Critical Care Medicine
- 1. Description of the problem
2. Emergency Management
- 3. Diagnosis
What's the evidence?
Also known as: Low blood pressure
Related conditions: Hypotension may be related to but is not synonymous with ‘shock.’
1. Description of the problem
What every clinician needs to know
Hypotension is blood pressure (BP) that is below that expected for a person in a particular demographic category. Neonates have systolic blood pressure (SBP) in the 50’s and that is entirely adequate for them, whereas a middle-aged person would be expected to have an SBP between 100 and 120 mmHg. Normal BP values are based on what has been measured in the general population and considered to be healthy values.
Certain parameters are used to alert the clinician to the possibility of an impending problem, and generally a drop in BP > 40 mmHg, an SBP < 90 mmHg, and mean arterial pressure (MAP) < 60 mmHg are considered hypotensive in an adult population. However, it is important to appreciate that low BP is not always a problem. The value of the BP must be interpreted in clinical context.
BP per se does not determine circulatory adequacy; organ function is dependent on the volume of blood passing through it per unit time: flow. However, BP is related to flow as described below in the section on pathophysiology. A person may have low BP with good flow, good organ perfusion, and no adverse symptoms; this would typically be a young, athletically conditioned person. Conversely, if a person has chronic hypertension and BP falls acutely to ‘normal’ values, organ hypoperfusion may occur.
In chronic hypertension autoregulatory mechanisms, particularly in the cerebral, coronary, and renal circulations, will be adapted to higher MAPs. Vessels may be atherosclerotic and therefore any decreases in pressure will invariably manifest as decreases in flow.
Hemodynamic shock is said to occur if there is evidence of organ hypoperfusion (see section on pathophysiology: oxygen delivery). There is a significant overlap between the causes of hypotension and shock. While it is fair to say that in a considerable proportion of shocked patients the BP will be low, a shocked patient may have BP that is in the population normal range or may have high BP.
The degree of organ dysfunction associated with low BP depends on the decrease in BP from the physiological norm to which the organs are adapted and the duration of the insult. Organ failure may occur hours or days in a previously asymptomatic patient; therefore, a proactive decision on what BP range will be tolerated should be made in anticipation of complications.
Is this person compromised as a result of this BP?
Hypotension in terms of the numerical definition may be asymptomatic but in a clinical context a patient is hypotensive if the low BP is associated with symptoms or clinical signs. Manifestations can be nonspecific, such as malaise; organ-specific (e.g. exacerbation of angina), or as the biochemical features of shock.
Examination points below are not exhaustive but highlight the most important aspects in the assessment of a hypotensive patient.
Start with a global assessment: does the patient look unwell? Is there any pallor or obvious signs of hemorrhage? Is the person thirsty?
Integument: Are the mucous membranes dry, is the skin wrinkled? Check the peripheries: are they cool or warm, what’s the capillary refill? Is there any mottling or diaphoresis?
CNS: Is the mental state altered? What’s the Glasgow Coma Scale score? Is there any confusion, disorientation or anxiety? This may be subtle and apparent only to relatives.
Respiratory: The respiratory rate is one of the most sensitive signs of inadequate organ blood flow. Count this yourself. Check for tracheal deviation and adequacy of breath sounds bilaterally. Auscultate for crackles.
CVS: Look at the jugular venous pressure. Check to see if heart sounds are audible and listen for added sounds. Check the ECG for rhythm, signs of LVH or ischemia.
Renal: A low urine output may imply hypovolemia but is often a late sign; therefore, normal urine output cannot exclude it.
Blood loss: The American College of Surgeons Advanced Trauma Life Support Classification of Hemorrhage Severity (
Sepsis: Typically the patient will be pyrexial and will have warm peripheries with bounding pulses. According to international consensus classification of sepsis, the presence of hypotension and/or organ dysfunction implies severe sepsis; hypotension resistant to treatment with fluids therefore requiring inotropes is septic shock.
Management should initially focus on resuscitation and stabilizing the patient if possible. A patient who is hypotensive due to a ruptured spleen will not be stabilized without surgical intervention; thus, establishing a cause is paramount in the initial assessment phase.
The most common cause of hypotension is hypovolemia; therefore, fluid resuscitation is indicated in almost all situations. The obvious exception is in the case of heart failure with pulmonary edema.
The aim in all situations should be to optimize preload before pursuing other treatment options such as vasopressors and inotropes.
2. Emergency Management
Resuscitation should always begin with assessment of the airway and breathing and administration of oxygen.
