All subspecialties are a product of their history, and Neurology and Internal Medicine are no exceptions. Classical Neurology evolved in Europe on the bedrock of clinical semiology and postmortem neuropathological correlation in places such as Pitié-Salpêtrière and Queen Square hospitals. The pioneers of Clinical Neurology in the United States and indeed other countries looked to these beginnings when they set up departments and training programs during the last century. From their point of view, a study of the brain and its meager relevant therapeutics had little in common with the wider Internal Medicine. Knowledge of Internal Medicine was not as crucial to the training of a neurologist at that time as neuropathology and repeated exposure to the intricacies of neurological examination. It is in this environment that the neurological method evolved in the image of neuropathological correlation studies: a lesion is localized by detailed neurological examination, differential diagnoses are generated, and individual hypotheses are tested. Apart from the intellectual elegance of this approach, any delay caused by the meticulousness and deliberate pace of progress hardly mattered. There were few time-sensitive therapies to be offered to the neurological patient, and for a long time, Neurology was seen as the very definition of therapeutic nihilism by other subspecialties.^1,2,3

These beginnings had predictable effects on the development of Neurology as a profession, particularly in North America. Most Neurology departments separated from Internal Medicine and, in some cases, Psychiatry to form independent entities with a separate curriculum. Internal Medicine training was relegated to a single year in the immediate postgraduate year of training for neurologists and much of the neurology content was removed or simplified in Internal Medicine training. Most of the practice of neurology was conducted either in the ambulatory setting or in the setting of hospital consultation to the primary care teams. This created a unique gap in American health care provision where the general physicians might not have the required expertise to manage neurological disease, and the neurologists, untrained in the treatment of multisystem disease, might in advertently neglect the general care of their inpatients.

While Neurology was predominantly a diagnostic and consultative specialty, these deficiencies were less palpable. But since effective therapies are now become available for acute neurological conditions, it is no longer adequate for the neurologists to treat their patients in consultation only or to only accept less critically ill patients into their service. The neurology inpatients now often suffer from multiple serious systemic conditions that complicate their care in the hand of the general neurologist. Recent changes to reimbursements and hospital recruitment policies have taken note of these changes, so that neurohospitalist medicine is one of the fastest growing areas within neurology. Additionally, there is an increased demand for neurological knowledge, for management of strokes and seizures in community settings where access to neurological consultation is either difficult or untimely. Hospitalists, trained as internists, critical care physicians, or emergency physicians, have had to retrain themselves to incorporate many more neurological conditions into their repertoire. Meanwhile, no corresponding changes to the respective curricula have occurred to adjust for this change.

In other countries, neurology training is part of advanced training in Internal Medicine, akin to Cardiology or Endocrinology in the United States. The numbers who are trained in Neurology are therefore less than what is the case in North America. The care of most neurological patients, outside of major centers, is in the hand of (often excellent) general physicians who deal with the most common neurological conditions and may have access to neurological consultation. For these physicians, the minutiae of rare neurological diseases are out of reach and a practical approach to common and critical conditions is prioritized.

A neurohospitalist, whether a neurologist or an internist who sees hospital neurological cases, is distinguished by the type of conditions she treats, the skill set she must possess, and her general approach to the diagnosis and treatment of patients.

What conditions are seen by neurohospitalists?⁴

The hospital practice of neurology concentrates on the 3 “C”s: the common, the critical, and the curious. Although neurology abounds in interesting conditions that have clear syndromes that may be explained by their genetics and functional neuroanatomy, the recognition and treatment of the majority of these are not time sensitive and may be deferred to the outpatient setting. Most of neurohospitalist practice may be summarized as follows:

1. 
   

   The Common: The most common neurological presentations to the hospital and the emergency department are strokes, seizures, headaches, exacerbations of multiple sclerosis (MS), and vertigo. One has to add to this list conditions that most neurologists do not consider neurological but are often consulted on: neurotrauma, syncope, loss of consciousness, back pain, and acute confusional states. An efficient and algorithmic approach to these conditions can streamline the high volumes and present the consulting team with consistency they can rely on for future referrals.

   
   

   Some of the common consults for patients admitted for other reasons include management of Parkinson disease (PD) while an inpatient, comatose patients, gait and balance problems, and management of other neurological conditions such as epilepsy before surgery to cite some examples.

   
   
2. 
   

