The incidence of depression is reported to be as high as 30 % of stroke survivors [26]. Selective serotonin receptor inhibitors (SSRIs) are the mainstay for treatment of depression and have also been studied for motor recovery in stroke survivors. These trials have reported less disability and neurologic impairment after stroke in both depressed and non-depressed patients treated with SSRIs [27]. Two large prospective randomized control trials are underway to confirm this effect.

Delirium is defined as a waxing and waning of attention, often with altered level of consciousness, disorganized thinking, or both. The incidence of delirium in stroke patients ranges from 13 to 48 % in clinical studies [28]. Delirium has been shown to predict worse outcomes including increased mortality in a variety of disease states including stroke [29, 30]. Patients should be screened for delirium daily during their admission; neurologically injured patients present a unique challenge for applying standard delirium scales. The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) has been validated in the stroke population and is an easy to administer, well-validated bedside assessment [31]. Some of the factors that predict patients who will develop delirium include older age, previous cognitive impairment, and baseline visual or hearing deficits. Patients with right-sided stroke that experience neglect are also more likely to experience delirium compared with infarcts in other territories [32].

Alcohol withdrawal delirium is common. A recent population based survey found that about a third of elderly patients with chronic medical problems drink alcohol regularly and about 6 % have risky drinking behaviors [33]. This reminds the clinician to take a complete social history in all patients hospitalized following stroke.

The treatment of delirium is focused on avoiding factors that may increase confusion including avoidance of benzodiazepines and other centrally acting medications, encouraging day and night orientation, early mobility, providing glasses for those with corrective vision problems, and treatment of dehydration and infections [34]. Antipsychotics, particularly haloperidol, can help to treat symptoms of hyperactive delirium but should only be used when patients’ behaviors are posing a harm to themselves or staff. Atypical antipsychotics, such as risperidone and olanzapine, have been studied and have a safer side effect profile than haloperidol but have not been directly studied in the stroke population [35]. All antipsychotic agents have an FDA Black Box warning for increased mortality for patients over 65 years old and have known cardiac side effects including prolonged QT and therefore should be used sparingly and in low doses only when necessary.

The appropriate acute blood pressure goal in stroke patients is a balance between maximizing cerebral perfusion and protection of the brain and other organs, especially the heart and kidneys, from hyperemia and its complications. The patient’s baseline blood pressure, impaired cerebral autoregulation and physiologic compensatory mechanisms make this ideal target individualized and complex. Some data suggest a U shaped curve associated with outcomes in stroke patients, with both hypertension and hypotension on presentation having worsened outcomes [36]. Other studies describe worsened outcomes with elevated blood in a more linear fashion [37].

The current guidelines from the American Heart Association for the treatment of acute ischemic strokes recommend careful lowering of the blood pressure in the acute setting to keep systolic blood pressure less than 220 and diastolic blood pressure lower than 120 unless there is a compelling reason to lower further for the protection of another organ system such as in myocardial ischemia or aortic dissection. If the blood pressure is higher, then reducing it by 15 % in the first 24 h is recommended. For patients who undergo thrombolytic therapies, a lower blood pressure is advised, specifically less than 185/110 [38].

In the acute setting prior to thrombolysis, attention is placed on lowering the blood pressure to less than 185/110. The current AHA/ASA guidelines recommend attempting to treat with intravenous labetalol with two doses prior to initiating a nicardipine infusion based on a safety study of aggressively lowering blood pressure using either agent prior to tissue plasminogen activator (tPA) [39]. Based on a retrospective study, patients treated solely with nicardipine may reach targeted blood pressure with less dose adjustments and faster than those treated with labetalol [40, 41].

When to further control blood pressure after acute ischemic stroke is more controversial, and this debate has being ongoing for more than 30 years. Many stroke patients were on antihypertensive agents prior to their cerebrovascular event. A randomized control trial of stopping or continuing antihypertensive medications was conducted with 4,071 patients in the first 24 h after stroke. There was no difference in neurologic outcomes at 14 days or 3 months after stroke, and the blood pressure was lowered by 9 mmHg systolic in the control group [42].

