Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Intensive care units (ICUs) represent a triumph of medicine: the ability to support and replace a large number of bodily functions. And yet, these actions are not cost free. Nearly every intervention has side effects and risks, all of which must be mitigated. While rapid decision-making is often possible and desired in the management of emergency center or ward patients, this approach can be prone to errors in the ICU environment and must be used in the appropriate clinical situations. In much of academic medicine, there are two overarching approaches to patients, those being a “problem-based” approach and a “systems-based” approach. In the ICU, the “systems-based” approach is preferred because of the need to thoroughly consider the patient’s needs and status from every angle. And this too is the layout of this chapter, covering each organ system and their common dysfunctions and treatments.
In the United States, critical care medicine has numerous educational pathways to include surgery, anesthesia, and internal medicine fellowships. Similarly, many larger hospitals have separate ICUs focused on patient subsets: cardiothoracic ICUs, medical ICUs, neuroscience ICUs, surgical ICUs, transplant ICUs, etc. Other smaller hospitals dispense of any such distinction. Each specialty brings a unique perspective and advantage to the field of critical care medicine. Surgical intensivists possess a strong grasp of the expected course of surgical diseases as well as comfort with many commonly needed procedures such as bronchoscopy, esophagogastroduodenoscopy, percutaneous tracheostomy, etc.
Finally, it must be emphasized that providing quality ICU care is not solely dependent on the intensivist, but also on the entire multidisciplinary ICU team. To the greatest extent possible, joint guidelines and standardization of care in conjunction with nursing, pharmacy, physical therapy, respiratory therapy, and other members of the healthcare team should be developed. It is well documented that this standardization of care improves outcomes. Although exceptions may exist to every guideline, having standard processes with the ability to adjust improves the quality of care. These guidelines are often unique to an institution based on best evidence and local expertise, capabilities, etc.
Pain is a ubiquitous but often underrecognized and inadequately treated symptom for the critically ill patient. In the ICU setting, sources of pain include traumatic injuries, burns, surgical wounds, underlying illness, and/or noxious stimuli (e.g., tracheal intubation, mechanical ventilation, invasive lines, etc.). Pain is a source of fear and anxiety that, if not properly addressed, contributes to physiologic alterations that can negatively impact patient outcomes and delay recovery. Although the sensation of pain is subjective, there are several measuring tools and scales to quantify pain as an objective data point that can be treated and easily reassessed for improvement. In patients who are able to communicate, the visual analog scale and numeric rating scale are used and provide reliable means for assessment and treatment; however, noncommunicative, altered, or comatose patients must rely upon the observant and attentive physician or nurse to identify the visual cues and physiologic alterations (e.g., tachycardia, hypertension, tachypnea, diaphoresis, etc.) that coincide with uncontrolled pain. If objective data are unavailable, one is better off to presume the presence of pain and provide appropriate treatment.
Opioids have traditionally been the first-line therapy for pain as they are centrally acting, highly effective for nonneuropathic pain, have a rapid onset, and can be administered via multiple routes (enteral, intravenous [IV], transdermal, etc.). Unfortunately, opioids are often overprescribed, leading to untoward side effects, patient abuse, and addiction. Deleterious side effects including nausea, vomiting, pruritis, sedation, respiratory depression, and bowel dysfunction result in unnecessary complications, prolonged hospital length of stay, and increased morbidity. Prolonged use can also result in physiologic dependence, tolerance, and opioid-induced hyperalgesia. Enhanced recovery after surgery protocols and the practice of utilizing multimodal therapies that work synergistically to alleviate pain and provide sedation are paramount in combating the opioid epidemic that currently plagues the United States. Combining the ability to perform regional anesthetic blocks along with nonopioid adjuncts with varying mechanisms of action such as acetaminophen, nonsteroidal anti inflammatory agents, gabapentinoids, tramadol, and muscle relaxants results in more effective analgesia and lower overall opioid requirements. The use of multimodal pain medications has become the standard of care to reduce opioid reliance and should be implemented widely.
In the same manner that pain is assessed and treated, agitation is quantified using various scoring systems to include the Richmond Agitation-Sedation Scale. When a patient is agitated, it is important that a thorough assessment of any underlying treatable etiologies is completed prior to initiating a pharmacologic intervention. Clinical guidelines have been recently updated and published by the Society of Critical Care Medicine as to the management of the agitated ICU patient. The most commonly used sedatives in the ICU include propofol, benzodiazepines, and dexmedetomidine. When deciding on which agent to use, considerations must include the expected duration and depth of sedation required, comorbidities, which may affect pharmacokinetics or metabolic clearance, and potential drug interactions with other medications the patient may already be receiving.
Propofol is a highly protein-bound, lipophilic molecule that is frequently utilized in the ICU. The exact mechanism of action is not fully elucidated, but it is thought to potentiate γ-aminobutyric acid (GABA) receptors resulting in amnesia but without analgesia. Because propofol is highly lipophilic, it readily traverses the blood-brain barrier and results in rapid onset of action in less than 1 minute as well as a short duration of action as it is quickly distributed into peripheral tissues and readily metabolized by the liver. Rapid clearance of propofol makes it the ideal drug for daily sedation weans to assess a patient’s neurologic status. Hypotension and cardiovascular depression are commonly seen with propofol administration, especially in the hypovolemic patient. The most serious potential side effects of propofol infusion include pancreatitis, hypertriglyceridemia, and propofol infusion syndrome. Propofol infusion syndrome is a rare complication associated with prolonged and/or high doses of propofol administration that is thought to be caused by mitochondrial respiratory chain inhibition or impaired mitochondrial fatty-acid metabolism. Clinical features include bradycardia, rhabdomyolysis, hyperlipidemia, hepatomegaly, and renal failure. Early recognition is critical because it has a high mortality rate and treatment is fairly limited to supportive care and discontinuing propofol administration.