Peripheral intravenous access is necessary: a wide-bore, short catheter in a large vein provides acceptable conditions to rapidly administer large volumes of fluid if indicated. If hemorrhage is suspected then at least two such catheters will be required. In some situations where rapid large-volume hemorrhage is anticipated, such as aortic aneurysm repair, the clinician may choose to insert a wider catheter such as a pulmonary artery catheter introducer.
However, central venous access is usually more time-consuming and can be difficult in a hypovolemic person, and it is inappropriate in the acute setting unless peripheral access cannot be established or inotropes are required.
Intra-arterial BP monitoring is useful as it provides continuous monitoring and the systolic pressure variation of the arterial trace can be used as an indicator of volume depletion. Furthermore, repeated sampling for indices of hypoperfusion is facilitated. Insertion of an arterial cannula is rarely required emergently and it is more appropriate to focus attention on stabilizing the patient.
Preload should be optimized with repeated administration of "fluid challenges." These are aliquots of 200-250 ml of fluid given rapidly, often over 10 minutes, in order to assess response. The efficacy of colloids over crystalloids remains unproven; theoretically colloids will remain in the intravascular compartment for longer, but ultimately both will redistribute throughout the extracellular space. In most institutions crystalloids are now used in the emergency department because of concerns about the potential for allergic reactions associated with the use of colloids.
Fluid responsiveness: Heart rate and BP are the most readily available parameters to use initially to assess fluid responsiveness; however, the clinician should be aware of their limitations. The value of central venous pressure (CVP) monitoring is controversial but is generally considered to be helpful in this situation. In health, the CVP as the surrogate of right atrial pressure correlates well with pulmonary artery occlusion pressure, a surrogate of left atrial pressure; however, in critically ill patients, many factors, both pulmonary and cardiac, can interfere with this relationship.
A very low CVP suggests intravascular volume depletion and a high value may suggest adequate filling pressure; however, such static measurements do not provide adequate information. It is the change in CVP following fluid challenging that is most useful; a lack of increase in CVP or an increase and subsequent decrease suggests fluid responsiveness, whereas a rapid steep increase suggests that the heart has reached the plateau phase of its Frank-Starling curve.
Hypovolemia can be absolute as with any cause of fluid loss or relative as with systemic inflammation where vasodilatation results in greatly increased vessel capacitance. It should not be surprising if a severely ill patient requires large volumes of fluid resuscitation (> 5 liters). It is not the total volume of fluid administered but the care with which it is given that is most important. Even in the absence of hemorrhage, transfusion of packed red cells may be required if large-volume resuscitation has caused significant hemodilution and subsequent anemia. However, this should not be considered emergency treatment.
A multicenter randomized controlled trial of restrictive (Hb 7-9 g/dl) versus liberal (Hb 10-12 g/dl) transfusion policy showed that there was no difference in mortality, but that changes in multiple organ dysfunction were significantly less in the restrictive group, with the possible exception of patients with acute coronary syndromes. These parameters do not apply to patients who are actively bleeding.
Central venous saturations (ScvO2) can be used as a guide to the adequacy of resuscitation and oxygen delivery. In early goal-directed therapy of patients (EGDT) with septic shock, targeting ScvO2 >70% or mixed venous saturation >65% was shown to improve survival. Mixed venous saturations can be obtained only if a pulmonary artery catheter is in place and are generally 2 mmHg lower than ScvO2 because of increased extraction of oxygen from the brain.
In sepsis, this relationship can be reversed and the difference widened. High venous saturations can also occur as a result of impaired oxygen extraction and utilization that can occur with sepsis; therefore, these values must be interpreted in a clinical context.
Hemorrhage: It is no longer acceptable to administer large volumes of fluid to a bleeding patient with the aim of restoring normal BP. It is preferable to administer fluid in aliquots of 20 ml/kg, aiming for the lowest pressure that will sustain organ perfusion until the source of the bleeding is controlled. In a prospective trial of immediate versus delayed fluid resuscitation in field casualties with penetrating trauma, there was better survival in the delayed group (70% vs. 62%); however, there is no evidence for this in the hospital setting, where surgical resources are to hand.
It is difficult to judge volume resuscitation in an exsanguinating patient and in this context it is necessary to administer large volumes of fluid and blood products while definitive treatment is underway. Hemodilution of clotting products, coagulopathy, hypothermia, and acidosis, a dangerous combination, should be anticipated in any massive hemorrhage.
A fluid-warming rapid infuser should be used and along with transfusion of red cells, replacement of clotting factors and fibrinogen should be undertaken without waiting for laboratory confirmation. Due to unremediable delays in obtaining these lab results, they lag behind the real clinical course. Thromboelastography (TEG) is more useful as it can be obtained rapidly and provides an indication of whole blood coagulation; unfortunately the available of TEG is usually limited to cardiac surgery units.
