We take a lot of modern medicine for granted. While we know that we benefit from a huge legacy of information gleaned from hard work, observation and serendipity, we rarely consider this as we battle to understand more complex problems; the expression of cytokines following blast injury, the interplay of genetics and proteomics, and the potential of stem cells for regeneration of damaged tissues. But before we discuss current challenges, let’s take a step back in time.
Imagine it is the year 1600 and the purpose of the heart has yet to be identified. William Harvey, an English physician, will propose a novel and contentious theory in his 1625 treatise “De Motus Cordis”; On the Motion of the Heart and Blood.1 More than 100 years later another Englishman, Stephen Hales, invasively measures a horse’s blood pressure and sees it rise and fall alongside the pulse. Another 100 years and Poiseuille (France) develops an invasive mercury manometer for humans. Non-invasive techniques are suggested and improved over the next 75 years; Vierordt (Germany) in 1855, Von Basch (Austro-Hungary) in 1881, Riva-Rocci (Italy) in 1896, von Recklinghausen (Germany) in 1901 and finally Korotkoff (Russia) in 1905.
Thus it took almost 300 years for a fledgling theory to develop into the sphygmomanometer, a device that has barely changed over the last 100 years. This is a common pattern seen in the advance of modern medicine; a worldwide effort; great leaps forward followed by incremental advance; occasional retreats, false dawns and possibilities recognised but unrealised for years or decades.
Of course, the pace of change in modern medicine, and science in general, has increased exponentially; the equivalent of 100 years of research undertaken in medieval Europe now occurs annually. Fortunately, the exponential rate of knowledge growth has become necessary, as every detail we uncover reveals how severely we had under-estimated the complexity of the systems that we seek to understand and control. The slow gradual processes where a century can pass before the next substantial advance belongs to a more genteel age of the ‘gentleman’ scientist / physician.
But some medical problems remain difficult to express in terms of current knowledge. For example, fibromyalgia and chronic fatigue syndrome where there is a lack of a rationalist scientific base to guide medical practice. The absence of objective disease markers results in clinicians dividing into believers and non-believers, based on their personal biases and observations. This begs the question; how do you advance science in the absence of a common language from which to commence a dialogue? Imagine trying to manage blood pressure without an investigatory tool.
This is a scenario we confront on a day to day basis in brain injury rehabilitation. Given the complexity of defining what constitutes ‘improvement’, it can be extremely difficult to determine measures that can accurately assess this nebulous concept, let alone determine cause and effect. If ‘patient X’ has a near-miraculous recovery, was it the effect of my treatment, my with-holding of other treatment, or just plain (genetic) luck on the patient’s part?
In the remainder of this piece, I would like to present a fairly personal experience with these sorts of problems; an often unrecognised disease, believers and non-believers, an absence of diagnostic criteria and an indeterminate association with eventual rehabilitation outcomes. The journey to date has been the result of 15 years of investigation, a doctoral thesis and a slowly maturing field of research. For reasons of historical accident, we termed this condition Dysautonomia, although we hope that consensus will adopt the term Paroxysmal Sympathetic Hyperactivity (PSH),2 a term coined by Dr Alejandro Rabinstein of the Mayo Clinic in Rochester. In this context, PSH refers to ‘a syndrome following acquired brain injury consisting of transient, paroxysmal and simultaneous increases in sympathetic nervous system function [evidenced by elevated heart rate, blood pressure, respiratory rate, temperature, and sweating] and motor [posturing] activity.’
Working in a 12 bed traumatic brain injury (TBI) rehabilitation service, we would see 2 or 3 cases of PSH a year; there was great excitement when we managed to recruit a case series of 7-10 patients in any prospective study. On occasions, however, research submissions to mainstream neurology journals were met with scathing reviews and it took some time to work out the root cause; lack of awareness of the condition and confusion regarding the nomenclature.
In 2010, my colleague Iain Perkes was lead author on the largest review of PSH to date.3 This review identified 389 cases across many forms of acquired brain injury, 80% of which were TBI. Many of these cases were hard to track down; they were pre-internet and published under 31 different names; Autonomic Dysfunction Syndrome,4 ‘sympathetic storming’,5 acute midbrain syndrome,6 paroxysmal autonomic instability with dystonia,7 and so on. Clearly there was a problem in need of a fix. Furthermore, a secondary review of the cases,8 soon to be published in Brain Injury, showed that few clinicians had attempted to use diagnostic guidelines to determine cases of PSH. Another problem in need of a fix.
So when you have a condition that doesn’t have accepted diagnostic features, how do you arrive at consensus? First, borrowing from the history of medicine, you need to think internationally; second you need to identify people with hands on clinical and/or research expertise. Third you need to have a way of tapping into their experience. Thus, a consensus process was begun with goals to produce:
- a single name,
- a standard definition,
- and diagnostic guidelines
To achieve this aim, I met with individuals who each had multiple PSH publications to establish a steering committee. These individuals were based in Australia, Europe and the US. Secondarily we established a larger working party of 30 individuals from as wide a range of interested parties as possible; medicine and allied health, acute and subacute fields, adult and paediatric practice. In order to produce consensus on a diagnostic system we aimed to adhere to published principles that are thought to be integral to a robust consensus process, namely:
“(a) a structured and unbiased method for enrolling expert participants, (b) an explicit method for information elicitation or item generation, (c) the use of a respected moderator who has no vested interest in the results, (d) predefined consensus thresholds, and (e) an explicit feedback loop that is operative after each iteration.” (p.152)9
The Steering Committee selected the Delphi method,10 described in the following quote.