Benzodiazepines are GABA receptor agonists that produce anxiolysis at low doses and sedation, amnesia, and cardiorespiratory depression at higher doses. The most common benzodiazepines used in the ICU setting are midazolam, lorazepam, and diazepam. Both midazolam and lorazepam can be administered as continuous infusions, whereas all three can be administered intermittently and have rapid onsets of action. The duration of action for each of these drugs is relatively short on initial administration as they are lipophilic and readily distributed to peripheral tissues. However, repeated administration or prolonged infusions result in saturation of adipose tissues and prolonged sedation even after discontinuation. Age, obesity, and altered hepatic or renal function all can alter benzodiazepine clearance and must be considered when deciding on the appropriate medication. The side effects of benzodiazepine administration include cardiovascular and respiratory depression, delirium especially in the elderly, as well as propylene glycol toxicity. Propylene glycol is used as the solvent for IV lorazepam and diazepam, and toxicity is rare. It is characterized as an anion gap metabolic acidosis, renal failure, and eventual multisystem organ failure. Treatment is limited to dialysis and discontinuing the offending agent.
Dexmedetomidine is a centrally acting α2- adrenoreceptor agonist that binds to receptors within the locus ceruleus to provide sedation and anxiolysis and receptors in the spinal cord to provide for analgesia. It has a relatively quick onset of action between 5 and 15 minutes and a duration of action between 60 and 120 minutes. Dexmedetomidine is hepatically metabolized by glucuronidation and the cytochrome P450 system, and, as such, alterations in dosing may be required when other cytochrome P450 altering medications are being given. One of the main advantages of dexmedetomidine is that it has no effect on respiratory drive, so patients can be sedated without the need for mechanical ventilation. Studies have also demonstrated that dexmedetomidine usage results in extubation almost 2 days sooner compared to midazolam use. The adverse effects of dexmedetomidine use include bradycardia, hypotension, atrial fibrillation, and reflex hypertension with abrupt cessation of administration. Currently, the US. Food and Drug Administration (FDA) states the use of dexmedetomidine for initial sedation should be limited to 24 hours with the idea that prolonged use can exacerbate reflex hypertension when stopped.
It is not uncommon for ICU patients to receive continuous infusions of sedatives as it increases patient comfort and provides a reliable and consistent level of sedation. That being said, randomized trials have demonstrated that protocol-driven daily sedation interruption decreases the duration of mechanical ventilation and length of stay in the ICU and improves the ability to perform daily neurologic assessments. Sedation holidays are also cost-effective as they decrease amounts of medication administration and reduce unnecessary diagnostic testing obtained to evaluate a patient’s neurologic status.
Assessing and accurately diagnosing alterations in a patient’s mental status is an often overlooked and underestimated aspect of critical care. Alterations in mental status among the critically ill can be hard to recognize and include a spectrum of disorders including delirium, encephalopathy, and coma. Patients who are elderly and those with preexisting mental illness or cognitive impairment are at increased risk for developing alterations in mental status when critically ill. When a patient has an acute change in mental status, it is important to rule out any organic or potentially reversible causes because the diagnosis of delirium of critical illness is otherwise a diagnosis of exclusion. Some examples of organic causes that can lead to an acute change in mental status include hypoxia, hypercapnia, hypoglycemia, medication side effects or withdrawal, infection, metabolic derangements, cerebrovascular accidents, seizures, and changes in intracranial pressure.
When a patient becomes confused, they lose their sense of orientation and are no longer able to identify who they are or where they are and/or will lack a general perception of time. Delirium is defined as confusion along with a disturbance in focus, attention, or awareness that occurs over a short period of time. The delirious patient will also develop cognitive deficits such as memory loss and difficulty with language or visuospatial abilities. Identifying and trying to reverse the effects of delirium are important because those who develop ICU delirium have an increased mortality, residual functional disabilities, and higher rates of dementia after discharge. In general, hyperactive delirium is easier and more quickly detected as these patients are agitated and restless. However, the opposite is true for patients who develop hypoactive delirium, as they are quiet, inattentive, and lethargic, which can be easily overlooked and ignored. The elderly and patients with prior mental illness or decreased cognition are at higher risk for developing hypoactive delirium and should be evaluated carefully.
Assessment for delirium has been standardized by utilizing either the Confusion Assessment Method for the ICU or the Intensive Care Delirium Screening Checklist. Both assessment tools have been validated in the ICU setting and can also be utilized for the mechanically ventilated patient. Utilizing either one of these assessment tools on a regular and scheduled basis allows for an objective assessment of a patient’s mental status and increases the chances for early detection of delirium.
When a diagnosis of delirium has been established, prompt intervention is important in order to begin the process of reorienting the patient and identifying any precipitating factors. The basis of treatment includes identifying and reversing any potential organic causes and providing supportive measures that bring the patient’s behavioral alterations back to baseline. Supportive measures serve to limit abnormalities around a patient’s environment that would not necessarily be present and to provide a calm, secure, and safe environment. Modifiable behavioral and environmental factors include limiting circadian rhythm abnormalities and maintaining normal sleep-wake cycles by opening blinds or shades during the day and turning off lights at night, reducing ancillary and unnecessary noises or noxious stimuli, avoiding the use of restraints when safe, mobilizing the patient, providing the patient with their eyeglasses or hearing aids if needed, and having family or friends help to reorient and reassure the patient. New-onset delirium has been documented to result in a statistically significant increase in the 90-day mortality rate, which increased in a graded manner when these patients were exposed to noxious stimuli, placed in restraining devices, or developed hospital-acquired conditions such as falls or pressure ulcers. Altering a patient’s environment and normalizing their daily lives can be readily done and drastically affects mortality and outcomes.