For the past two decades, there has been emerging evidence that inhibiting the renin angiotensin aldosterone system (RAAS) has benefits on a variety of cardiovascular outcomes in addition to lowering blood pressure [43]. As a result, these agents, or thiazide diuretics are typically the first choice for secondary prevention when an oral agent is to be started.

Hypotension is occurs at presentation in less than 1 % of patients with acute ischemic stoke [36]. Some of these strokes occur when systemic hypotension is present in a patient in the setting of a fixed intracranial or cervical stenosis. The pattern of stroke is often consistent with a watershed distribution [44]. In general, strokes that present with low blood pressure have a much worse prognosis than those that present with elevated blood pressure [36]. The choice for treatment of hypotension is limited to animal data only, and there is no data to recommend vasopressors in this setting. Occasionally hypotension is due to an effect of an underlying condition, such as an arrhythmia or aortic dissection, which is the etiology of stroke.

Atrial fibrillation or flutter is the known cause of stroke in about 20 % of patients presenting with acute ischemic infarcts. An additional 25 % of patients have a suspected embolic source that cannot be identified (i.e., cryptogenic stroke) [45]. Detection of cardiac arrhythmias influences the secondary prevention strategy moving forward and an aggressive search for such rhythms should be part of routine stroke evaluation. An electrocardiogram is standard for all patients being admitted with an acute ischemic event. While some arrhythmias are detected with the initial EKG, paroxysmal atrial fibrillation is more likely to cause a stroke than persistent atrial arrhythmias, necessitating further cardiac telemetry [46]. The presence of paroxysmal supraventricular tachyarrhythmias other than atrial fibrillation are also an independent risk factor for stoke [47]. Observational studies have shown an additional 7 % of patients with idiopathic etiologies are found to have atrial arrhythmias with repeated EKG’s up to 48 h from stroke [48]. The incidence of atrial fibrillation was found to be about 30 % of patients with implantable cardiac defibrillators that are capable of monitoring atrial rhythm; patients with cryptogenic stroke have a 15–20 % risk of atrial fibrillation with 21–30 days of cardiac monitoring [49–51]. Another study of patients with recently implanted defibrillators that can detect atrial rates found 10 % of patients had an atrial rate of greater than 160 beats per minute sustained for at least 6 min within 3 months of implanting the device; these patients had a higher incidence of stroke and other embolic events in the subsequent 2 years [52].

Clinicians caring for patients with atrial fibrillation can use the validated CHADS2 scoring system helps clinicians determine stroke risk and potential benefit of antiplatelet or anticoagulation agents. The presence of congestive heart failure, hypertension, or diabetes or age greater than 75 are each worth 1 point in the score, and stroke or TIA is 2 points [53]. Anticoagulation has been recommended for CHADS score of 1 or more and aspirin therapy is recommended in patients with 0. In the age of directed anticoagulation agents for non-valvular atrial fibrillation, there may be expanded recommendations for full anticoagulation [54, 55].

The benefit and timing of therapeutic anticoagulation in ischemic stroke patients remains a source of much debate. The AHA recommends starting all patients with stroke and atrial fibrillation on anticoagulation if there are no contraindications [45]. Starting too early may increase the risk of hemorrhagic transformation; starting too late may increase the risk of further ischemic events. The Heparin in Acute Embolic Stroke Trial (HAEST) randomized 449 patients with acute ischemic strokes and presenting with atrial fibrillation within the first 30 h of symptoms onset to either low molecular weight heparin, dalteparin 100 IU/kg, subcutaneously twice a day or aspirin 160 mg orally for the first 14 days after stroke. The primary outcome was functional neurologic outcome. A secondary outcome was hemorrhagic transformation. There was no difference in the functional outcome at 14 days and 3 months but there was a statically significant increase in larger symptomatic hemorrhages in the group that received low molecular weight heparin [56]. A further meta-analysis with more than 22,000 patients included evaluating the risks or benefits of anticoagulation within the first 14 days of stroke and found no net difference in patients who were started on heparin compared with those who were started on aspirin or placebo [57]. With this knowledge, patients are typically started on aspirin on admission or 24 h after thrombolytic therapy and then anticoagulation with a bridge upon discharge. The aspirin may be stopped after the anticoagulation is at a therapeutic level.

