Novel tools for hemodynamic monitoring in critically ill patients with shock

INTRODUCTION — Central venous and pulmonary artery catheters (PAC) are invasive tools that have traditionally been used for hemodynamic monitoring in patients who present with shock. However, these tools have drawbacks and inaccuracies. Thus, several, less invasive, novel technologies are available or being investigated for use to assess parameters such as cardiac output, intravascular volume status, responsiveness to intravenous fluid administration, and tissue perfusion. They can potentially be used in the emergency department, intensive care unit, and operating room when caring for patients with shock or hypovolemia.

This topic will discuss novel techniques for hemodynamic monitoring. The evaluation and treatment of shock, central venous pressure and PAC monitoring are discussed separately. (See "Evaluation and management of suspected sepsis and septic shock in adults" and "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock" and "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)

GENERAL PRINCIPLES

Deficiencies of standard techniques — For many years, the gold standard for hemodynamic monitoring was the pulmonary artery catheter (PAC). However, several studies have demonstrated that the PAC fails to improve outcome in critically ill patients and may be associated with harm. In the late 1990s, central venous pressure (CVP) monitoring via central venous catheterization (CVC) emerged as a less invasive alternative that was incorporated into guideline management of sepsis; however, this practice has also been questioned [1-3]. (See "Pulmonary artery catheterization: Indications, contraindications, and complications in adults" and "Evaluation and management of suspected sepsis and septic shock in adults".)

CVP and PAC monitoring suffer from the following inadequacies:

●Inconsistent prediction of fluid responsiveness – Both CVP and pulmonary alveolar occlusion pressure have been shown to have poor predictive value for predicting fluid responsiveness (arbitrarily defined as an increase of at least 15 percent in cardiac output [CO] in response to a 500 mL bolus fluid challenge, as measured by PAC) [3-6]. Furthermore, CVP is affected by a number of other physiologic derangements, including valvular regurgitation, right ventricular dysfunction, pulmonary hypertension, and variation in intrathoracic pressure with respiration.

●Complications associated with invasiveness – CVCs and PACs require central venous access and have been associated with a number of complications, including arrhythmias, injury to vascular or cardiac structures, catheter-associated bloodstream infection, pneumothorax, and venous thromboembolism. (See "Pulmonary artery catheterization: Indications, contraindications, and complications in adults", section on 'Complications' and "Central venous catheters: Overview of complications and prevention in adults".)

●Data interpretation is difficult – Data from CVCs and PACs may be challenging to interpret both due to the lack of standardization of technique and the hemodynamic complexity of patients receiving them. Several studies have documented poor interobserver reliability and challenges interpreting intravascular pressures from PACs, even among trained intensivists [7,8]. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)

Indications — A plethora of techniques aimed at overcoming the deficiencies associated with standard hemodynamic monitoring tools have been developed, many of which use complex imaging technology and computer algorithms to estimate the following:

●Fluid responsiveness and volume status (see 'Volume tolerance and fluid responsiveness' below)

●Cardiac output (see 'Cardiac output' below)

●Tissue perfusion (see 'Measurement of tissue oxygen saturation' below and 'Measurement of microcirculatory blood flow' below and 'Tissue perfusion' below)

While no large randomized trials of resuscitation guided by noninvasive hemodynamic monitors have been conducted, a systematic review of 13 trials enrolling over 1600 subjects found that such practice was associated with reduced mortality, ICU length of stay, and duration of mechanical ventilation [9]. A subanalysis of the ANDROMEDA-SHOCK trial suggests that assessment of fluid responsiveness was possible in 80 percent of patients. Non-fluid responsive patients received lower fluid volumes, exhibited less positive fluid balances, and received more vasopressors, but had no difference in mortality or organ failure [10]. These findings will likely only increase the interest in use of these tools in critically ill patients.

Limitations — The devices discussed in this topic have several limitations, some of which explain why their use is not widespread. First, many of these tools use proprietary algorithms or imaging technology, making it difficult to confirm their validity. Second, some devices produce data that still require complex interpretation which can be challenging even for expert clinicians. Third, no large randomized trials of resuscitation guided by noninvasive hemodynamic monitors have been conducted.