If patient behavior interferes with their care or they are disoriented to the point of potentially harming themselves or others, low-dose antipsychotic medications such as haloperidol can be safely utilized. Benzodiazepine use should be limited to cases of sedative or alcohol withdrawal as they can precipitate delirium, especially in the elderly population. When sedative infusions are necessary, dexmedetomidine has been found to decrease the rate of delirium by more than 20% when compared to using midazolam, although no one agent is ideal and the proper choice of sedative needs to be made on a case-by-case basis. The medical workup of delirium should include a thorough review and cessation of any unnecessary medications, physical and neurologic examination with possible computed tomography (CT) imaging of the brain, and evaluation for potential infectious etiologies with sampling of cerebrospinal fluid with lumbar puncture when clinically indicated. If no obvious etiology is found, one must consider magnetic resonance imaging (MRI) of the brain, electroencephalogram, measuring drug or toxin levels, or prophylactic supplementation with vitamin B 12 or folate if there is concern for alcoholism.
Often confused with delirium, which describes the mental manifestations of the disease, encephalopathy is a term to describe an altered mental state per the underlying pathophysiologic process. Encephalopathy develops as a syndrome of overall brain dysfunction and can occur from many organic and inorganic causes that directly induce brain injury or remotely affect the brain from other systemic causes. Encephalopathy can be described as acute or chronic based on the timing and potential reversibility of the syndrome. Chronic encephalopathy is slow to progress and results in structural changes in the brain that are usually irreversible. Acute encephalopathy can potentially be reversed with a return to baseline functional status if the inciting insult is removed or treated in a timely fashion. Examples of causes of encephalopathy include chronic traumatic encephalopathy, Wernicke-Korsakoff syndrome, heavy metal poisoning, electrolyte abnormalities, liver failure, medications, and sepsis.
At the furthest end of the spectrum of altered mental status is coma. A comatose patient is unarousable and unaware of their environment. Most cases of coma that present to the emergency department are due to trauma, cerebrovascular accidents, metabolic derangements, medications, seizures, and infections. The Glasgow Coma Scale (GCS) is a neurologic assessment tool that provides a reliable and objective measurement of a patient’s conscious state. It is composed of three elements (eye, verbal, and motor responses) and scaled from 3 to 15. The lowest possible score is 3 when a patient has no response to stimuli, and the highest possible score is 15 when a patient is fully awake and interactive. When comatose patients have a GCS score of 8 or lower, their brain injury or dysfunction is deemed severe and they should be intubated as they cannot reliably protect their airway. As is the case with most causes of altered mentation, treatment for a comatose patient is based on identifying and treating any reversible organic causes and providing supportive care.
One of the leading causes of disability in the United States, traumatic brain injury (TBI) is a devastating and life-altering event that often leads to heavy familial and socioeconomic burden. It is the most common cause of death and disability in people between the ages of 15 and 30, and the most severe cases result in prolonged periods of coma and unresponsiveness. The severity of TBI is classified using GCS scoring, as it is simple, reproducible, and a prognosticator for outcomes. Neuroimaging is utilized to identify pathologic injuries such as skull fractures, cerebral contusions, hemorrhage or hematoma, and diffuse axonal injury. Primary and secondary brain injuries are the two phases during which neuronal injury occurs. Primary brain injury occurs during the initial insult, whereas the focus of critical care management of TBI is to limit secondary brain injury. Secondary brain injury is a consequence of pathologic and physiologic alterations that manifest after the initial injury usually due to concomitant multiorgan injury. Examples of causes of secondary brain injury resulting in further neuronal injury and death include ischemia, hypoxia, hypotension, cerebral edema, acidosis, and elevated intracranial pressure.
Management and treatments are focused on optimizing intracranial pressure and blood pressure in order to maintain adequate cerebral perfusion, avoiding hypoxia, and maintaining normothermia and normoglycemia. Often, antiseizure medications are prescribed as antiseizure prophylaxis, although the ideal medication, dosage, and duration of treatment are not clearly established. Recovery after TBI can be a prolonged and lengthy process. Approximately 10% to 15% of patients with severe TBI (GCS <8) are discharged in a vegetative state, with approximately 50% regaining consciousness by 1 year. Medication adjuncts to accelerate and maintain long-term recovery are sparse. A randomized control trial investigating the use of amantadine has shown that, over a 4-week treatment period, patients showed accelerated recovery; however, over the long-term, there was no difference between the treatment and placebo groups. Long-term prognosis and recovery are variable and difficult to predict and are highly dependent on the severity of TBI, patient comorbidities, and postinjury complications.
Cardiovascular issues commonly encountered in the ICU can be separated broadly into three categories: cardiac arrhythmias, shock, and myocardial ischemia. All of these areas are affected in the setting of primary cardiac dysfunction such as heart failure or a myocardial infarction (MI) and may be affected by extracardiac disease processes (e.g., pulmonary embolism [PE] can lead to right heart failure, hyperthyroidism can lead to arrhythmias, and increased metabolic demand can lead to cardiac ischemia). This section begins by briefly covering normal physiology, continues on to disorders encountered in these three areas and their treatments, and concludes with invasive and noninvasive methods of monitoring the heart and fluid status.
