C

CE: Namrta; MCC/21306; Total nos of Pages: 6;

MCC 21306

REVIEW

CURRENTOPINION Clinical assessment of peripheral circulation

opyright © 2015 Wolters

1070-5295 Copyright � 2015 Wolte

Alexandre Lima and Jan Bakker

Purpose of review

Monitoring of the peripheral circulation can be done noninvasively in contrast to the more traditionalinvasive systemic haemodynamic monitoring in the intensive care unit. Physical examination of peripheralcirculation based on clinical assessment has been well emphasized for its convenience, accessibility, andrelation to the prognosis of patients with circulatory shock. The purpose of this article is to highlight themain findings according to recent literature into the clinical applications of the peripheral perfusionassessment in patient management.

Recent findings

Clinical assessment of peripheral circulation includes physical examination by inspecting the skin for palloror mottling, and measuring capillary refill time on finger or knee. Studies have addressed the capillary refilltime assessment in adults and its relation to normal range, body site, effect of skin temperature, and itsreliability among examiners. These are easily applicable methods in many circumstances, and it has beenused for predicting unfavourable outcomes in critically ill adult patients. Current studies are ongoing todetermine the effects of different interventions on the clinical parameters of peripheral circulation incritically ill patients during shock resuscitation.

Summary

The feasibility and reproducibility of the clinical assessment of peripheral circulation are substantial, andreliance on capillary refill time, skin temperature, and mottling score must be emphasized and exploited.Incorporating therapeutic strategies into resuscitation protocols that aim at normalizing these peripheralcirculation parameters are being developed to investigate the impact of peripheral perfusion-targetedresuscitation in the survival of critically ill patients.

Keywords

capillary refill time, intensive care medicine, mottling, prognosis, shock

Department of Intensive Care, ErasmusMC, Rotterdam,The Netherlands

Correspondence to Professor Jan Bakker, MD, PhD, FCCP, ErasmusMC University Medical Centre, Dept Intensive Care Adults, P.O. Box2040 - Room H625, 3000 CA Rotterdam, The Netherlands. Tel: +317033629; e-mail: jan.bakker@erasmusmc.nl

Curr Opin Crit Care 2015, 21:000–000

DOI:10.1097/MCC.0000000000000194

INTRODUCTION

The recent advances in diagnostic and monitoringtechnologies have helped intensivists to betterunderstand the complex pathophysiology of acutecirculatory failure. The power and objectivity pro-vided by these new technologies might cause us tothink that physical examination in the intensivecare setting has become obsolete. Much emphasisis given to the global variables of perfusion, whereasrelatively little is said about less vital organs suchas skin and/or muscle. One may argue about theclinical significance of monitoring perfusion ofthese nonvital organs in which blood flow is notcrucial for immediate survival. As the introductionof gastric tonometry has proved to be of prognosticvalue, the field of regional (non)invasive monitor-ing gained wide interest, and studies started toaddress the importance of monitoring peripheralvascular beds more susceptible to hypoperfusion,such as skin, muscle, and gastrointestinal tract.Observations on the behaviour of the peripheral

Kluwer Health, Inc. Una

rs Kluwer Health, Inc. All rights rese

circulation alterations permit the recognition oftwo broad phases during the development of shock,irrespective of initiating factors. There is an initialperiod during which compensatory mechanismspredominate, when the neurohumeral response-induced vasoconstriction preserves the perfusionof the vital organs at the expense of decreasedperfusion in the peripheral tissues. Blood flow vari-ations, therefore, follow a similar response patternin the skin, muscle, and gastrointestinal vascularbeds, which makes these tissues highly sensitive fordetecting occult tissue hypoperfusion during acutecirculatory shock [1–4]. With the progression of

uthorized reproduction of this article is prohibited.

rved. www.co-criticalcare.com

Cop

CE: Namrta; MCC/21306; Total nos of Pages: 6;

MCC 21306

KEY POINTS

� Reliance on simple methods, such as CRT, skintemperature, and mottling score, must be emphasizedand exploited.

� Studies were able to demonstrate that CRT>5.0 s orthe presence of mottling at physical exam in adultpatients with either septic or nonseptic shock was ofgreat value to predict more severe organ dysfunction.

� The effects of common resuscitation strategies (i.e., fluidresuscitation, vasopressors, or vasodilators) should takeinto account the dynamic assessment of peripheralcirculation.

� Therapeutic strategies have been incorporated intoresuscitation protocols aiming at normalizing CRT andmottling to investigate the impact of peripheralperfusion-targeted resuscitation in the survival ofcritically ill patients.