   The Critical: Neurological emergencies—although relatively rare outside of stroke, central nervous system (CNS) infections and epilepsy—are nonetheless important and time-sensitive conditions with which a neurohospitalist must be familiar. Strokes, seizures, myelopathies, neuromuscular emergencies, movement disorder emergencies, and rapidly progressive dementias fall in this category. A working knowledge of dementias and delirium is often necessary for diagnosing rapidly progressive dementia and dealing with confused patients. Whenever necessary skills do not exist in the hospital environment in which the patient presents, transfer to a hospital with higher available expertise should be considered.

   
   
3. 
   

   The Curious: One of the more difficult set of consults that the neurohospitalist receives belongs to the category of patients who have isolated or atypical neurological symptoms. The patient may have numbness, weakness, movement disorder, speech or visual problems. A working knowledge of the basic physiology in each case can allow the neurologist to be able to distinguish the factitious from the real, and to direct the patient to the appropriate resources.

What is the skill set needed to become a neurohospitalist?

1. 
   

   Neurological knowledge: A good knowledge of the 3 C’s of hospital neurology forms the basis of practice.

   
   
2. 
   

   General medical knowledge:

   
   
   
   
   1. 
      

      Critical care: Neurological patients with critical conditions are looked after in intensive care units (ICUs). Where a neurological ICU is not available, the hospital neurologist often co-manages these patients along with the critical care physician. A working knowledge of the setup and equipment in the ICU is a necessary part of the job.

      
      
   2. 
      

      General inpatient care: Neurological patients may have risk factors for systemic disease. The neurologist should be able to at least recognize the signs and symptoms of a patient with systemic disease that requires immediate attention. Furthermore, some common day-to-day management of inpatient problems may save considerable time and improve efficiency.

   
   
   
   
3. 
   

   Diagnostic skill set:

   
   
   
   
   1. 
      

      Electrodiagnostics: Interpreting electroencephalograms (EEGs) is important to the management of inpatient neurology. Recognition of epileptiform activity, often not subtle, can be taught in a brief period of time. Identifying normal variant prevents unnecessary treatment. Some access to EEG reading, even if remote, can be arranged when the hospitalist does not feel prepared to interpret EEGs. Electromygraphy is used in the diagnosis of several acute neuromuscular diseases but not in the day-to day management of neurological patients.

      
      
   2. 
      

      Imaging: Working knowledge of stroke imaging is also important. Computed tomography (CT) and CT angiograms (CTAs) form part of the acute management of strokes. Radiological input in some cases may delay care. For other purposes, being able to interpret one’s own images is definitely time saving and efficient. Unlike the radiologist who has to be versed in the rare and the subtle, the neurohospitalist’s task is to be familiar with the rudiments of radiological signs as well as those which pertain to our narrower field of interest.

      
      
   3. 
      

      Lumbar puncture (LP): The neurohospitalist is often the last resort after all others have failed to find their way into the cerebrospinal fluid space.

   
   
   
   
4. 
   

   Therapeutic skill set: Administration of tissue plasminogen activator (tPA), familiarity with antiepileptic drugs (AEDs), and ability to manage pain are some of the core therapeutic skills.

What is the neurohospitalist approach?

The neurohospitalist approach combines the neurological method with that of general medicine. Here we present one possible approach to the problems of hospital neurology. The two parameters upon which initial decisions are made are risk and probability. The neurohospitalist would approach the differential diagnoses according to these two parameters:

1. 
   

   Risk: The diagnoses that are potentially catastrophic if not treated in a timely manner should be prioritized. For example, the combination of some facial sensory changes and vertigo is statistically most likely due to relatively benign causes. The patient may have benign positional vertigo and hyperventilate in anxious response to the discomfort caused by the spinning sensation. However, missing basilar artery thrombosis, though a less likely diagnosis, would be unacceptable, so that the initial diagnostic inquiry should be directed toward brainstem vascular pathology in a patient with high enough pretest probability.

   
   
2. 
   

   Probability: The probability of other differential diagnoses is a function of their semiology and presence of risk factors. So, for example, any focal neurological sign in a diabetic smoker in their 60s, however atypical, should prompt further investigation. Atypical symptoms, for example sensory changes, in a young healthy 20-year-old is less likely to prompt urgent imaging. Conversely, a highly typical syndromic presentation, such as one consistent with a classic lateral medullary syndrome, regardless of risk factors, should raise the possibility of vascular pathology in the posterior fossa.