Physiologic principles

Upstream versus downstream monitors — A greater understanding of tissue and cellular hypoxia as a cardinal feature of shock has led to the concept of "upstream" and "downstream" indicators of organ perfusion [11].

●Upstream – "Upstream" ("macro") markers assess flow and pressure in the heart, vena cava, pulmonary artery, and aorta and are the traditional variables that have been used to assess the hemodynamic status of critically ill patients. The majority of existing hemodynamic monitors are upstream monitors. (See 'Volume tolerance and fluid responsiveness' below.)

●Downstream – Shock with end-organ dysfunction occurs at the capillary and tissue levels [12]. Tools have been developed that follow alterations in tissue oxygenation and microvascular blood flow. These techniques are known as the "downstream" (or "micro") markers of resuscitation. (See 'Measurement of tissue oxygen saturation' below and 'Measurement of microcirculatory blood flow' below.)

Heart-lung interaction during mechanical ventilation — The underlying physiologic principle common to a number of the monitoring tools discussed in this topic is the heart-lung interaction [13]. During the inspiratory phase of positive pressure ventilation, intrathoracic pressure increases, passively increasing right atrial pressure, causing venous return to decrease and the vena cava to distend. If both the right ventricle (RV) and left ventricle (LV) are fluid responsive, this leads to decreased RV output and, after two or three heartbeats, decreased LV output [13,14]. In preload-dependent patients, cyclic changes in LV stroke volume (SV) and its coupled arterial pulse pressure are seen, and the magnitude of the changes is proportional to volume responsiveness. The inverse is true during spontaneous negative pressure breathing, though this has not been well-studied. (See 'Pulse contour analysis (fluid responsiveness)' below.)

VOLUME TOLERANCE AND FLUID RESPONSIVENESS — The devices listed in this section rely on the principles that underlie the heart-lung interaction during mechanical ventilation and/or visually assess flow in the heart and major vessels (ie, "upstream" monitors). Volume tolerance generally refers to the notion that the administration of intravenous fluid resuscitation will cause no harm to the patient. Volume responsiveness implies an improvement in cardiac output (often by 10 to 20 percent) and, ideally, tissue perfusion, after the administration of intravenous fluid. (See 'Physiologic principles' above.)

Pulse contour analysis (fluid responsiveness)

Pulse pressure variation (PPV) — Pulse pressure (ie, the difference between systolic and diastolic arterial blood pressure) varies with respiration induced by positive pressure ventilation. Variation in pulse pressure is thought to be an indicator of a patient's position on the Frank-Starling Curve, a curve that denotes a patient's response to pre-load (ie, fluid responsiveness) (figure 1) [15]. Patients operating on the flat part of the curve are insensitive to changes in preload induced by mechanical ventilation and thus have a low variation in the pulse pressure, indicating a lack of fluid responsiveness. In contrast, patients operating on the steep portion of the curve, are sensitive to cyclic changes in preload induced by mechanical ventilation and hence, exhibit greater variation in the pulse pressure (ie, fluid responsive).

Numerous studies have demonstrated that a PPV of at least 13 to 15 percent is strongly associated with volume responsiveness [4,14,16]. As an example, one systematic review of 29 studies reported a higher area under the receiver operating characteristic curve (AUROC) for PPV compared with CVP (0.94 versus 0.55) as an indicator of fluid responsiveness (sensitivity and specificity were 0.88 each) [16].

PPV is typically calculated as the ratio of the maximum pulse pressure (systolic blood pressure minus diastolic blood pressure; PPmax) minus the minimum pulse pressure (PPmin) to the mean pulse pressure (PPmean), usually averaged over three or more breaths. Although it can be measured from pressures derived from manual cuff-inflation, measurements are generally more accurate when an arterial catheter is used such that the latter is preferred (figure 2):

A number of commercially available devices and monitors capable of measuring PPV exist, and several use complex proprietary algorithms to additionally calculate stroke volume (SV) and cardiac output (CO).