Cardiopulmonary monitoring

circulatory shock and after initiation of appropriatetherapy, the active participation of the peripheralcirculation in supporting tissue perfusion becomesless striking and ultimately disappears. Some patientsenter a phase of stability, and peripheral circulationalterations may no longer reflect the acute com-pensatory mechanisms. Other factors, such as mech-anical ventilation, the use of vasopressors orvasodilators, sedatives, and opiates, can distort theneurohumoral physiologic response. Nevertheless,abnormalities in peripheral circulation may stillpersist, although systemic haemodynamic stabilityhas been reached. Moreover, the persistence of thesealterations has been associated with worse outcomes[5,6]. Therefore, some argue that following normal-ization of perfusion parameters, global systemicparameters are of less importance [7]. Today,monitoring of the peripheral circulation can be donenoninvasively in contrast to the need of intravascularcatheters, transoesophageal probe insertion, or bloodanalysis used in more traditional haemodynamicmonitoring in the ICU. Space limitations preventedus from discussing all noninvasive techniques avail-able; however, we provide a more detailed descrip-tion of the commonly used noninvasive methods ofmonitoring peripheral perfusion elsewhere [8].

From all these noninvasive techniques available,physical examination of peripheral perfusion basedon clinical assessment has been well emphasizedfor its convenience, accessibility, and relation tothe prognosis of patients with circulatory shock.The purpose of this article, therefore, is to highlightthe main developments into the clinical assessmentof peripheral perfusion as reflected in recentliterature.

yright © 2015 Wolters Kluwer Health, Inc. Unaut

2 www.co-criticalcare.com

CLINICAL EXAMINATION OF PERIPHERALCIRCULATION

Clinical examination of peripheral circulationallows rapid and repeated assessment of criticallyill patients at the bedside. Peripheral circulation canbe easily assessed performing a careful physicalexamination by touching the skin or measuringcapillary refill time (CRT). The cutaneous vascularbed plays an important role in the thermoregulationof the body, and this process can result in skinperfusion alterations that have direct effects on skintemperature and colour, that is, a cold, clammy,white, and mottled skin, associated with an increasein CRT. In particular, CRT is defined as the timerequired for a distal capillary bed (i.e., the nail bed)regaining its colour after pressure has been appliedto cause blanching. This concept was first intro-duced by Champion et al. [9] in 1981 as a com-ponent of the international trauma severity score forthe rapid and structured cardiopulmonary assess-ment of trauma patients. As CRT assessment is aneasily applicable method in many circumstances, ithas been used for assessing peripheral perfusionand for predicting unfavourable outcomes in bothcritically ill adults and paediatric patients. Over thepast 30 years, the definition of a delayed CRT hasbeen debated in the literature [10]. Most of what weknow, however, about this simple clinical test comesfrom studies performed in the paediatric popu-lation. For instance, in paediatric patients, severalsystematic reviews of clinical predictors of severecardiovascular dysfunction have highlighted theCRT as a warning signal of a cardiovascular distress[11

&

]. Although much has been published in paedi-atrics, these findings cannot be transferred easilyto adult critically ill patients. In addition, veryfew studies have addressed the CRT assessment inadults and its relation to normal range, body site,effect of ambient or skin temperature, and itsreliability among examiners.

Effect of skin temperature on capillary refilltime

Assessing the skin temperature will assist inevaluating the cause of sluggish capillary refill.Maintaining the same heart rate, cardiac output,and core temperature in a healthy population ofeight volunteers, our group showed that the pres-ence of peripheral vasoconstriction as a result ofbody surface cooling could significantly increasethe CRT to a magnitude of more than 150%(>11.0 s) [12]. Thus, assuming normal core tempera-ture, decreased skin blood flow as cause of delayedCRT can be estimated by measuring skin tempera-ture, as cold extremities reflect constriction of

horized reproduction of this article is prohibited.

Volume 21 � Number 00 � Month 2015

C

CE: Namrta; MCC/21306; Total nos of Pages: 6;