Does the patient have low BP?
Measuring the BP
For noninvasive BP measurement: Ensure that you’re using the correct size cuff for the diameter of the arm. The width of the cuff should be equivalent to the diameter + 20%. A small cuff will overestimate the BP and a large cuff will give a lower reading, leading to misdiagnoses and inappropriate treatment.
For intra-arterial BP: Take into account which vessel the arterial cannula is in. Due to gravity effects femoral cannulae will give higher readings than those placed in the upper limbs. More peripheral cannulae give a narrower trace with a higher amplitude; thus, the systolic pressure in the dorsalis pedis is higher than that in the femoral, and radial artery SBP is higher than aortic SBP.
Ensure that the arterial pressure trace is adequate and that the tranducer is at the correct level (i.e., that of the right atrium). If the trace looks damped, check the fluid column for air bubbles or blood clots, flush the catheter and check that the fluid bag in the transducer equipment is sufficiently inflated. Damping will overestimate the diastolic BP (DBP) and underestimate the SBP, whereas over-resonance will underestimate the DBP and overestimate the SBP. Nevertheless, in both these situations the MAP remains largely unchanged.
Take at least two BP measurements interspaced by at least several minutes. Check BP in both arms. Check that all pulses are present. Is it the SBP, DBP, or both that is low? Is the pulse pressure high, low or normal? Check supine and standing or sitting BP if possible to look for a postural drop, which usually implies hypovolemia.
Is this level of BP normal for this patient?
Ask what the patient’s normal BP is. Look at previous hospital or GP records. Patients often do not know what their usual BP is but may know if they have low or high pressure. Look at medication: are there any antihypertensives there? Many drugs used to treat hypertension are also used to treat angina (e.g., beta-blockers), so ask if they are for high BP or ischemic heart disease. Establish risk factors for hypertension: ethnicity, age, smoking, obesity, diabetes mellitus, and associated disorders such as cerebrovascular, cardiovascular, or peripheral vascular disease.
Assessment of potential causes: This will be based largely on the history, which will determine the focus of the examination.
The most common cause of hypotension is inadequate preload due to hypovolemia.
Is there evidence of volume depletion?
Reduction in cardiac output can result from inadequate SV or HR
Preload is the end diastolic ventricular volume, which determines the pre-systolic length of the cardiac muscle fibers. The Frank-Starling principle demonstrates that the tension developed on contraction relies on and is proportional to the initial length of the fiber up to an optimal length. Increasing volume beyond this point will not increase sarcomere length but results in reduction of ventricular compliance.
In clinical terms it isn’t practical to measure end diastolic volumes; however, the filling pressures of the ventricles are used to represent these volumes. Central venous pressure acts as a guide to right atrial and therefore right ventricular (RV) filling, whereas pulmonary capillary wedge pressure corresponds to left atrial and thus left ventricular (LV) filling (in the absence of pathology such as mitral stenosis).
Practically, the use of pulmonary artery catheters is rare; thus, right heart filling pressures are usually used as a guide to left ventricular pressures. RV pressure will not accurately correlate to LV pressures in the presence of elevated pulmonary vascular resistance (i.e., pulmonary hypertension, ARDS).
Intravascular volume depletion
Decreased fluid intake: mental or physical disability; elderly, organic disease (e.g., dysphagia); nausea associated with acute illness or chronic disease and chemotherapy
Increased losses: excessive sweating, diarrhea, vomiting
Third-space fluid sequestration: bowel obstruction, pancreatitis, sepsis, burns
Polyuria: osmotic diuresis with untreated diabetes mellitus, diabetic ketoacidosis (DKA) or hyperglycemic hyperosmolar syndrome (HHS), where fluid deficit can be enormous
Hemorrhage: In a postoperative or trauma patient, hypotension is due to blood loss unless proved otherwise.
Other causes of inadequate preload
Increased intrathoracic pressure causes reduction in venous return to the RA: mechanical ventilation particularly with high positive end-expiratory pressure (PEEP), tension pneumothorax
Obstructive: pulmonary embolus, mitral stenosis, cardiac tamponade; suspect this particularly after cardiac surgery.
Right ventricular failure
Disorders of heart rate and rhythm
Contractility is the intrinsic ability of the intact ventricle to do work or the pump action of the heart. Factors that can impair contractility:
Acidemia: As pH is a logarithmic scale, the degree of myocardial impairment increases exponentially with decreases in pH.