“Delphi is a consensus method developed by the Rand Corporation to utilize expert opinion for forecasting when insufficient information is available to make a knowledge-based decision . The Delphi process is being used in medicine with increasing frequency…. for “harnessing the opinions of an often diverse group of experts on practice-related problems” . Delphi is particularly well suited to the development of consensus-based guidelines .”11
The strengths of the Delphi method include its egalitarian nature; all responses are weighted equally and those being surveyed are unaware of the responses of other participants. This effectively reduces the potential for influential individuals to dominate proceedings, as might occur at round table discussions. The Delphi method is also economical, in that it requires relatively few resources other than the time of the surveyed experts and a coordinator to distribute and collate questionnaires.
The predefined consensus thresholds are taken from the following excerpt:
"For scales which are used as research tools to compare groups, a may be less than in the clinical situation, when the value of the scale for an individual is of interest. For comparing groups, a values of 0.7 to 0.8 are regarded as satisfactory. For the clinical application, much higher values of a are needed. The minimum is 0.90, and a=0.95….. is desirable."12
With respect to determining a diagnostic system, two options presented themselves; the probabilistic and the polythetic systems. The probabilistic system assigns a diagnostic likelihood for the condition, rather than providing a definitive diagnosis. This is distinct from a polythetic system which requires a minimum number of features to be present to arrive at the diagnosis.10 The probabilistic method was chosen, primarily due to the difficulty and importance of excluding other causes of excessive sympathetic nervous system activity such as sepsis and sedation withdrawal, particularly in the acute clinical setting. At present, the first round of data has been collated and is being analysed. We hope that finalised data will be available to present at the International Brain Injury Association meeting in Scotland in March, 2012.
Why spend so much effort trying to arrive at consensus? Data suggests that up to 25% of patients in ICU following severe TBI have the condition, at least temporarily.13 Does this have long-term consequence? Some of my colleagues think it doesn’t, but the brain has a long memory for insults. Emerging data in repetitive sports concussion or other forms of mild TBI suggests that there can be major long-term sequelae for injuries that would have been considered trivial in the past. If this is the case, could the prolonged hypersympathetic drive of PSH cause unnecessary secondary brain damage? Unsurprisingly, based on research to date I would argue that it does, but if so, how might we more effectively treat it? There are a range of pharmaceutical interventions that can mitigate the features of PSH, but what are the respective risks of these treatments?
Hopefully, these questions won’t take us another hundred years to answer.
A/Prof Ian Baguley MBBS PHD FAFRM
Senior Staff Specialist, Research Team Leader
Brain Injury Rehabilitation Service, Westmead Hospital
With assistance from
Iain Perkes MD
Brain Injury Rehabilitation Service, Westmead Hospital
- Booth J. A short history of blood pressure. Proc R Soc Med 1977; 70(11): 793–799
- Rabinstein AA. Paroxysmal sympathetic hyperactivity in the neurological intensive care unit. Neurol Research 2007; 29(7): 680-682
- Perkes I, Baguley IJ, Nott MT, Menon DK. A review of paroxysmal sympathetic hyperactivity after acquired brain injury. Ann Neurol 2010; 68(2): 126-135
- Rossitch E, Bullard DE. The autonomic dysfunction syndrome: aetiology and treatment. Br J Neurosurg. 1988;2:471-478.
- Lemke DM. Sympathetic storming after severe traumatic brain injury. Crit Care Nurse. 2007;27:30.
- Becker R, Benes L, Sure U et al. Intrathecal baclofen alleviates autonomic dysfunction in severe brain injury. J Clin Neurosci. 2000;7:316-319.
- Blackman JA, Patrick PD, Buck ML, Rust RS, Jr. Paroxysmal autonomic instability with dystonia after brain injury. Arch Neurol. 2004;61:321-328.
- Perkes IE, Menon DK, Nott MT, Baguley IJ. Paroxysmal Sympathetic Hyperactivity after acquired brain injury: a review of diagnostic criteria. Brain Injury, in press
- Ferguson ND, Davis AM, Slutsky AS, Stewart TE. Development of a clinical definition for acute respiratory distress syndrome using the Delphi technique. J Crit Care 2005; 20(2): 147-154.
- Graham B. Diagnosis, diagnostic criteria, and consensus. Hand Clin 2009; 25(1): 43-48, vi
- Arnold GL, Koeberl DD, Matern D, Barshop B, Braverman N, Burton B, et al. A Delphi-based consensus clinical practice protocol for the diagnosis and management of 3-methylcrotonyl CoA carboxylase deficiency. Mol Genet Metab 2008; 93(4): 363-370
- Bland JM, Altman DG. Cronbach's alpha. BMJ 1997; 314(7080): 572
- Baguley IJ, Slewa-Younan S, Heriseanu RE, et al. The incidence of dysautonomia and its relationship with autonomic arousal following traumatic brain injury. Brain Inj 2007; 21: 1175–1181