While encouraging, this technique is limited to patients who are mechanically ventilated, receiving ≥8 mL/kg of tidal volume, in sinus rhythm, and not spontaneously triggering the ventilator, factors that limit its general applicability in the intensive care unit [13]. In addition, sensitivity for volume responsiveness is decreased in patients ventilated with tidal volumes of ≤6 mL/kg, as the cyclic changes induced by mechanical ventilation are less pronounced [17]. Although patients with intra-abdominal hypertension often have markedly abnormal respiratory system compliance, PPV may be accurate in this setting and data in humans are limited [18,19].

Stroke volume variation (SVV) — SV is linearly related to pulse pressure. SVV functions on the same physiologic principle as PPV (see 'Pulse pressure variation (PPV)' above). Studies have consistently found that SVV >10 percent is associated with fluid responsiveness [13,19-22]. As an example, in a study of 40 mechanically ventilated liver transplant patients, an SVV threshold of >10 percent discriminated fluid responsive patients with a sensitivity and specificity of 94 percent each [20].

Analogous to PPV, SVV is typically defined as the ratio of the maximum (SVmax) SV minus the minimum SV (SVmin) to the mean SV (SVmean), averaged over several respiratory cycles.

The SV can be calculated from the arterial pressure waveform if the arterial compliance and systemic vascular resistance are known, values typically derived from an arterial catheter. SVV is typically measured by several commercially available devices. SVV can also be determined by measuring aortic blood flow velocity using esophageal Doppler, bioimpedance, and bioreactance technology, which are discussed below. (See 'Thoracic electrical bioimpedance or bioreactance' below.)

SVV has the same limitations as PPV (see 'Pulse pressure variation (PPV)' above). Although there is evidence that suggests that SVV may also be applied to spontaneously breathing patients, this has not been validated [23]. Patient position may also affect SVV accuracy. In one study, the 30 degree head-up and prone positions were associated with increased SVV because of the associated decreased SV associated with these positions [24]. Another study reported poor correlation with pulmonary artery catheter assessment of volume status [25].

A meta-analysis of 11 randomized trials (total 1015 patients) reported that measuring fluid responsiveness using the SVV resulted in a shorter length of hospital stay and (weighted mean difference -1.96 days, 95% CI -2.34 to -1.59) [26]. In addition, there was a reduction in mortality that was not significant (odds ratio 0.55, 95% CI 0.3-1.03).

Oximetric waveform variation — Using the same principles as PPV and SVV, variation in the plethysmographic waveform of the pulse oximeter has been proposed as a predictor of fluid responsiveness. The pleth variability index (PVI) is an automated algorithm that has been shown to modestly predict fluid responsiveness in the operating room [27-29]. However, PVI was not associated with fluid responsiveness in two intensive care unit-based studies and has not been systematically studied in the emergency department setting [29,30].

Passive leg raising or fluid bolus challenge — Many devices used for measuring CO, PPV, or SVV may be combined with a provocative maneuver to assess whether or not a patient is fluid responsive. (See 'Cardiac output' below and 'Pulse pressure variation (PPV)' above and 'Stroke volume variation (SVV)' above.)

Provocative maneuvers include:

●Intravenous fluid bolus – These parameters may be measured before and after a small "test" bolus of intravenous fluid (250 to 500 mL administered over 5 to 10 minutes) to assess whether a patient is fluid responsive.

●Passive leg raising (PLR) – PLR is thought to provide a bolus of the patient's own intravascular blood from the capacitance veins of the lower extremities into the thorax (figure 3) [31]. PLR is accomplished by the following steps:

•Position the patient in the semi-recumbent position with the head and torso elevated at 45 degrees.

•Obtain a baseline measurement (eg, baseline of CO).

•Lower the patient's upper body and head to the horizontal position and raise and hold the legs at 45 degrees for one minute.

•Obtain subsequent measurement.

•Tidal volume challenge – PPV and SVV do not reliably predict volume responsiveness during low tidal volume ventilation. A small study demonstrated that transiently raising tidal volume from 6 to 8 mL/kg (using predicted body weight) is safe and the absolute change in both PPV and SVV predicts fluid responsiveness [32].