Myocardial ischemia and infarction
Cardiomyopathy: dilation, hypertrophic, viral myocarditis, sepsis, myocardial stunning post cardiac surgery or cardiac arrest
Drugs: negatively inotropic drugs (e.g., beta-blockers, calcium channel antagonists)
The physiological definition of afterload is the ventricular wall stress developed during contraction. Arterial pressure during systole is an indirect index of wall tension; however, in the presence of aortic stenosis or hypertrophic obstructive cardiomyopathy, there will be a gradient between the ventricular pressure and that in the aorta. In this context factors that increase afterload may cause hypotension; however, increases in systemic vascular resistance (SVR) manifest as augmented BP.
Factors that reduce SVR resistance:
Systemic inflammatory response: sepsis, burns, major trauma, panceatitis
Rewarming of a ‘cold’ patient (e.g., post cardiac surgery, therapeutic cooling after cardiac arrest)
Adrenal insufficiency: relative insufficiency (e.g. critical illness), absolute insufficiency (e.g. Addison’s disease or bilateral adrenalectomy)
Regional anesthesia: spinal or epidural anesthesia
General anesthesia: postoperative drug redistribution from the third to first compartment
Immune-mediated: anaphylaxis and anaphylactoid reactions
Drugs: vasodilators, ACE inhibitors, opioids
Disorders of heart rate and rhythm
Bradycardia: if stroke volume (SV) remains constant, reduction in heart rate (HR) will lead to a reduction in cardiac output.
Atrio-ventricular dysynchrony (e.g., complete heart block, atrial fibrillation). Atrial contraction contributes up to 25% to ventricular filling. Loss of this atrial ‘kick’ effectively results in reduction in preload.
Tachycardia: will also result in reduced preload due to reduced diastolic filling time. The majority of ventricular filling occurs in early diastole and only a small proportion of filling occurs in late diastole (diastasis). As HR increases, initially diastasis is reduced, so HRs of up to 150 beats per minute are usually well tolerated; thereafter, the initial filling is compromised. Coronary artery perfusion occurs during diastole; therefore, ischemia can ensue if diastolic time is significantly shortened.
Drugs: negative chronotropes (e.g., beta-blockers)
What diagnostic tests should be performed?
Routine investigations should include hemoglobin for evidence of blood loss, coagulation profile for reversible causes of hemorrhage, and a group and save sample in case blood transfusion is required. Severe hypotension can cause hypoxic-ischemic injury to vital organs such as the brain, the liver, and kidneys; therefore, a full biochemistry profile is required. An elevated urea may suggest hypovolemia or upper gastrointestinal hemorrhage.
Arterial blood gas analysis: Lactate and base deficit are the most useful indicators of hypoperfusion. If an arterial sample is not obtainable then a venous sample will provide all the required information in terms of organ perfusion.
ECG: is essential and will determine the nature of any arrhythmias, and may show evidence of myocardial ischemia or infarction.
CXR: may show evidence of heart failure, or pneumothorax
Echocardiography: this can be performed rapidly at the bedside for initial assessment as a detailed echo exam requires expertise and can be time-consuming. A focused echo can exclude cardiac tamponade and can give an indication of volemic status, right and left ventricular contractility and any regional wall motion abnormalities. A formal departmental echo can be obtained once the patient is stabilized.
Focused sssessment with donography for trauma (FAST): this is a basic ultrasound examination of the abdomen (and pericardium) with the aim of identifying the presence of free fluid. In the context of trauma if the fluid is due to hemorrhage this may guide the decision for further imaging or immediate surgical intervention.
Is there an indication for cardiac output (CO) monitoring? This should be considered in any patient who has not responded to initial therapy and where problems with cardiac contractility are suspected to guide inotropic support.
The choice of CO monitor will depend on availability within an institution, the question that is proposed and the clinician’s confidence with its use and interpretation of data. Most modern CO monitors provide the following parameters -- arterial pressure, HR, SV, CO, SVR -- and all can provide continuous monitoring and have been validated against the gold standard pulmonary artery catheter. Below is a summary of the pros and cons of various devices.
Provides flow time (FTc = corrected to HR) to reflect afterload and mean acceleration (MA), which reflect contractility. Can be difficult to keep the Doppler probe focused. Can’t be used in non-sedated/ventilated patients.
LiDCO: lithium dilution cardiac output
This is a non-invasive CO monitor that can be calibrated rapidly by connecting to any arterial catheter. It provides information on the dynamic parameters of stroke volume variation (SVV) and pulse pressure variation (PVV), which are better indicators of volemic status than static parameters.