The heart is a two-pump circuit in sequence. All blood that goes out of the right ventricle to the pulmonary circulation must then be pumped out of the left ventricle to the system circulation, with important exceptions that arise in the setting of congenital cardiac anomalies and a minor exception from bronchial arteries and veins. There is a vast difference in resistance between the pulmonary vascular bed faced by the right heart and the systemic vascular bed faced by the left heart, with the pulmonary pressures and resistance being significantly lower. The heart’s ultimate function is to supply oxygenated blood to the tissues of the human body. This ability is captured by the oxygen delivery (Do 2 ) equation:
CO is defined as the blood flow put out by the heart per unit time, typically expressed in liters/min. An average, CO is 4 to 6 L/min. Overall, one observation is evident: the primacy of the hemoglobin and oxygen saturation in determining the carrying capacity of oxygen by the blood and the relatively trivial contribution of dissolved oxygen.
Vascular resistance is the collective resistance of all vessels including arteries and veins against the flow of blood, and there are two such resistances: the systemic vascular resistance (SVR) faced by the left ventricle and the pulmonary vascular resistance (PVR) faced by the right. The relationship between flow, pressure, and resistance is defined by Ohm’s law:
These relationships guide decisions about fluid and vasopressor therapy, which are elucidated later. The lower PVR consequently requires a smaller right heart muscle volume (lower pulmonary artery pressure) to supply the lungs with the same CO as the rest of the body. This means that the right heart is unable to maintain its CO in the face of large, acute rises in PVR, and this lack of reserve has significant consequences both in trauma surgery when a trauma pneumonectomy is performed and in thromboembolic disease in the lung. In both these situations, a sudden rise in PVR, particularly in the context of a low preload, can lead to cardiovascular collapse.
The heart muscle is supplied primarily by the coronary arteries. During systole, the subendocardial vessels experience retrograde flow, and thus the heart is primarily supplied during diastole. This has important implications for cardiopulmonary resuscitation, as failure to allow for full recoil of the chest may reduce blood supply to the subendocardium during resuscitation.
The heart’s rhythm is controlled by pacemaker cells. The primary node is the sinoatrial node, which is influenced by sympathetic stimulation from the sympathetic trunk mainly arising from the T1–T4 spinal levels, which stimulate positive chronotropy. Parasympathetic stimulation results in negative chronotropy and is mediated via the vagus nerve. The heart has a series of escape pacemakers, which are, in order, the atria, the atrioventricular node, and the ventricles themselves. As long as the sinoatrial node paces above the intrinsic rate of these escape pacemakers and as long as those impulses are transmitted through the atrioventricular node and to the ventricles, impulses from the sinoatrial node control the heart rate.
Postoperative atrial fibrillation is common, occurring in 8% of major surgeries and 45% of cardiac ones. Recent work has noted that the risk of thromboembolism after the development of atrial fibrillation after noncardiac surgery is similar to that of patients with nonvalvular atrial fibrillation. However, while the 2016 European Society of Cardiology guidelines now recommend anticoagulation for postoperative atrial fibrillation following cardiac surgery, they do not address other major surgical procedures.
For atrial fibrillation with rapid ventricular response, the acute management is dependent on the hemodynamic stability of the patient. If the patient is acutely unstable, immediate electrocardioversion is mandated as it is for any tachyarrhythmia causing acute instability. In the context hemodynamic stability, pharmacologic methods should be employed, including amiodarone, beta blockers, and calcium channel blockers. Amiodarone is favored in the context of heart failure, as it does not depress cardiac function as is done by beta blockers and calcium channel blockers. That being said, a recent retrospective review documented that metoprolol had the greatest success in the treatment of acute atrial fibrillation, defined by rate control without the need for a second agent. Other studies have noted more rapid control of atrial fibrillation with diltiazem in both the emergency department and the ICU but also noted an increased rate of hypotension with its use relative to amiodarone. More recent data from the emergency medicine literature suggests that procainamide may be an efficacious option for cardioversion of atrial fibrillation (Ottawa Aggressive Protocol). Acutely, digoxin alone is not recommended owing to its slow onset and comparative lack of success in controlling atrial fibrillation.
Two points of caution should be noted. Traditionally, it was held that cardioversion of an atrial rhythm into sinus could safely occur up to 48 hours after initiation of the rhythm. Thereafter, either a transesophageal echocardiogram to verify lack of clot formation in the left atrium or 4 weeks of anticoagulation were recommended prior to cardioversion, except in cases of acute hemodynamic compromise. However, recent work suggests that the safe period of cardioversion may be much shorter, as little as 12 hours. While all drugs used to treat atrial fibrillation may induce cardioversion into sinus rhythm, amiodarone is especially prone to do so. Thus, it should be used with caution in patients with longer time periods of atrial fibrillation due to its higher tendency to result in a return to sinus rhythm. Second, in patients with an accessory pathway, such as Wolf-Parkinson-White syndrome, atrial fibrillation with preexcitation may develop. The use of calcium channel blockers, beta blockers, amiodarone, and digoxin is contraindicated in such instances due to the risk that following suppression of the atrioventricular node, the accessory pathway will result in an exacerbated tachycardia. In this scenario, ibutilide or procainamide is recommended by the most recent guidelines.
Multifocal atrial tachycardia is most commonly associated with hypomagnesemia. Pulmonary insufficiency, hypokalemia, and coronary artery disease are other known precipitating factors. If correcting hypomagnesemia and hypokalemia (in that order) are not effective, then beta blockers and calcium channel blockers should be tried. Interestingly enough, data show that an empiric push of 6 mg of IV magnesium sulfate terminated multifocal atrial tachycardia 88% of the time, regardless of serum magnesium levels, a result possibly explained by a systemic deficiency of magnesium with normal blood levels.