Given the similarities in the pathophysiology of cerebral and cardiac ischemia, all patients admitted with cerebral ischemia are at risk for coronary disease. Troponins are used as a marker of cardiac damage but are often elevated in other medical conditions including renal insufficiency, pulmonary embolism, sepsis, hypertensive emergencies, and stroke [58]. About 20 % of patients with acute ischemic stroke have elevated troponins on admission [59]. Older patients and those with larger strokes are more likely to have an elevated troponin as well as serum creatine kinase level, and these markers are associated with an increase in stroke mortality [60].

Determining the clinical significance of elevated cardiac markers can be challenging. One study described 834 consecutively admitted stroke patients with troponins measured at admission; if elevated, an EKG was performed and troponins were measured again 3 h later. The patients were divided into two groups, those with greater than 30 % increase in the troponin level and those levels that remained constant. There were no patients in the constant group that met established criteria for myocardial infarction (MI) but half of the patients with an increase in the troponins did. Of all 834 patients, only 29 (3 %) had concurrent cerebral and cardiac ischemia, consistent with previous studies [58].

During a time of decreased cerebral perfusion, hypoxemia is the most rapid cause of cellular death. There is limited data for the appropriate level of oxygenation for patients experiencing a stroke. It is reasonable to keep oxygen saturation greater than 94 % during the time of acute ischemia, which can be achieved in the vast majority of patients with supplemental oxygen administered via nasal cannula [38].

When oxygenation cannot be maintained with less invasive measures, intubation is indicated. Intubation is also indicated for patients who have a decreased level of consciousness with a Glasgow Coma Scale (GCS) of less than 10 and is required for patients with a GCS of less than 8, as they are unlikely to maintain appropriate ventilation and oxygenation. If the partial pressure of oxygen becomes less than 60 mmHg with supplemental oxygen or if the partial pressure of carbon dioxide becomes greater than 60, intubation is also recommended [38]. For stroke patients who require ventilation based on these needs, there is a 70 % 1-year mortality [61].

Hypercapnic respiratory failure may occur in stroke patients with decreased levels of consciousness and inability to protect their airways. Additionally, obstructive sleep apnea (OSA) is an independent risk factor for ischemic stroke and leads to an elevated CO₂ level at baseline; any structural neurologic damage, including stroke, can worsen the hypercarbia [62]. When the carbon dioxide level rises, the systemic and cerebral vasculature dilates. For patients who are dependent on collateral flow to perfuse an area of tissue experiencing ischemia, this change is particularly dangerous since dilation of normal blood vessels shunts blood away from tissues experiencing ischemia [61]. In one study of patients with a history or reported symptoms of OSA who were admitted with stroke, those treated with noninvasive mechanical ventilation had a 6 % absolute decrease in mortality during the admission [63].

The decision to extubate a neurologically injured patient is often more difficult than the decision to intubate. For patients who were intubated for a primary lung problem, there are studies to support when a patient is likely to be successful extubated. However, there is limited data for neurologically injured patients who are often incapable of following verbal commands. The decision falls on the global impression of the physicians and respiratory therapist. Typically if bulbar functions are preserved, then extubation is attempted. Fifteen to 35 % of intubated stroke patients will fail to wean from the ventilator and require a tracheostomy and long-term ventilation. Some small studies have shown benefit of early tracheostomy, similar to larger studies in the general ICU population [64].

One of the most common medical complications of stroke is pneumonia. Aspiration is typically the mechanism, presenting as an opacification of the gravity dependent areas of the lungs on radiographs. See Fig. 6.3. Reported incidences range from 4 to 20 % of stroke patients in the hospital [65]. The presence of pneumonia in a hospitalized stroke patient increases the cost of hospitalization by on average over $27,000, adding an additional $459 million dollars to health care cost in the United States each year [66, 67]. The development of pneumonia increases mortality by five times in acute stroke patients [65].

