|Home | About | Journals | Submit | Contact Us | Français|
Monitoring of tissue perfusion is an essential step in the management of acute circulatory failure. The presence of cellular dysfunction has been a basic component of shock definition even in the absence of hypotension. Monitoring of tissue perfusion includes biomarkers of global tissue perfusion and measures for assessment of perfusion in non-vital organs.
The presence of poor tissue perfusion in a shocked patient is usually associated with worse outcome. Persistently impaired perfusion despite adequate resuscitation is also associated with worse outcome. Thus, normalization of some perfusion indices has become one of the resuscitation targets in patients with septic shock.
Although the collective evidence shows the clear relation between impaired peripheral perfusion and mortality, the use of different perfusion indices as a resuscitation target needs more research.
Monitoring of tissue perfusion is an essential step in the management of acute circulatory failure. The presence of cellular dysfunction has been a basic component of shock definition even in the absence of hypotension . The ideal parameter for tissue perfusion should be rapid, non-invasive, and easily measured without the need of advanced skills. Evaluation of tissue perfusion includes clinical evaluation as well as biomarkers. Popular biomarkers of tissue perfusion such as serum lactate and central venous oxygen saturation are indicators of global tissue perfusion. Monitoring of peripheral circulation especially in non-vital organs added new insights for monitoring of tissue perfusion. Peripheral non-vital organ perfusion deteriorates earlier and improves later than vital organs. Assessment of peripheral circulation has become easier after introduction of new non-invasive devices as well as clinical scoring systems.
In this article, we provide a comprehensive updated review for all the available indices for monitoring of tissue perfusion. The advantages and disadvantages of each method will be mentioned. Thoughts for future research and gaps in literature will also be highlighted (Table 1).
The balance between oxygen delivery (DO2) and oxygen consumption (VO2) considered the mainstay of understanding the concept of tissue perfusion and the development of organ dysfunction. In a steady state, the VO2 constitutes only 25% of DO2. In a shock state, the VO2 increased out of proportion of DO2 to the point that DO2 falls below a critical threshold where the VO2 is dependent on DO2. Below that point, organ perfusion will be critically impaired and transition to anerobic metabolism will occur .
Because the main determinant of DO2 is the cardiac output (CO), most patients’ evaluation and resuscitation strategies depend on evaluating and optimizing this parameter. Assessing CO is performed by invasive, minimally invasive, and non-invasive methods . More recently, surviving sepsis campaign recommended the routine use of echocardiography during initial assessment of patients with septic shock .
Lactate is the end product of anerobic glycolysis . Serum lactate level increases in states of cellular hypoxia or low peripheral perfusion; thus, serum lactate level is considered a surrogate of cellular perfusion .
Lactate is the most frequently used marker of tissue perfusion . Lactic acidosis is a predictor of in-hospital mortality in septic shock [7, 8]. Increased lactate clearance during resuscitation of septic shock was associated with improved outcomes .
It is minimally invasive but needs available blood gas analysis.
Mixed venous oxygen saturation (SvO2) is an indicator for adequate oxygen delivery (DO2) ; thus, the change in SvO2 reflects the change in cardiac output as long as other determinants of DO2 are stable . In healthy individuals, SvO2 value is 75% while in critically ill patients, its value is 70% . Central venous oxygen saturation (ScvO2) is another indicator for oxygen delivery; however, the evidence on the correlation between ScvO2 and SvO2 is not clear [12, 13]. ScvO2 reflects perfusion status in the upper body and is not affected by blood coming from the lower body nor coronary sinus . In healthy individuals, ScvO2 is lower than SvO2 ; however, in shocked patients, ScvO2 may exceed SvO2 by up to 20% due to redistribution of blood to the upper body [16, 17].
Although ScvO2 and SvO2 are not interchangeable, ScvO2 is still useful as a predictor of outcome in septic shock. Persistently low ScvO2 during resuscitation is associated with poor outcomes in septic shock patients [18, 19]. Despite its clinical value, maintenance of ScvO2 is no longer a resuscitation goal in the new surviving sepsis campaign guidelines .
Despite the limited value of ScvO2 under general anesthesia, it was reported as a valuable measure in cardiac surgery patients where lower ScvO2 values were associated with more complications ; moreover, in a randomized controlled trial, maintenance of ScvO2 of at least 70% was associated with lower complications in cardiac surgery patients .