Atrial flutter is commonly caused by the same disorders that give rise to atrial fibrillation. It is also not infrequent following the treatment of atrial fibrillation with amiodarone. It is an unstable rhythm and has the potential to spontaneously degenerate into atrial fibrillation or to revert to normal sinus rhythm, particularly if the underlying factors have been addressed. The management of atrial flutter is similar to that of atrial fibrillation, with rate and rhythm control options, but electrocardioversion is the preferred therapy. Antiarrhythmic drug therapy is also an option and may be selected for stable patients who are too high risk to undergo the sedation that would typically be required prior to electrocardioversion. Additionally, ibutilide is FDA approved for the conversion of atrial flutter to normal sinus rhythm and has shown superiority to amiodarone and procainamide in this setting. It should be remembered that all antiarrhythmic drugs have proarrhythmic tendencies and ibutilide is no exception with a significant risk of torsades de pointes. It must be used with caution in patients at high risk for torsades de pointes, and patients should be in a monitored setting following its use. IV magnesium given alongside ibutilide both enhance its ability to break atrial flutter and prevent torsades de pointes.
There are numerous other subtypes of supraventricular tachycardias, of which the most common is atrioventricular nodal reentrant tachycardia. The pathophysiology is a reentrant circuit. While sometimes difficult to distinguish from sinus tachycardia and ventricular tachycardia, blocking the atrioventricular node by using vagal maneuvers can reveal the underlying rhythm. Agents acting at the atrioventricular node (adenosine, beta blockers, and calcium channel blockers) are all options to terminate these rhythms. It is important to distinguish a paroxysmal supraventricular tachycardia with a block from a ventricular tachycardia, as both may present with a widened QRS. A history of heart disease or operation portends a ventricular tachycardia. Furthermore, adenosine will aid in distinguishing the rhythms, as it will stop a supraventricular tachycardia but not a ventricular tachycardia. Additionally, ventricular tachycardia will not respond to beta or calcium channel blockers.
Ventricular arrhythmias are rare in young patients and those without a history of heart disease. Options for treatment in stable patients include lidocaine, amiodarone, and procainamide; these are also adjuncts to consider if initial defibrillation for stable or unstable patients fails to convert a ventricular tachycardia into a sinus rhythm. Additionally, a recent study demonstrated the superiority of procainamide to amiodarone in the conversion of stable ventricular tachycardia. Seeing as procainamide is a therapeutic option for atrial fibrillation with rapid ventricular response, supraventricular tachycardias, and ventricular tachycardias, it may be considered a go to option if one is uncertain of the rhythm. While some patients with ventricular tachycardia may appear stable, they are at high risk for sudden deterioration and must be monitored closely and treated expeditiously.
Fundamentally, a monomorphic ventricular tachycardia indicates that ectopic beats are arising from one often ischemic focus in the ventricles, typically secondary to coronary artery disease. Polymorphic ventricular tachycardia indicates either multiple foci of ectopic beats or a more commonly global dysfunction. The latter is a rhythm classically known as torsades de pointes, a feared rhythm for which a predisposing factor is a prolonged QT interval. This can be caused by a variety of drugs and by an inherited condition. Ironically, many antiarrhythmics, including procainamide, lidocaine, and ibutilide, prolong the QT interval, as do commonly used antipsychotic drugs such as haloperidol. The treatment is focused on unsynchronized cardioversion if unstable and aggressive administration of magnesium.
While the treatment of tachycardia involves extensive pharmacologic options backed up with electrical cardioversion, the treatment of bradycardia focuses more on electrical pacing, either transcutaneous or transjugular, as well as reversing the underlying cause. Such causes of bradycardia include acute spinal cord injury, MI, hypoxia, and various toxicologic states, as well as global dysfunction from severe sepsis. Bradycardia is most prevalent in general ICU populations, as it is a sign of profound cardiac dysfunction and sometimes a periarrest rhythm. The acute management includes atropine, epinephrine, and pacing; however, these are temporary supports (apart from spinal cord injury), and the priority must be reversal of the underlying cause.
Sinus bradycardia may be a normal resting rhythm for many fit, young individuals, and no treatment is required. It may also be seen in profound shock states, as a periarrest rhythm. Sinus bradycardia secondary to a spinal cord injury can occur in high (cervical) spinal cord injuries and should be treated with atropine and concomitant vasopressor therapy to treat neurogenic shock as required. Similarly, sinus bradycardia associated with MI, commonly seen in inferior wall infarctions due to involvement of the sinoatrial node, can be treated with atropine as well. Sinus bradycardia secondary to the use of dexmedetomidine requires a different mindset. If this occurs, treatment of the resultant hypotension with vasopressors is required until the drug wears off, and/or pacing may be required if the hypotension is significant and unresponsive to moderate vasopressor use. Atropine and epinephrine will be ineffective with this etiology due to the α1-antagonist effect of dexmedetomidine.
Junctional or ventricular bradycardia is due to profound atrioventricular node dysfunction. There is no atrial activity apparent. Treatment with temporary pacing may be required if the CO falls. Atropine is not effective in this scenario because it acts upon the dysfunctional atrioventricular node.
The most severe hemodynamic alteration is shock, which is a condition of circulatory failure resulting in end organ dysfunction secondary to reduced perfusion. A shock state itself is indicated by signs of end-organ dysfunction such as rising lactate, altered mental status, falling urine output, and liver enzyme elevation. Each form of shock demands different responses, and when different forms of shock are combined, the appropriate management can be challenging. There are numerous forms of shock, classified broadly into four categories: distributive, hypovolemic, cardiogenic, and obstructive.