Several studies have described predictors for the development of stroke-associated pneumonias. Pneumonias are associated with older patients and those who have worse strokes, higher NIHSS scores, and those that involve the brainstem [68]. Endotracheal intubation is a major risk for pneumonia, especially when prolonged ventilation is required. Even brief intubations for endovascular procedures have been associated with an increased risk of pneumonia [69].

The presence of dysphagia is the single biggest risk factor for aspiration pneumonia. The AHA’s Get With the Guidelines Program has placed an emphasis on screening for dysphasia in all stroke patients prior to any oral intake including fluids or medications [65]. This type of screening has been shown to aid the early identification of patients at risk so that they may undergo modifications to reduce the incidence of aspiration pneumonia, in turn decreasing length of stay, cost, and mortality rates [70]. Any health-care provider may perform and document testing for dysphagia [65]. Even though there is no standard assessment for dysphagia, there are several studies that have looked at key features of the tests including direct observation of patients swallowing water. Other important components of an assessment may include observing for an abnormal volitional cough, an abnormal gag reflex, dysarthria, dysphonia, cough or throat clearing after swallowing, and voice change after swallowing [70]. If there is a suspicion that the patient may not be able to swallow safely, the patient should not take anything orally until further assessed by a speech and language pathologist.

The treatment of stroke patients with aspiration pneumonia is no different than other hospitalized patients. Nasal and oral floras are the most commonly found bacterial pathogens. Empiric antibiotics should cover both Gram positive and Gram-negative organisms [68]. If the patient has been hospitalized for two or more days in an acute care facility within 90 days; lives in a nursing home or long-term care facility; has attended a hospital or hemodialysis clinic, or has received intravenous antibiotic therapy, chemotherapy, or wound care within 30 days of infection, then their pneumonia is considered health care-associated. For these patients, empiric coverage should include antibiotics targeting multidrug resistant organisms according to the local bacterial resistance patterns [71].

Food choices can influence many risk factors for stroke such as hypertension, diabetes and dyslipidemia, and patients who have chronic medical problems are often malnourished. The nutritional status of patients who are admitted with acute ischemic stroke likely changes rapidly.

Once dysphagia is recognized, tube feedings should begin. The result of being underfed after a stroke can have serious consequences including increased infections and bedsores and has an association with worsened neurologic outcome [72]. Nutritionists are charged with the challenge of determining the resting metabolic needs for each patient, and there are a several equations that can be used to determine the appropriate caloric needs in critically ill states. Each disease process uses a corrective factor to help predict the increased demands due to illness. For example, sepsis, malignancy, burns and trauma require remarkable additional nutritional support due to the high catabolic demands of these states. There is very little data on the metabolic needs of patients with acute stroke. One small prospective study of 27 stroke patients found that 40 % of them had a negative nitrogen balance when evaluated using urine nitrogen excretion, suggesting they were being underfed by using a standard equation for calculating nutritional needs [73]. In practice, calculating the nitrogen balance or oxygen consumption is rarely done; nutritionists more often rely on end markers of nutritional status such as serum pre-albumin or C-reactive protein levels.

While some patients regain their ability to swallow, others are left with long-standing dysphagia or impaired consciousness and are unable to meet their nutritional needs orally. Patients with higher NIHSS scores, larger infarct size, those who are more critically ill, or require long term ventilator support are more likely to need prolonged nutritional support [74]. For these patients, surrogates are often required to make a decision regard long-term artificial nutrition. Some wish to withhold artificial nutrition and succumb to the stroke; others elect to have a percutaneous endoscopic gastrostomy (PEG) tube placed. This decision-making involved is often complex and reliant on imprecise data regarding long-term recovery prognostication.

When patients are admitted with acute ischemic stroke, intravenous fluids are often quickly administered for a variety of reasons. These patients are not allowed fluid intake by mouth until swallowing function can be assessed and hydration may be desired prior to receiving intravenous contrast dye. Insuring the body has enough intravenous volume can theoretically improve perfusion to the ischemic brain tissue.