Although poorly defined, a 10 percent increase in CO has been shown in several studies to predict fluid responsiveness [31,33-37] whereas a reduction in SVV and PPV is expected in those who are fluid responsive when provocative maneuvers are used. One meta-analysis of 23 studies reported a sensitivity of 86 percent and a specificity of 92 percent for PLR for predicting fluid responsiveness; the predictive value of PLR was best when a flow variable such as CO was used in conjunction with PLR [38]. Another review of 50 studies also reported that augmentation of CO (or other related parameters) on PLR had a likelihood ratio of 11 and specificity of 92 percent, better than other measurements of fluid responsiveness including physical examination, central venous pressure, and respiratory variation in vena cava diameter [39]. (See 'Vena cava assessment' below.)

Point-of-care ultrasonography — Although point-of-care bedside ultrasonography (POCUS) is not traditionally considered a monitoring device, evaluation of the lung and heart are important components of the evaluation of hemodynamically compromised patients [40-45]. Although the data on its use are limited, a randomized trial demonstrated that use of POCUS for patients with undifferentiated hypotension in the emergency department did not reduce mortality [46]. Data that supports the use of POCUS in critically ill patients who present with shock or trauma are discussed separately. (See "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock" and "Emergency ultrasound in adults with abdominal and thoracic trauma" and "Indications for bedside ultrasonography in the critically ill adult patient".)

POCUS techniques that assess volume status are discussed here while point-of-care echocardiography to assess CO is discussed below. (See 'Point-of-care echocardiography' below.)

Vena cava assessment — Vena cava diameter and dynamic measures of vena cava collapse have been proposed as tools for estimating intravascular volume status.

Because there is no valve between the vena cava and right atrium, fullness of the vena cava is thought to correlate with increased right atrial pressure [47,48]. During spontaneous breathing, a decrease in intrathoracic pressure with inspiration draws blood from the vena cava into the heart, leading to collapse of the vessel. Conversely, during positive pressure ventilation, increased intrathoracic pressure pushes blood from the heart into the vena cava, leading to distention of the vessel. The magnitude of these changes has been proposed to correlate with intravascular volume status and fluid responsiveness.

Typically, the inferior vena cava (IVC) is identified in its longitudinal axis in the subcostal view as it enters the right atrium. Diameter should be measured approximately 2 centimeters from the junction of the IVC and right atrium.

Static measurement of IVC diameter and variation with spontaneous respiration has been shown to correlate with central venous pressure (CVP) [47,49,50]. A change in IVC diameter with respiration of 12 to 18 percent has been associated with fluid responsiveness (defined as an increase in CO of >15 percent after a fluid bolus) in mechanically ventilated patients [51,52]. However, in a systematic review of 20 studies that examined the caval index (IVC collapsibility or distensibility) and six that examined the IVC diameter, the sensitivity was only 71 percent and specificity 75 percent [53]. There was significant heterogeneity among the included studies. However, we propose that clinicians should not use the IVC as a sole indicator of volume responsiveness. IVC ultrasound may be most useful at extremes, and when taken into context with prior probability and other tools assessing likelihood of volume tolerance and responsiveness.

A general disadvantage of this technique, and POCUS in general, is that it requires training, and images obtained are operator-dependent. Several patient factors such as pulmonary hypertension, valvular regurgitation, and right ventricular dysfunction may also confound findings. In addition, obtaining accurate serial measurements may be time consuming and difficult.

Lung ultrasonography — Advocates of the concept of "fluid tolerance" believe that patients should receive fluid resuscitation until they develop signs of volume overload, such as pulmonary edema [31,54]. Radiographic and clinical signs of pulmonary edema and clinical evidence of anasarca are late signs of volume overload and poor endpoints for fluid resuscitation [31]. Sonographic assessment of B-lines, indicative of interstitial or alveolar pulmonary edema, and measurement of extravascular lung water (EVLW) are techniques that may aid in the assessment of early volume overload, but is poorly studied [55-57]. (See "Bedside pleural ultrasonography: Equipment, technique, and the identification of pleural effusion and pneumothorax".)

