensus being reached regarding the typical course of symptoms and the utility of diagnosis. Both psychological and physiological processes underlie symptoms, though more research is needed to delineate their respective contributions and interactions. Understanding the mechanisms behind post-mTBI symptoms helps to inform intervention. As will be discussed in later chapters, n-3 PUFA may exert effects on symptoms via physiological pathways, while information provision affects psychological processes. Assessment of Post-Concussion Symptoms The assessment of PCS depends on the diagnostic criteria used. The vast majority of studies employ the ICD-10 criteria and rely on self-report symptom measures such as the RPQ , and the Post-Concussion Symptom Checklist. As with all self-report measures this introduces the possibility of several biases. One bias particularly common with post-concussion symptom reporting is the ‘good old days’ bias. After suffering mTBI many individuals under-report premorbid symptoms, misperceiving their premorbid functioning to be better than the average person (Iverson, Lange, Brooks, & Ashton Rennison, 2010; Lange, Iverson, & Rose, 2010; Yang et al., 2014). Unsurprisingly, Iverson and colleagues (2010) found that this effect was more pronounced in participants who failed an effort test. Those who failed the Test of Memory Malingering (TOMM) under reported premorbid symptoms, and reported more current symptoms, than both mTBI participants who passed the TOMM, and healthy controls (Iverson et al, 2010). As outlined in above sections, various neuropsychological tests are used to evaluate cognitive symptoms following mTBI. Tests of memory, attention, processing speed and executive function are common. According to Heitger and colleagues’ (2009) study however, oculomotor testing had a stronger correlation than the neuropsychological measures with post-concussion symptom load and problems with activities of daily living. Both Heitger et al. (2009) and Cifu et al. (2015) propose eye movement tracking as an effective measure to 42 distinguish PCS sufferers from healthy or recovered controls, however the required equipment is not yet commonly possessed by concussion clinics and more research may be necessary before this method can be considered for standard practice. Treatment of Post-Concussion Symptoms Current treatment for symptoms after mTBI includes rest and information regarding the course of injury, with more specialist treatment such as occupational therapy or psychological therapy offered for individuals who are identified by the concussion service as requiring further input (The New Zealand Guidelines group, 2006; ACC, 2019). For treatment of ongoing symptoms after mTBI, the New Zealand Guidelines Group (2006) recommend symptom-based approach to minor problems, and offering the injured individual reassurance and information regarding symptom management and strategies. For clinically significant TBI, which may encompass PCS, the New Zealand Guidelines Group (2006) posit that rehabilitation should be functionally oriented, provide information and support to family/whānau/carers of the injured individual, and provide access to psychological assessment and intervention when required. A Canadian paper by Marshall and colleagues (2015) recommended a similar symptom-based approach. Their paper updating clinical practice guidelines for the treatment of persisting symptoms following mTBI recommended: initiating rehabilitation of any cognitive impairments found during formal cognitive assessment, or if learning compensatory strategies would facilitate the individual returning to work or other functional activity; including both compensatory and remediation approaches to cognitive rehabilitation; and where possible, informing employers or teachers of potential temporary alternative duties or working hours in order to avoid anxiety related to cognitive difficulties and experiencing repeated errors or setbacks in work or school. They also recommended Cognitive Behavioural Therapy (CBT) for ongoing mood, anxiety and/or sleep difficulties, gradual return to sport/exercise and leisure 43 activity as symptoms allow, physiotherapy and/or neurology for peripheral and vestibular problems, and pharmacotherapy for headache. MTBI and PCS are heterogeneous conditions; every case involves a unique combination of forces acting on an individual’s unique anatomy (Taber & Hurley, 2013) and is likely moderated by his or her unique personality, beliefs, and coping styles. Thus, there is no ‘one size fits all’ solution to intervention and as the New Zealand Guidelines Group (2006) recommend, a multi-disciplinary approach is favourable. ACC manage the rehabilitation of head injuries caused by accidents in New Zealand, and employ a ‘concussion service’ for claimants who have suffered mild or moderate TBI. These concussion services are non-governmental organisations (NGOs) that apply to ACC to provide services for a contracted price. The concussion service is offered from a biopsychosocial perspective, and each service provider must include a minimum of a medical specialist with qualification in neurology, or internal medicine with a focus on brain injury; a clinical neuropsychologist; and an occupational therapist (ACC, 2019). Each service provider must also have access to other professionals