Neurogenic shock occurs secondary to the loss of sympathetic tone after a spinal cord injury. The overall etiology is decreased vascular resistance; however, there are two variants based on the location of the spinal cord injury. In lower spinal cord injuries, below C5, the hypotension causes an appropriate reflex tachycardia. In higher spinal cord injuries, C5 and above, the patient is often bradycardic, in that the heart does not respond appropriately to the increased vagal tone due to the lack of sympathetic innervation to the heart. This results in a “warm” shock. Treatment consists of vasopressor support to maintain blood pressure. While norepinephrine is considered the first-line therapy, no agent has been proven to be superior.
Of note, neurogenic shock does not equate to spinal shock. Spinal shock is not a hemodynamic phenomenon and results in the temporary loss of reflexes following a spinal cord injury, most commonly the bulbocavernosus and cremasteric reflexes. It is temporary in nature.
Septic shock arises from inflammatory mediators released by the body in response to bacterial or fungal pathogens and is considered a dysregulated immune response. Traditionally, sepsis was defined as a systemic inflammatory response syndrome (SIRS) accompanied by a suspected source of infection, with SIRS being defined as derangements in two or more of four parameters: white blood cell count, temperature, heart rate, and respiratory rate. Septic shock was then sepsis unresponsive to fluid resuscitation requiring vasopressor support to maintain the blood pressure. While still commonly used, these definitions have been superseded by new guidelines that base the definition of sepsis and septic shock on sequential organ failure assessment (SOFA) scores. Specifically, organ dysfunction represented by an increase in SOFA score of 2 or more represents sepsis, while sepsis with hypotension despite fluid resuscitation and/or a serum lactate greater than 2 despite a lack of hypovolemia represents septic shock.
Previous sepsis management guidelines included immediate lactate measurement (and other laboratory values), an empiric 2-L fluid bolus, empiric antibiotic administration following cultures being drawn, appropriate measurement and titration of fluid resuscitation to central venous pressure and central venous oxygen saturation, and potentially a Swan-Ganz catheter to accurately capture hemodynamic variables. This was known as an early goal-directed therapy. Subsequent randomized controlled trials suggest that this bundle has no benefit; however, proponents maintain that these studies are flawed in that much of early goal-directed therapy has become standard of care. This issue is even more complicated as many aspects of early goal directed therapy have been indoctrinated as quality metrics, meaning failure to comply results in the hospital or clinician being penalized. What is widely accepted is that early recognition and administration of antibiotics lead to increased survival, although this must be balanced against the risk of increasing antibiotic resistance and the harms associated with administering antibiotics to patients who do not require them. Early source control of the infection is also positively associated with outcomes.
While it may seem logical that vasopressors are the first choice for septic shock, given that the derangement is systemic vasodilatation, this is counterbalanced by the fact that a critical portion of the pathophysiology of septic shock is leakage of fluid into the interstitial spaces due to an increase in vascular permeability. Consequently, increasing vasoconstriction alone will not by itself reverse the shock state. Traditionally fluid resuscitation followed by vasopressor therapy only when fluids no longer produced an increase in blood pressure (fluid responsiveness) has been standard. This is currently being studied in a major randomized controlled trial.
The most recent updated guidelines for the overall management of sepsis were released by the Surviving Sepsis Campaign in 2017. It controversially continued to recommend an empiric crystalloid bolus of fluid to all patients. Less controversially, it is recommended continued close monitoring, a target mean arterial pressure of 65 mm Hg, and using lactate as a measure of tissue hypoperfusion and as a guide to resuscitation. The use of central venous pressure to guide fluid resuscitation has been largely discredited, owing to its inability to reliably predict fluid responsiveness. In its place, the guidelines recommend the use of so-called dynamic parameters, including passive leg raise, fluid challenges, pulse pressure variation in response to mechanical ventilation, and other techniques. Many devices and techniques exist to predict fluid responsiveness, but ultimately, clinical judgment must still be used to guide fluid resuscitation and vasopressor therapy in septic shock states.
Currently, when crystalloid fluid resuscitation is not effective at raising blood pressure, the vasopressor of choice is norepinephrine, which has been shown to be superior to the other vasopressors. Epinephrine can be substituted if norepinephrine is inadequate; however, it may lead to falsely elevated lactate concentrations and difficulty in using this as an endpoint for resuscitation. Phenylephrine has been associated with higher in-hospital mortality in septic shock.
When norepinephrine requirements increase significantly (past a value of approximately 5 mcg/min), low-dose vasopressin should be added. This is based upon research suggesting a relative vasopressin deficiency in inflammatory vasodilatory shock. The dosing is 0.04 units/min, a relatively low dose thought to correct this relative deficiency, and may decrease norepinephrine requirements to maintain the blood pressure. High-dose vasopressin is not recommended. When high doses of norepinephrine and added vasopressin are insufficient, epinephrine may then be added, with no data suggesting a clear benefit. Dobutamine may also be used to augment tissue perfusion if the CO is low.
The use of corticosteroids in septic shock has been controversial for over two decades. The current recommendation by the Surviving Sepsis Campaign is for low-dose steroids in vasopressor-dependent, volume-replete septic shock. Additional therapies being investigated include a bundle of vitamin C, thiamine, and hydrocortisone, the “Marik” protocol, which showed dramatic results in a single-center before and after trial. Similarly, foundational concepts such as initial fluid resuscitation versus vasopressor use and even the meaning of lactate as a resuscitation endpoint remain in dispute and are being investigated. The critical care physician is advised to remain abreast of the literature in this rapidly changing arena.