In addition to the need to central venous catheter, ScvO2 is usually less informative under general anesthesia because the use of neuromuscular blockers decreases oxygen consumption; moreover, the use of high fraction of inspired oxygen usually (FiO2) corrects hypoxemia and increases DO2 .
The difference between PCO2 in central venous blood and PCO2 in arterial blood is known as central-venous-arterial CO2 gap (P (v-a) CO2). P (v-a) CO2 has been considered as an indicator of the adequacy of venous blood flow to wash out CO2 in peripheral tissues . Elevated P (v-a) CO2 (above 6 mmHg) occurs in cases of decreased systemic blood flow. Normalization of P (v-a) CO2 during resuscitation was associated with normalization of serum lactate .
P (v-a) CO2 is negatively correlated with cardiac output in septic shock patients ; thus, P (v-a) CO2 is a useful parameter to assess the adequacy of tissue perfusion during resuscitation of patients in septic shock . Persistence of high P (v-a) CO2 during early resuscitation of septic shock is associated with poor outcomes .
In cases with decreased ScvO2, a high P (v-a) CO2 denotes that tissue hypoperfusion is due to inadequate cardiac output, whereas normal P (v-a) CO2 denotes that improving cardiac output is unlikely to improve the oxygen delivery to the tissues . In cases of normal ScvO2 and P (v-a) CO2 with evidence of anerobic metabolism (i.e., elevated serum lactate), manipulation of macrocirculation would be not effective to improve the condition because this denotes mitochondrial or microcirculatory disturbances .
Assessment of P (v-a) CO2 needs the presence of both arterial and central venous lines.
Left ventricular longitudinal strain (LVS) is defined as “the percentage distance shortening of the endocardium along its length” . The traditional parameter for assessment of left ventricular function is ejection fraction; however, this parameter could be confounded by cardiac preload and heart rate; moreover, it is highly operator dependent; thus, LVS appears as a novel method for evaluation of systolic function that could help in early detection of myocardial ischemia .
In a prospective observational study , the ability of LVS as a measure for oxygen delivery was evaluated. LVS was associated with elevated serum lactate and low ScvO2 in patients with septic shock. The findings of the aforementioned study suggest that LVS could be a non-invasive surrogate for assessment of tissue perfusion.
There is no available evidence.
Skin temperature is a traditional sign of peripheral vasoconstriction. Critically ill patients with cold skin temperature have lower cardiac index, lower Svo2, and higher serum lactate compared to patients with warm skin temperature . Temperature gradients are better subjective methods for assessment of peripheral perfusion [33, 34].
Central-to-toe temperature (T c-toe) is the difference between central temperature measured at the tympanic membrane and temperature at the ventral surface of the big toe measured by a skin probe. T c-toe gradient has been used as a measure of peripheral vasoconstriction; however, it has the disadvantage of being affected hypothermia as well as room temperature .
The temperature gradient between fingertip and forearm (T skin-diff) has also been reported as a marker of peripheral vasoconstriction. T skin-diff is the difference between the temperatures at the index and at the radial side of the forearm measured using two skin probes. T skin-diff has the advantage of not being affected by the ambient temperature because the change in ambient temperature affects both fingertip and forearm equally [34, 35].
T c-toe more than 7 °C and T skin-diff more than 0 °C are indicators for vasoconstriction . T skin-diff more than 2 and 4 °C indicates the presence of intermediate and severe vasoconstriction respectively . There is an association between T skin-diff and poor outcomes in abdominal surgery patients  and in patients with acute circulatory failure.
The persistence of peripheral perfusion alterations after reversal of therapeutic hypothermia are associated with poor outcomes in cases of out-of-hospital arrest . In cardiac surgery patients, the transition from peripheral vasoconstriction to vasodilatation (measured by skin surface gradient) is associated with shorter time to extubation after both normothermic and hypothermic cardiopulmonary bypass .
It is a non-invasive parameter. T c-toe is affected by hypothermia and room temperature.
Skin mottling is defined as “patchy skin discoloration.” Skin mottling usually manifests around the knees and might extend to other sites of peripheral circulation such as fingers and ears . Skin mottling is a result of heterogeneous small vessel vasoconstriction . Skin mottling is an easily assessed sign of peripheral hypoperfusion .