such as optometrists, various types of counsellors, cultural advisors, and advocacy services. Individuals are referred to the concussion service via primary health care professionals (e.g., a general practitioner [GP], or a medical or allied health professional from a district health board [DHB] if the individual presented to a hospital) or by an ACC case manager. ACC (2019) stress the individual nature of injury, and as such each individual’s pathway through the service is dependent on recommendations made after a comprehensive assessment. The organisation also places emphasis on educating clients about mTBI in order to prevent PCS, though explicitly states within the services operational guidelines that clients should not be provided such education if the diagnosis has not yet been confirmed. If treatment within the service is indicated, this may include medical consultation, psychological counselling, and/or allied health and vocational services specific to the individual’s goals and circumstances (ACC, 2019). 44 Summary Mild traumatic brain injury (mTBI) is caused by mechanical forces to the head which disrupt normal brain function. Traditionally mTBI has been referred to as concussion, a term that continues to be used in some services and by some researchers though is frequently considered to be at the ‘milder’ end of the mTBI spectrum and heal within weeks. Regardless of terminology, the injury involves primary and secondary components, as complex injury cascades begin in the acute period and continue into the post-acute period. Many individuals suffer from physical, cognitive, and affective symptoms in the acute period, but recovery is usually rapid, occurring within days or weeks for most individuals. A small percentage of mTBI sufferers however, experience symptoms that persist beyond this time. Chronic symptoms tend to include fatigue, headache, and difficulties with memory and attention. There is debate around the classification and cause of ongoing symptoms. While chronic symptoms have traditionally been attributed to psychological causes, recent research has outlined physiological deficits and processes that contribute to symptoms and suggest ongoing cerebral impairment. A biopsychosocial approach to treatment is usually taken, and in New Zealand multidisciplinary teams are engaged to provide intervention. Interventions include education, reassurance, and psychotherapy when necessary, as well as allied health input depending on the individual’s circumstances. This approach adheres to guidelines for individualised treatment, however given the significant minority who suffer ongoing symptoms and an increasing understanding of the mechanisms and aetiology of symptoms, there is scope for adjunct intervention. One possible adjunct intervention is omega-3 polyunsaturated fatty acid supplementation, discussed in the next chapter. 45 Chapter Two: Omega-3 Polyunsaturated Fatty Acids Omega-3 polyunsaturated fatty acids (n-3 PUFAs) are a class of nutrients essential to the human body. Fats perform many functions in the body: they provide the most energy per gram (37kJ/g) of all the macronutrients; provide a means of storing energy; insulate the body and its organs; form part of cell membranes, hormones, and other chemical messengers; and aid the absorption of fat soluble vitamins (Whitney, Rolfes, Crowe, Cameron-Smith, & Walsh, 2011). Fatty acids consist of a hydrocarbon chain with a carboxyl group at one end (named the alpha end) and a methyl group at the other (named the omega end); different types of fats have different saturation levels – the saturated fats contain the maximum number of hydrogen atoms as can be attached to the carbon chain (the carbon chain is ‘saturated’ with hydrogen atoms), the unsaturated fats contain at least one carbon that is not saturated with hydrogen, and at this point a double bond exists with the next carbon in the chain (Whitney et al., 2011). The present research focuses on the long chain n-3 PUFAs (LC n-3 PUFAs) eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) as these have shown promise in various arenas of health research, including cognitive health. These fatty acids consist of hydrocarbon chains 20 (EPA) and 22 (DHA) carbon atoms in length. Each has a carbon to carbon double bond three carbons from the omega end of the chain, hence the term ‘omega- 3’ and contains further double bonds – or points of unsaturation – hence the term polyunsaturated (Lunn & Theobald, 2006). See Figure 2.1 for a depiction of these molecules. 46 Figure 2.1. The chemical structure of EPA and DHA. Each line represents the bond between carbon atoms; double lines represent double bonds. This illustrates the polyunsaturated nature of these fatty acids. Adapted from Whitney et al., (2011). While most fats can be synthesised in the human body, alpha linolenic acid (ALA) – the precursor to both EPA and DHA – and linoleic acid (LA) cannot; these are termed essential fatty acids as they must be obtained from the diet (Whitney et al., 2011). Although EPA and DHA can be synthesised from ALA, the rate of synthesis is very low and may not meet an individual’s requirements, making them ‘conditionally essential’ nutrients – it is essential to obtain them from the diet when synthesis is inadequate (Whitney et al., 2011). Estimates of the amount of ALA converted to DHA in