Femoral vein diameter — Preliminary studies measuring femoral vein diameter in mechanically ventilated patients suggested acceptable correlation with central venous pressure measurements but additional studies are warranted to validate these findings [58].

Doppler of portal, hepatic, or renal veins — A novel method of assessing volume tolerance is measurement of Doppler flow in the portal, hepatic, or renal veins. Several studies have documented that pulsatile flows in these vessels may be markers of venous congestion and can be associated with end-organ injury [59,60].

CARDIAC OUTPUT — Several invasive and noninvasive technologies have been developed to measure cardiac output (CO).

Arterial pulse waveform analysis — Several commercially available devices calculate CO based upon the arterial pulse waveform derived from an arterial catheter. A study of 17 postoperative patients that compared some of these devices to the thermodilution method using a pulmonary artery catheter (PAC) found that, although the devices produced similar mean CO values, their dynamic responses and trends correlated poorly with each other [61].

Lithium dilution-based devices — Lithium-based dilution devices use the lithium dilution method for calibration and subsequent measurement of cardiac output. Lithium is injected via a central or peripheral vein, and a lithium analyzer is connected to an arterial line, which measures the wash-out curve over time, generating a curve similar to the thermodilution curve of a pulmonary artery catheter (PAC; that is also used to calculate the CO). Based upon this initial calibration, the root mean square method applied to the arterial pressure signal is used for subsequent measurements, so no additional lithium injections are required. Correlation between lithium dilution and thermodilution has been reported to be acceptable [11,62]. Recalibration must be performed after significant hemodynamic changes or other interventions that alter vascular impedance. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults", section on 'Indicator thermodilution method'.)

Thermodilution-based devices — This device uses pulse contour analysis of the aortic transpulmonary thermodilution curve for initial calibration. Typically, a small volume of cold saline is injected into the central vein. Various hemodynamic parameters can be obtained through analysis of variations in blood temperature taken by the temperature sensor of an arterial catheter. One device produces CO data by determining the area under the systolic arterial waveform with some evidence suggesting that its performance is comparable to thermodilution derived from a PAC [63,64]. As an example, a retrospective study of 46 patients with subarachnoid hemorrhage complicated by Takotsubo cardiomyopathy reported good correlation between pulse contour cardiac output analysis (PiCCO)-based measurements of CO and echocardiography [65]. However, in another study of 25 postoperative patients, thermodilution-based measurements did not correlate well with PAC measurements of CO but reliably tracked changes in CO over time [64].

Arterial waveform-based devices — These devices transduce multiple pressure points along the arterial pressure curve and calculate cardiac index using these data combined with vascular resistance data (calculated from age, gender, height, weight, and body surface area) [66]. Although some studies suggest improved performance among more contemporary devices [20,31,61,67-72], several studies have shown that the devices inconsistently predict CO, fluid responsiveness, and SV in hemodynamically compromised patients receiving fluid boluses or vasoactive agents [66,70,73-75]. These devices may be more promising in the operating room (OR) setting among high-risk patients undergoing noncardiac surgery [76].

Thoracic electrical bioimpedance or bioreactance

●Thoracic electrical bioimpedance (TEB) – Using low-voltage electrodes placed on the chest wall, electrical impedance (ie, opposition of flow to an electrical current) across the thoracic cage is measured. The higher the fluid content within the chest cavity, the lower the impedance, since fluid conducts electricity. As the heart cycles through systole and diastole, the volume of blood in the thorax changes, and this can be measured electrically and extrapolated to determine CO [11].

While early studies demonstrated poor correlation between TEB and invasive measures of CO [11,77], studies since then report improved accuracy, in patients who have recently undergone cardiac surgery [78-83]. It has traditionally been thought that TEB is inaccurate during states of volume overload, but one study demonstrated that it performs well in patients with decompensated heart failure [84].