PiCCO: Transpulmonary indicator dilution technique. This is a relatively non-invasive device but requires a special arterial catheter that must be inserted into the brachial or femoral artery. Also provides not only SVV but also intrathoracic blood volume (ITBV), extravascular lung water (EVLW), and global end-diastolic volume (GEDV).
Pulmonary artery catheter
Relatively invasive, but useful if a discrepancy between right and left ventricular filling pressures is suspected as it provides pulmonary artery occlusion pressures (PAOP). It is often the preferred CO monitor in cardiac ICUs. After obtaining wedge pressures, the tip needs to be in the main PA and the position checked on CXR. Its use is associated with significant infection risk and morbidity; therefore, it shouldn’t remain in situ for more than 3 days.
The relation between pressure, flow and resistance in the cardiovascular system can be determined by applying Ohm’s law, which states that current (i.e. flow) through a conductor is proportion to the voltage (i.e. pressure) and inversely proportional to resistance (I = V/R or V = IR).
Mean Arterial Pressure = Cardiac Output x Systemic Vascular Resistance
Cardiac output = Stroke Volume x Heart Rate
A change in any of the physiological parameters above may result in a decrease in blood pressure. SV is determined by preload, contractility and afterload. Decreases in preload or contractility and increases in afterload will reduce SV; however, cardiac output can be maintained if there is a compensatory tachycardia. HR and contractility are influenced by sympathetic and parasympathetic activity and catecholamines.
With each contraction of the LV, the ejected fraction distends the arterial wall, generating the systolic pressure, which is inversely proportional to the capacitance of the artery. BP then gradually falls as the blood redistributes to the peripheral arterioles. The lowest level reached is the diastolic component, which is directly related to both SVR and capacitance.
Flow through a single vessel that is >0.5 mm in diameter will be laminar and therefore can be construed by applying the Hagen-Poiseuille principle. This states that flow is proportional to the pressure drop and the radius but inversely proportional to the length of the vessel and viscosity of the fluid that flows through it.
This illustrates why a central venous catheter is not a good device to use in situations where rapid volume resuscitation is required; it is long and has a small radius and therefore high resistance. The maximum flow rate through it is much less than that through a wider, shorter peripheral cannula.
The vascular resistance is the sum of all forces that oppose flow through a vascular bed, of which there are two main constituents. The frictional component: viscosity dictates friction between the blood and vessel wall; and drag between fluid layers. The reactive component: the energy required to convert kinetic energy of the pump action to potential energy in the elastic vessel wall.
Clinically it is much harder to measure flow than it is to measure pressure, and that’s why pressure is used as a surrogate marker of perfusion.
Ultimately the function of the circulation is to deliver oxygen to the tissues, which depends upon the following factors:
Oxygen Delivery (DO2)= Cardiac Output (CO) x Arterial Oxygen Content (CaO2)
CaO2 = [O2 Saturation(%) xHb (g/dl) x 1.34(mls/g)] +[PaO2 (kPa) x 0.023(ml/kPa/dl)]
1.34 represents the affinity of hemoglobin for oxygen
0.023 represents the solubility of oxygen in plasma
Shock can occur both as a result of impaired oxygen delivery due to hemodynamic inadequacy or carbon monoxide poisoning or impaired utilization, such as that which occurs in mitochondrial toxicity from sepsis or cyanide poisoning.
What's the evidence?
Fink, MP, Abraham, E, Vincent, JL, Kochanek, PM. Textbook of Critical Care. Elsevier Saunders. 2005. pp. 27-29.
Power, I, Kam, P. Principles of Physiology for the Anaesthetist. Arnold Publishers. 2001. pp. 99-158.
Spahn. "ACS/ATLS, American College of Surgeons/Advanced Trauma Life Support". Critical Care. vol. 11. 2007. pp. R17.
Levy, MM, Fink, M, Marshall, JC. "SCCM/ESICM/ACCP/ATS/SIS International Sepsis definitions Conference". Critical Care Medicine. vol. 34. 2003. pp. 1250-6.
Rivers, E, Nguyen, B, Havstad, S. "Early Goal Directed Therapy in the treatment of severe sepsis and septic shock". New England Journal of Medicine. vol. 345. 2001. pp. 1368-77.
Kwan, I, Bunn, F, Roberts, I. "Timing and volume of fluid administration for patients with bleeding". Cochrane Database of Systematic Reviews. 2003.
Herbert, PC. "Transfusion Requirements in Critical Care TRICC investigators for the Canadian Critical Care Trials Group. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care". New England Journal of Medicine. vol. 340. 1999. pp. 409-417.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.