MCC 21306

Clinical assessment of peripheral circulation Lima and Bakker

cutaneous vessels that ultimately decreases theamount of blood volume within peripheral vascu-lature. By contrast, peripheral vasodilation hasthe opposite effect. Inducing peripheral vasodila-tion with nitroglycerin infusion in patients withshock after haemodynamic stabilization and withmuch delayed CRT resulted in significant decreasein CRT in 51% toward normal compared withbaseline values [13]. One should pay attention tohow to evaluate skin temperature. In this regard, atemperature gradient can better reflect changes incutaneous blood flow than the absolute skintemperature itself [8]. Body temperature gradientsare determined by the temperature differencebetween two measurement points such as periph-eral-to-ambient, central-to-toe, and forearm-to-fin-gertip (Tskin-diff). Increased vasoconstriction leadsto decreased skin temperature and limited abilityto regulate core temperature before hypothermiaoccurs. Consequently, core temperature is main-tained at the cost of low skin blood flow, resultingin an increased central-to-peripheral temperaturedifference. As peripheral temperature may be influ-enced by ambient temperature, Tskin-diff may be amore reliable measurement, as the two skin tem-peratures are exposed to the same ambient tempera-ture [14]. Experimental studies have suggested aTskin-diff threshold of 08C for initiating vasocon-striction and of 48C for severe vasoconstriction[15,16]. Therefore, assessing skin temperature bytouching the extremities or measuring a bodytemperature gradient can assist a physician to rec-ognize a clinically acceptable CRT. If the extremitiesare cold, one should expect a delayed CRT. On thecontrary, warm extremities indicate adequatecutaneous blood flow and one should expect anormal CRT, whereas a delayed CRT in this con-dition suggests cutaneous microcirculatory derange-ment [17].

What is a normal capillary refill time inhealthy persons?

The normal range of CRT in adults is still debated.Only two studies have addressed the normal rangeof CRT in the adult population. Schriger and Baraff[18] reported that the upper limit of a normal CRT is4.5 s. In a more recent study, Anderson et al. [19]assessed the difference in normal range of CRT withage in an adult population. The study included 1000participants, and CRT was found to be stronglydependent on age, sex, and ambient temperature,with the upper limit of normal (defined by the 95thpercentile) of 3.5 s. The authors also emphasizedthat using 2.0 s as the upper limit of normal wouldhave misclassified 45% of participants in their study.

opyright © 2015 Wolters Kluwer Health, Inc. Una

1070-5295 Copyright � 2015 Wolters Kluwer Health, Inc. All rights rese

Taken together, these findings show that thedelayed return of more than 2.0 s seems to be oflimited value in adult critically ill patients andnot specific enough to identify patients in cardio-vascular distress.

What is normal capillary refill time incritically ill patients?

Many studies already showed that a CRT >5.0 sfollowing initial haemodynamic optimization dis-criminated haemodynamically stable patients withmore severe organ dysfunction and higher odds forworsening organ failure [20–23].

Serial evaluations of CRT have more prognosticvalue than one time point measurement. In a popu-lation of 111 postoperative patients, delayed CRTwas more often seen in patients with severe com-plications (grade III–V Clavien–Dindo classificationsystem) [23]. Before surgery, there were no differ-ences between groups in CRT. However, at firstday after surgery, CRT was significantly longer(CRT>5.0 s) in patients who subsequently devel-oped severe complications, and this difference per-sisted over time until the fourth day. In anotherrecent study, Hernandez et al. [20] reported that in apopulation of 41 critically ill patients with sepsis-related circulatory dysfunction, 34 (85%) who weresuccessfully resuscitated exhibited normal CRT(<5.0 s) within 6 h, even before normalization oflactate levels.

In severe cardiovascular dysfunction, an evenlonger CRT may have more prognostic value. VanGenderen et al. [22] performed a controlled obser-vational study of mild systemic hypothermia inpatients following out-of-hospital cardiac arrest.In this study, CRT was shorter in survivors at admis-sion and improved even further directly afterrewarming, whereas in nonsurvivors a prolongedCRT persisted during the rewarming period.A positive likelihood ratio was calculated afterrewarming and indicated that CRT response exceed-ing 11.5 s was obtained at least 1.8–17.0 times moreoften in nonsurvivors than survivors. In addition,the authors also found that after rewarming toachieve normal core temperature, SOFA scorewas significantly higher in patients who had aCRT>11.5 s. This was largely attributed to thesignificantly higher scores for respiratory, cardio-vascular, and renal system, although this of coursedoes not indicate causality.

Similar observations were reported in septicshock patients. In a recent article, Ait-Oufellaet al. [24

&&

] investigated the skin CRT in a selectedgroup of septic shock patients. CRT was evaluated in59 patients at intensive care admission and after

uthorized reproduction of this article is prohibited.

rved. www.co-criticalcare.com 3

Cop

CE: Namrta; MCC/21306; Total nos of Pages: 6;

MCC 21306

Cardiopulmonary monitoring

initiation of vasopressor therapy within 24 h. SkinCRT of finger and knee was a strong predictorof 14-day mortality. Patients who persisted withCRT>5.0 s had an odds ratio of dying in 14 daysof 9 [95% confidence interval (CI) 1–61] whenmeasured in the finger and odds ratio of 23 (95%CI 11–1568) when measured in the knee. Amongpatients who died, both knee and index CRTs didnot change and remained prolonged in 82% of thepatients despite resuscitation, demonstrating thatabnormal signs of peripheral perfusion in patientswith no other clinical signs of shock are predictive ofprogressing organ dysfunction. These results alignwith two previous findings of the same group. Toanalyse mottling objectively, Ait-Oufella et al. [25]recently developed a clinical scoring system (from 0to 5) based on the area of mottling from the knees tothe periphery. This group reported that a highermottling score within 6 h after initial resuscitationwas a strong predictive factor of 14-day mortalityduring septic shock, and this was independent ofsystemic haemodynamics. In a second study, apply-ing the same mottling score system, the same groupfound that tissue oxygen saturation measuredaround the knee in septic shock was associated withan increase of the mottling score [26].