●Thoracic bioreactance – Thoracic bioreactance is a modification of bioimpedance technology, designed to increase the "signal-to-noise" ratio [66]. Bioreactance technology determines the "phase shift" in alternating current voltage across the thorax. It is proposed that the phase shift almost exclusively depends on pulsatile flow and is therefore less influenced by other intravascular and extravascular fluid in the thorax. Because the overwhelming majority of pulsatile flow in the thorax comes from the aorta, the bioreactance signal correlates with aortic flow.

One commercially available device uses four electrode patches each consisting of two electrodes and calculates CO separately for the right and left side of the body, with the final CO being the average of these two values [66]. This device reports a number of hemodynamic parameters, including CO and SV.

Several studies have demonstrated that CO determined by bioreactance technology correlates with measurements using pulmonary artery catheter thermodilution or pulse contour analysis [33,66,85-87]. However, other studies have reported a poor correlation between bioreactance and thermodilution techniques [88,89]. One observational study demonstrated that this technology, when incorporated into usual clinical care, was associated with lower fluid balance, fewer days requiring mechanical ventilation, shorter ICU length of stay, less hemodialysis, and a shorter time on vasopressors [90].

Electrocautery and external pacemakers interfere with the bioreactance signal, limiting its use in certain locations (eg, OR) and certain patients [66]. Severe aortic insufficiency or other thoracic aorta pathology may impact accuracy.

Aortic Doppler — Aortic Doppler measures blood flow velocity in the aorta by means of a Doppler probe, which may be inserted blindly into the esophagus (esophageal Doppler) or placed on the anterior chest wall (ie, transcutaneous Doppler). With esophageal Doppler (usually performed under sedation in a ventilated patient), the CO is calculated based on the diameter of the aorta, the distribution of the CO to the descending aorta, and the measured flow velocity of blood in the aorta. Transcutaneous devices use Doppler to calculate CO, using a proprietary algorithm that determines velocity-time integral (VTI) measurements in the left and right ventricle outflow tracts. Esophageal Doppler uses similar proprietary algorithms.

Esophageal Doppler has been used with success to guide fluid management in the OR [66,91-93]. Studies of the transcutaneous device have produced varying results [94-97].

The major limitation of this technology is that the Doppler waveform is highly dependent on correct positioning, as it must be well aligned with the direction of blood flow. Poor positioning tends to underestimate true CO.

Point-of-care echocardiography — In addition to a global assessment of ventricle and valvular function in patients with hemodynamic compromise, CO can be assessed using bedside echocardiography. (See "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock" and "Indications for bedside ultrasonography in the critically ill adult patient", section on 'Basic critical care echocardiography'.)

Cardiac output may be calculated by determining the velocity-time integral (VTI) of the spectral Doppler envelope (or tracing), most commonly at the level of the left outflow tract (LVOT). Most ultrasound machines equipped with pulsed wave Doppler may be used. First, the cross sectional area (CSA) of the LVOT is measured, typically in the parasternal short-axis view. Next, a pulsed wave Doppler signal of the LVOT is obtained, usually in the apical five-chamber view to determine the LVOT VTI, from which the CO can be calculated when the heart rate (HR) is known:

Many devices contain software that automate these calculations. Analogous measurements may be taken of the carotid artery.

Trained physicians are capable of determining CO using these methods with fair reliability [98-100]. CO determination using carotid blood flow has been shown to be feasible in a number of disease states, including cardiac arrest [33,101-105]. Serial measurements before and after fluid bolus have been associated with volume responsiveness [34,35].

Point-of-care echocardiography is highly operator-dependent. Serial measurements may be challenging, as slight changes in patient or transducer position may lead to large variations in measurements.

TISSUE PERFUSION — In compensated shock, macrocirculatory measures such as arterial pressure and cardiac output (CO) may be normal in the face of markedly abnormal oxygen delivery and utilization [13]. Devices have been developed to measure indices of shock at the tissue level. (See 'Upstream versus downstream monitors' above.)

Measurement of tissue oxygen saturation — Tissue oxygen saturation (StO₂) measurement using near-infrared spectroscopy (NIRS) has been proposed as a downstream hemodynamic monitoring tool to survey the microcirculation and assess the balance of oxygen delivery and consumption at the tissue level.