Other etiologies of distributive shock include anaphylactic and endocrine shock. The former occurs in response to an allergic stimulus, with the first-line treatment being epinephrine. Optionally, adjuncts such as Benadryl and steroids may be used. While the former does not manage airway symptoms and may cause hypotension, the latter takes several hours to be effective and has never proven to be of benefit.
Endocrine shock results from severe myxedema coma from thyroid deficiency and Addisonian crisis from acquired or iatrogenic hypothalamic-pituitary-adrenal axis suppression. The diagnosis depends on an accurate history and physical examination in both cases. Often, until laboratory tests rule out coexisting Addison disease, the treatment for myxedema coma includes empiric steroids alongside levothyroxine and liothyronine.
Hypovolemic shock is characterized by increased SVR and decreased CO, with the latter being secondary to decreased preload. This is a so-called “cold shock,” meaning the skin is cold and clammy from the vasoconstriction. For hypovolemic shock from dehydration or fluid losses, such as from prolonged physical activity in warm temperatures or excessive gastrointestinal (GI) losses and lack of oral intake, the treatment is relatively straightforward to include fluid resuscitation with crystalloid.
In hemorrhagic shock, blood products are the resuscitative fluid of choice. In these situations, crystalloid administration leads to increased coagulopathy and can increase the blood pressure, resulting in more bleeding, which is exacerbated by the aforementioned coagulopathy. Two resuscitation strategies are possible: empiric resuscitation with packed red blood cells (PRBCs), fresh frozen plasma (FFP), and platelets in a ratio designed to roughly approximate whole blood; or resuscitation based upon analysis of clotting via thromboelastography (TEG) or rotational thromboelastometry, which are tests that purport to accurately measure derangements in coagulation and guide resuscitation strategy. The first strategy, administration of blood products in an empiric ratio, is commonly employed in many centers as part of a massive transfusion protocol to be administered until control of bleeding can be obtained and/or resuscitation endpoints have been met. Alongside this are adjuncts such as the administration of tranexamic acid, which too is highly controversial. It should be noted that a strict 1:1:1 ratio was shown to be no better than a 1:1:2 ratio of FFP to platelets to PRBCs in patients with severe trauma.
Cardiogenic shock is a “cold” shock in that it is characterized by decreased CO due to intrinsic failure of the heart as a pump, with compensatory vasoconstriction. Numerous causes for heart failure resulting in cardiogenic shock exist, but the most commonly seen is the acute or long-term sequelae of coronary artery disease. This shock state has multiple variants to include diastolic versus systolic failure and left heart versus right heart failure. Valvular obstruction may also be a cause of dysfunction. Heart failure progresses through three stages. Initially, the filling pressure of the ventricle increases, but contractility is preserved at the expense of increased pressure and congestion in the lungs. Next, the stroke volume begins to fall, but an increase in heart rate preserves CO. Finally, CO begins to fall.
The mainstay of right heart failure therapy is straightforward. This includes fluid boluses until the central venous pressure (or wedge pressure if available) is above 15 mm Hg, followed by inodilator therapy with dobutamine or milrinone. This being said, fluid therapy must be used judiciously, as dila-tation of the right ventricle may cause the septum to bow out into the left ventricle, resulting in decreased left ventricular function, a phenomenon called interventricular interdependence. Inodilators both dilate the vasculature, reducing blood pressure, and promote CO by increasing contractility. They are ideal choices when the CO is low and the SVR high, but often are not options if systemic blood pressure is low. For this reason, inodilators in conjunction with vasopressors such as norepinephrine are a commonly pursued strategy, with the inodilator titrated to the CO and the vasopressor titrated to an appropriate systemic blood pressure.
The two questions in left heart failure are: what is the patient’s blood pressure and is the patient fluid overloaded? While diuretics historically have been given to almost all patients in left heart failure on the theory that such patients are past the inflection point on the Starling curve, in reality, patients who are in cardiogenic shock may be fluid overloaded, underloaded, or euvolemic. Alternative measures should be used to determine volume status, including weight, pedal edema, inferior vena cava ultrasound, and other newer noninvasive tools.
The other decision point is blood pressure. If the patient is hypertensive, then nitroglycerin, nitroprusside, or nicardipine may be used. All three will decrease the afterload, allow for forward flow of blood to peripheral tissues, reduce the myocardial oxygen demand, and protect the heart from ischemic damage. Nitroprusside has the risks of worsening coronary ischemia and of causing cyanide toxicity and so is less preferred compared to the other agents. If the blood pressure is normal in a state of cardiogenic shock, inodilators may be used. Vasodilators can also be used with caution as long as the blood pressure is maintained. Finally, if both the blood pressure and CO are low, epinephrine or dopamine infusions may be tried; however, due to their peripheral vasoconstriction, they can further increase the afterload and worsen the patient’s condition. This state has an extremely high mortality, and, often times, mechanical circulatory support is one of the few options left. These options include intra aortic balloon pumps, left ventricular assist devices, and extracorporeal membrane oxygenation (ECMO). If a facility does not have these resources in-house, the patient should be considered for transfer to a center with these capabilities, possibly by having a mobile unit from the accepting facility arrive at bedside and placing the patient on mechanical circulatory support prior to transfer.
The etiologies of obstructive shock include tension pneumothorax, cardiac tamponade, constrictive pericarditis, and massive PE. In all cases, the treatment is interventional in that they require removal of the cause of the obstructive shock. Options include a needle decompression or thoracostomy tube in the case of pneumothorax; pericardiocentesis or thoracotomy in the case tamponade; and heparinization and systemic or catheter-directed thrombolysis in the case of PE. A high index of suspicion is required in making these diagnoses.