the body range from less than one percent, to up to nine percent (Arterburn, Hall, & Oken, 2006). A limiting factor in the conversion of the n-3 PUFAs is competition for the enzymes necessary for this process. The omega-6 (n-6) fatty acids, present in many foods such as nuts, grains, and vegetable oils, use the same enzymes for similar conversion processes, thus the presence of these fats can impede EPA and DHA synthesis, particularly when the ratio of n-6 to n-3 is high (Whitney et al., 2011). The optimal ratio of n-6 to n-3 fats is approximately 1-2:1, however the typical ‘western diet’ has a widely varied but largely out of proportion ratio of 10-25:1 (Lewis, 2016). In addition, males convert less ALA to EPA and DHA than females (Arterburn et al., 2006), and individual differences in stress hormones, genes, vitamin and mineral deficiency, 47 and alcohol consumption can also limit the synthesis of EPA and DHA in the body (Uauy & Valenzuela, 2000). Due to these limitations, dietary sources of EPA and DHA are recommended (Lunn & Theobald, 2006). The richest sources of these fatty acids are oily fish such as salmon, tuna, and mackerel (Whitney et al., 2011), as well as algae sources (Lunn & Theobald, 2006; Weiser, Butt, & Mohajeri, 2016). The remainder of this chapter will focus on the role of the LC n-3 PUFAs in the body – particularly the brain. Information regarding the relationship between a high LC n-3 PUFA diet and mortality and morbidity, n-3 PUFA consumption in New Zealand and abroad, and studies using EPA and DHA supplements in human populations will be reviewed. The chapter concludes with a specific focus on EPA and DHA supplements for mTBI. A Brief History of Dietary Omega-3. Diet plays an essential role in health and development through the lifespan and may have played a central role in human evolution. A high quality diet was both necessary for, and a consequence of the evolution of the large human brain (Kyriacou, Parkington, Marais, & Braun, 2014). A key dietary factor for the development of the large brain in the human species was fish and/or shellfish due to their concentration of iodine, iron, and DHA (Cunnane, 2005; Cunnane, Plourde, Stewart, & Crawford, 2007). Indeed, excavation of stone age sites have unearthed evidence of abundant and intentional fish and shellfish intake by both humans and Neanderthals (Cunnane et al., 2007; Kyriacou et al., 2014; Stringer et al., 2008). Given that the nutrients present in these foods, particularly DHA, are essential for the development of brain structure and function, the exploitation of such marine resources may have played an essential role in making humans what we are today. Omega-3 Fatty Acids and Brain Structure and Function. The brain is a fat-rich organ; DHA represents over 90% of its n-3 PUFA, and 10-20% of its total lipids (Weiser et al., 2016). While EPA exerts acute actions in the brain and is absorbed in approximately equal proportions to DHA, very little EPA exists in brain tissue as it is rapidly oxidised and removed, 48 or converted into other fatty acids (Weiser et al., 2016). Therefore, n-3 PUFAs’ effects on brain structure and function are considered to be significantly dependent on DHA. DHA begins to accumulate in neural tissues at a high rate from the third trimester of gestation, continuing at this rate to the end of the second year of life (Lauritzen et al., 2016; McCann & Ames, 2005). High levels are then maintained throughout the lifespan. Individual differences exist in DHA accretion, with an important factor being dietary consumption. For infants, this includes consumption via maternal DHA stores, which are transferred through the placenta and breast milk (Innis, 2007). Later in life, DHA must be consumed in its complete form, or synthesised following the consumption of other n-3 PUFAs, though as outlined previously the synthesis process is inefficient. Brain DHA has an estimated half-life of 2.5 years, as it is metabolised at a rate of approximately 4mg per day (Barceló-Coblijn & Murphy, 2009; Weiser et al., 2016). As DHA is mainly synthesised in the liver and the rate of synthesis in the brain is very low, brain DHA levels are maintained through delivery via the circulating blood (Chen et al., 2015). DHA is concentrated in the cerebral grey matter, and stored primarily in membrane phospholipids (Brenna & Diau, 2007). It cycles in and out of the membrane from phospholipids to the intracellular free fatty acid pool, providing a mechanism to help meet the energy demands of rapid growth and other cellular challenges (Chen, Green, Orr, & Bazinet, 2008; Weiser et al., 2016). DHA can influence membrane fluidity, lipid raft function, neurotransmitter release, transmembrane receptor function, gene expression, signal transduction, myelination, neuroinflammation, and neuronal growth and differentiation (Innis, 2007). Thus it plays several important roles in the structure and function of the brain throughout the lifespan. DHA deficiency has been associated with alterations in normal neurodevelopment, neural inflammation, and increased mortality and morbidity (Muldoon et al., 2010). 49 Dietary Omega-3 Fatty Acids and Mortality and Morbidity The Mediterranean Diet. Dietary patterns high in LC n-3 PUFA have shown an association with health and longevity. Possibly the most well known and most widely researched