StO₂ is measured transcutaneously using NIRS via a number of commercially available devices that measure tissue absorbance values in a defined range of wavelengths.

StO₂ with vascular occlusion test (VOT) has been shown to predict outcome and organ dysfunction in patients with sepsis and congestive heart failure in two small studies, and preliminary studies have demonstrated its usefulness in trauma patients [106-108].

However, the value of StO₂ value is limited because StO₂ remains within normal range until shock is quite advanced. The addition of a dynamic ischemic challenge such as the VOT (application of a tourniquet or sphygmomanometer above systolic arterial pressure for brief, defined intervals) may improve the predictive ability of StO₂ to identify tissue hypoperfusion [109].

Measurement of microcirculatory blood flow — There is considerable interest in shock-induced microcirculatory dysfunction, most notably in the case of sepsis.

The sublingual mucosa is the preferred means to evaluate the microcirculation in critically ill patients because it shares embryological origin with the splanchnic circulation and can be easily accessed at the bedside.

Imaging of the sublingual microvasculature is typically obtained using advanced microscopy techniques, such as sidestream dark field imaging, or by near-infrared spectroscopy (NIRS).

Early studies demonstrated alterations in microvascular flow in patients with sepsis and cardiogenic shock [12,110]. Multiple subsequent studies have demonstrated that alterations in sublingual microcirculatory blood flow are associated with poor outcome among patients with septic shock [111-117].

Capillary refill versus lactate — Lactate has long been touted as a marker for tissue perfusion in shock states (see "Causes of lactic acidosis"). In a multi-center trial, a resuscitation strategy targeted to normalization of capillary refill time was compared with a strategy of normalization of serum lactate (or decrease by 20 percent). Tissue perfusion assessments, with normal capillary refill time were defined as less than or equal to three seconds, and were performed every 30 minutes, compared with serial lactate measures, which were performed every two hours. The peripheral perfusion strategy was associated with less organ dysfunction at 72 hours, less intravenous fluid administration, and a possible 28-day mortality benefit [118,119].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sepsis in children and adults".)

SUMMARY AND RECOMMENDATIONS

●Scope – Central venous and pulmonary artery catheterization (PAC), the traditional tools used for hemodynamic monitoring of patients who present with shock, are invasive and frequently inaccurate. (See 'Introduction' above and 'Deficiencies of standard techniques' above.)

●General principles – Several, less invasive, novel technologies are available or being investigated for use to assess hemodynamic parameters such as cardiac output (CO), intravascular volume status, responsiveness to intravenous fluid administration, and tissue perfusion. Although novel, many of these tools use proprietary algorithms or imaging technology rather than direct measurements, data interpretation can be challenging, and no randomized studies have demonstrated improvement in clinical outcome with their use. (See 'General principles' above.)

●Volume tolerance and fluid responsiveness tools – Several devices have been proposed for assessing volume status and fluid responsiveness. These include measurements of pulse pressure variation (PPV) and stroke volume variation (SVV) as well as ultrasonography of the vena cava and lung. Measurements of PPV and SVV typically require an arterial catheter; they are also limited to patients who are in normal sinus rhythm, mechanically ventilated, and not spontaneously breathing. Ultrasonography is noninvasive but more time consuming and expertise-dependent. Their performance compared with PAC monitoring is variable. (See 'Volume tolerance and fluid responsiveness' above.)

●Cardiac output tools – Several invasive and noninvasive technologies have been developed to measure CO including devices that analyze the arterial waveform, devices that measure thoracic bioimpedance and bioreactance, aortic Doppler, and point-of-care echocardiography. Arterial pulse waveform analysis requires an arterial catheter while the others are noninvasive but require complex algorithms or operator skill. Their performance compared with standard echocardiography or PAC-determined CO is variable. (See 'Cardiac output' above.)

●Tools to assess tissue perfusion – There are few commercially available "downstream" monitoring tools capable of noninvasively assessing oxygen delivery and utilization at the tissue level (eg, near-infrared spectroscopy). Data are lacking regarding their value in the evaluation and management shock at the tissue level. (See 'Tissue perfusion' above.)