Both myocardial ischemia and MI are feared entities in the perioperative period, both of them portending significant morbidity and mortality. The lack of ischemic symptoms or electrocardiogram (EKG) changes associated with a rise in troponins is not a sign of safety, as mortality remains high. Two fundamental types of MI exist. Type I MI is based upon atherosclerotic plaque rupture, and consequent ischemia and infarction of muscle that was being supplied by that blood vessel. In contrast, type 2 MI is based upon a mismatch of the supply of blood and the heart’s demand for it and is also commonly referred to as demand ischemia. The treatments for each type of MI flow naturally from their causes: revascularization for type I MI and reduction of cardiac oxygen demand for type 2 MI.
Type 1 MIs can be broken down into ST elevation MI (STEMI) and non-STEMI (NSTEMI). As its name indicates, a STEMI classically involves symptoms and elevations in cardiac enzymes as well as evidence of ST elevations on EKG, while an NSTEMI is similar but without EKG evidence of infarction. Unstable angina is a symptom of ischemia, but without cardiac biomarker elevation indicating injury. These critical distinctions drive differences in the immediate management, and it should be noted that unstable angina and NSTEMI are indistinguishable in the first 6 hours as that is how long it takes troponins to become positive after cardiac injury. All three together form part of the spectrum of acute coronary syndrome.
The treatment of a type I MI in the postoperative period focuses on a percutaneous coronary intervention with thrombolysis as an option if timely percutaneous coronary intervention is unavailable. In NSTEMI, intervention can be delayed in some cases up to 72 hours. However, the overall intervention decision is complicated by the added burden of deciding on the risks and benefits of full-dose systemic anticoagulation in postoperative patients. In certain populations, such as those who recently underwent neurosurgical procedures, the risks of stroke or death from hemorrhage make percutaneous coronary intervention with its attendant anticoagulation unacceptable, while in other patients, the risks can be accepted.
In addition to the primary intervention, the complications of an acute MI must be managed. Cardiogenic shock should be managed as previously discussed. Nitroglycerin, either sublingual or IV, can be given for hypertension and for chest pain. The distinction between a right- and left-sided MI is critical, as giving nitroglycerin to reduce the afterload and myocardial oxygen demand in the setting of presumed left-sided MI may reduce the preload, resulting in right-sided heart failure if in fact the patient is suffering a right-sided MI. This could have catastrophic consequences. Pain control with any IV opiate such as morphine can be used to control chest pain symptoms if not relieved by nitroglycerin. The patient should be given aspirin and additional anticoagulant medications as specified by local protocol and depending upon the course of therapy chosen. Beta blockers should be initiated if there are no signs of cardiogenic shock, as these are cardioprotective. However, if the patient is hypotensive or has a decreased ejection fraction or bradycardia, avoid these. Above all else, have the defibrillator close by, pads preferably on the patient, ready to shock any life-threatening arrhythmias or pace the patient if they go into bradycardia secondary to heart block.
ICUs in many ways were defined by the mechanical ventilator. The first ICU was arguably established by Dr. Bjørn Aage Ibsen in 1953 in response to a polio outbreak in Denmark. The use of positive pressure ventilation, initially supplied by medical students working in shifts, in combination with intubation, prevented secretions from causing aspiration pneumonitis and pneumonia, saving hundreds of lives. This basic combination of intubation and mechanical ventilation continues to represent a common function of critical care units of all types, and familiarization with both airway management and mechanical ventilation is fundamental to the practice of critical care medicine.
Respiratory physiology is characterized by two linked processes, oxygenation and ventilation. Oxygenation refers to the addition of oxygen (O 2 ) to the blood stream from the air, which is typically at a concentration of 21%, also known as the fraction of inspired oxygen (FiO 2 ). Ventilation is the clearance of carbon dioxide (CO 2 ) from the blood stream, after the latter has been generated by cellular respiration. It can help to think of these two processes as entirely separate, although in reality, this abstraction breaks down at extremely low minute ventilations.
Both processes rely upon air coming down the oral cavity, into the trachea, through the bronchi and into the lung parenchyma, where blood brought from the pulmonary artery goes through the pulmonary capillaries. This alveolar-capillary interface is where gas exchange occurs. If too much blood flow relative to oxygenation capacity is present, it is referred to as shunt physiology. Air exchange in areas that do not have sufficient blood supply is known as dead space physiology. Some amount of both of these is normal: less than 10% of total CO does not participate in gas exchange and 20% to 30% of total ventilation does not equilibrate with blood. Increases in shunt fraction occur secondary to asthma, to distention of alveoli from pulmonary edema or pneumonia, to atelectasis, or to PE, where excessive CO flows through nonembolized regions. Dead space ventilation takes place when the alveolar-capillary interface is destroyed by emphysema, when the CO is low, or when air overdistends the alveoli during positive pressure ventilation. Oxygen delivery was previously discussed in the “Cardiovascular System” section and is not further discussed here.
In respiratory physiology, compliance is the increase in volume of a lung in response to a given pressure applied to it. Diseased, fibrotic lungs in interstitial lung disease or acute respiratory distress syndrome (ARDS) have low compliance. Critically, in the era of mechanical ventilation, low compliance can be a viscous cycle. An initial insult from pneumonia, pulmonary contusion, severe systemic disease like pancreatitis, or other cause can result in a decrease in compliance. Using mechanical ventilation, air is forced into the lungs to deliver a set amount of ventilation. This results in barotrauma, or trauma to the lungs from increased pressure. This barotrauma further decreases compliance. Thus, preventing barotrauma is a key goal of the ARDSNet protocol, which is discussed later.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here