A six-grade scoring system (ranging from 0 to 5) has been introduced by Ait-Oufella and colleagues depending on the extent of skin mottling around the knee . Skin mottling score (SMS) ranges from 0: no mottling, 1: a coin sized mottling area at the knee center, 2: a mottling area that does not extend above the upper margin of knee cap, 3: a mottling area localized to the lower thigh, 4: a mottling area reaching the upper thigh below the groin fold, to 5: extensive mottling reaching above the groin fold . SMS could be subjective estimate of the severity of peripheral hypoperfusion . Higher SMS (scoring 4 to 5) was associated with poor patient outcome in general ICU population  and in patients with septic shock . A good correlation was reported between changes in both mottling score and skin perfusion (assessed by laser Doppler) during resuscitation of septic shock patients . Prolonged skin mottling more than 6 h was associated with ICU mortality independently from severity scoring systems .
There is no available evidence.
It is a non-invasive parameter. It has a limited value in burns, amputations, and in dark skin.
CRT is defined as “the time needed for skin’s color to return to baseline on a finger’s tip after application of blanching pressure” . CRT can be clinically measured over the fingertip [35, 43] or over the knee area . CRT is a measure of peripheral capillary blood flow .
CRT showed a good correlation with urinary output and serum lactate levels . Prolonged CRT has been associated with poor outcomes in septic shock patients  and in non-selected critically ill patients  with a cutoff value that ranged between 2.5 and 5 s. Normalization of CRT was associated with improved survival rate in septic shock patients .
Prolonged CRT has also been associated with poor outcomes in patients following abdominal surgery .
It is a non-invasive parameter. Index CRT has the disadvantage of inter-rater variability. Knee CRT has a limited value in dark skin.
PPI represents “the ratio between the pulsatile and non-pulsatile component of the light reaching the pulse oximeter” . As the pulsatile (arterial) flow is the only portion affected with vasoconstriction and vasodilatation, PPI has been considered as a numerical non-invasive measure for peripheral perfusion. PPI decreases in states of hypoperfusion due to decreased pulsatile component with a constant non-pulsatile component of blood flow .
The change in PPI reflects the change in other measures of hypoperfusion such as temperature gradients , lactate, and P (v-a) CO2 . In critically ill patients, PPI less than 1.4 is a marker of hypoperfusion ; also, PPI less than 0.6 is an independent factor for 30-day mortality . In septic shock patients, PPI less than 0.3 predicted vasopressor therapy  and below 0.2 predicted mortality .
PPI is a non-invasive measure; however, it needs a special pulse oximeter. PPI is characterized by high skewness  and high inter-individual variability ; thus, it is more useful in trend monitoring for follow-up rather than being used as a single measure .
StO2 is measured using near-infrared spectroscopy (NIRS) . Measurement by StO2 is based on the difference in light absorption between oxy- and deoxyhemoglobins . StO2 represents the hemoglobin oxygen saturation in the tissues . The normal value for StO2 is 87% which represents the sum of arterial, venous, and capillary blood oxygen saturation . The most common sites for measurement of StO2 are thenar and frontal. StO2 decreases in states of hypoperfusion and hypoxia .
Thenar StO2 decreases in situations of hypoperfusion according to shock severity [51, 52]. In trauma patients, thenar StO2 could discriminate patients in severe shock . Low thenar StO2 during initial resuscitation of multiple trauma patients correlate with future development of multiple organ dysfunction .
StO2 could predict the need of blood transfusion in trauma patients . In cardiac surgery, cerebral oxygenation below 50% predicted poor postoperative cognitive dysfunction ; moreover, cerebral oxygenation showed good correlation with jugular venous saturation .
StO2 is a non-invasive parameter; however, it has some limitations: firstly, StO2 showed high variability among healthy volunteers, patients with mild and moderate shock; thus, its value is restricted to patients with severe shock ; secondly, this high variability makes StO2 a useful parameter for trend monitoring rather than a single measure value . Thirdly, StO2 is of low value in patients with septic shock [6, 57].
Subcutaneous partial oxygen pressure (Ptco2) was previously measured using invasive subcutaneous probes ; however, newer technology facilitated Ptco2 measurement using non-invasive transcutaneous probes [59, 60]. Ptco2 has been more popular in neonates than adults . The limited use of Ptco2 in adults is because of the lack of agreement between its value and arterial oxygen tension due to the thicker epidermis in adults compared to neonates .
Despite its low accuracy in adults, Ptco2 has been reported as a useful predictor for outcomes in critically ill emergency patients [59, 60]. A newer parameter, the oxygen challenge test (OCT), was reported as a tool for early diagnosis of poor peripheral perfusion [48, 62, 63]. OCT the Ptco2 response to increasing Fio2 to 1 for a 5-min duration. The Ptco2 increased with increasing FiO2 in non-shocked patients, whereas Ptco2 poorly responded to increasing FiO2 in shocked patients. An increase of 21 mmHg in Ptco2 after OCT was associated with better patient outcome . Poor response to OCT was a predictor of mortality when reported before [62, 63] or after  resuscitation. In a randomized controlled trial, Hu et al.  investigated the use of OCT as a resuscitation target in patients with severe sepsis; they reported that a response of 25 mmHg or more to OCT would improve the patient outcome.
There is no available evidence.
Although it is a non-invasive measure, Ptco2 has some limitations. Ptco2 needs special monitor and electrodes. Moreover, there is no definitive cutoff value. The value of OCT is restricted to intubated patients only.
Visualization of sublingual microcirculation using handheld microscopes has recently gained widespread attention. Direct visualization of microcirculation focuses on vascular density, flow, and proportion of perfused vessels . Assessment of microcirculation has been a research tool not applicable in the clinical bedside assessment. With the introduction of more advanced microscopes (CytoCam), bedside real-time assessment of microcirculation has been available with adequate agreement with the conventional image analysis .
In a cross-sectional multicenter observational study, Lima et al.  investigated the inter-rater reliability for 45 physicians and 16 nurses in subjective assessment of sublingual microcirculation; participants were asked to categorize the videos into good, bad, and very bad microcirculation. Results of the aforementioned study showed good agreement between the participant assessment and the conventional analysis . A recent protocol for real-time assessment of microcirculation was recently introduced by Naumann et al. . Point of care microcirculation (POEM) is a 5-point scoring system that could be used for assessment of flow and heterogeneity with accepted inter-rater agreement .
It is still under research.
There is no available evidence.
Assessment of sublingual microcirculation is a non-invasive procedure. However, it has the disadvantage of the need of an expensive device.
Although normalization of central indices of tissue perfusion (such as lactate and ScvO2) has been an essential target of resuscitation of patients with septic shock , studies validating the value of targeting an improved peripheral perfusion markers are sparse. In an observational study, Hernandez and colleagues reported that early recovery of peripheral perfusion indices (such CRT, T c-toe, and P (v-a) CO2) is a marker of successful resuscitation of septic shock patients. In an animal study, Van Genderen et al.  reported a relationship between central and peripheral markers of perfusion during resuscitation of experimental shock; however, the rate of normalization of perfusion markers was dependent on the underlying type of shock. A randomized controlled trial (RCT) investigated the perfusion-based approach during resuscitation of 30 septic shock patients was conducted by the same group of authors . The aforementioned study reported that the perfusion-based approach resulted in lower fluid replacement, less organ dysfunction, and shorter length of stay . Larger RCTs are recommended to validate the perfusion-based approach and to set more clear steps of resuscitation.
Monitoring of tissue perfusion includes biomarkers of global tissue perfusion, measures for assessment of perfusion in non-vital organs, and direct visualization of sublingual microcirculation.
Three main clinical implications for perfusion indices were reported. First, the presence of poor tissue perfusion in a shocked patient is usually associated with worse outcome. Second, persistently impaired perfusion despite adequate resuscitation is associated with worse outcome. Third, normalization of some perfusion indices (lactate and central venous oxygen saturation) has become one of the resuscitation targets in patients with septic shock.
Direct visualization of sublingual circulation represents a promising tool for evaluation of peripheral perfusion. Simple and user-friendly protocols are being introduced to facilitate the use of sublingual circulation as a bedside tool for assessment of shocked patients.
Although the collective evidence shows the clear relation between impaired peripheral perfusion and mortality, the use of different perfusion indices as a resuscitation target was not adequately investigated. Most of the data concerning the perfusion indices are extracted from observational trials with a few number of randomized controlled trials.
We would like to thank Dr. Ahmed Shash professor and head of Department of Anesthesia and Critical Care Medicine, Cairo University for his continuous support.
This work was funded by Cairo University.
AH was responsible for writing the manuscript and collection of the data. AM was responsible for conception of the topic and revision of the manuscript. HN helped in writing and revising the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Ahmed Hasanin, Email: moc.liamg@ninasahdemahomdemha.
Ahmed Mukhtar, Email: email@example.com.
Heba Nassar, Email: moc.oohay@10rassanabeh.