More recently, Coudroy et al. [27&

] conducted anobservational study applying this same mottlingscore to investigate the incidence of mottling ina large cohort of 791 critically ill patients and itsimpact on ICU mortality. Skin mottling was definedas a red-violaceous discoloration of the knee visuallyassessed by the nurses regardless of skin temperatureor CRT. This study also included an interobserveragreement assessment between the nurse andthe principal investigator. Although the authorsstudied the nurse–investigator agreement in only16 patients, the reliability of qualitative assessmentof mottling in patients with nonextensive mottling(mottling score <2) was very good (Cohen k of 0.87,95% CI 0.63–1.0). The authors reported a mottlingincidence of 29% in all patients and 49% in patientsadmitted for acute circulatory failure. Patients withskin mottling had more severe disease with higherSOFA and higher SAPS II score. In addition, patientswith mottling had significantly more organ support(mechanical ventilation, vasopressor infusion, orrenal replacement therapy). Using multivariateanalysis, the occurrence of at least one skin mottlingepisode (mottling score>1) over the knee wasassociated with ICU mortality independent fromSAPS II [27

&

].These studies suggest that mottling of the skin

is easily recognized and is often encountered incritically ill patients. Pallor, mottling, and cyanosisare key visual indicators of reduced skin circulation.

yright © 2015 Wolters Kluwer Health, Inc. Unaut

4 www.co-criticalcare.com

It is defined as a bluish skin discoloration thattypically manifests near the elbows or knees andhas a distinct patchy pattern. Mottling is the resultof heterogenic small vessel vasoconstriction and isthought to reflect abnormal skin perfusion. Thisscoring system is very easy to learn, has good inter-observer agreement, and can be used at the bedside.

Reliability of capillary refill time

One concern among physicians is the suspectedlack of reproducibility of CRT within and betweenobservers. In the study by Ait-Oufella et al. [24

&&

],skin CRT of finger and knee was evaluated in59 patients at intensive care admission and afterinitiation of vasopressor therapy within 24 h. Twointensivists blinded to patient treatment quantifiedthe CRT, and the interrater concordance was 80%(73–86) for finger CRT and 95% (93–98) for kneeCRT. In another recent study, van Genderen et al.[23] evaluated 1038 consecutive daily CRT measure-ments in 173 postoperative patients, reporting asubstantial overall agreement between trainedobservers, instructed intensivists, and ICU nursesto determine prolonged CRT (CRT>5.0 s) with aCohen k between-rater correlation coefficient of0.91 (95% CI 0.80–0.97) on first postoperativeday, 0.81 (95% CI 0.65–0.93) on the second post-operative day, and 0.74 (95% CI 0.52–0.89) on thethird postoperative day. With a substantial inter-rater agreement of CRT measurement, these comp-lementary studies provide convincing evidenceof CRT reproducibility and support the strategy ofbedside evaluation of peripheral perfusion withCRT.

CLINICAL APPLICATIONS IN PATIENTMANAGEMENT

From the above, we can conclude that conventionalhaemodynamic parameters must be combined withthe clinical assessment of peripheral circulation tocontinually monitor a critically ill patient as anattempt to optimize resuscitation. Although themechanisms involved in shock resuscitation arenot yet fully understood, it is clear that the persist-ence of abnormal peripheral circulation is associatedwith worse patient outcomes. It is likely that inter-ventions specifically aimed at the peripheral vascu-lar bed would have a greater effect on regionalperfusion. However, before developing specifictherapy that targets peripheral perfusion, it is essen-tial to evaluate the effect of different therapeuticstrategies in critically ill patients, such as fluid resus-citation, vasopressor, or vasodilator therapy. Theconcept of treating peripheral perfusion in critically

horized reproduction of this article is prohibited.

Volume 21 � Number 00 � Month 2015

C

CE: Namrta; MCC/21306; Total nos of Pages: 6;

MCC 21306

Clinical assessment of peripheral circulation Lima and Bakker

ill patients originated in the 1990s with clinicaltrials of different types of therapy, includingvasodilators (prostacyclin and N-acetylcysteine),targeting splanchnic perfusion as assessed bygastric tonometry [28]. These studies demonstratedan improvement in gastric perfusion suggestingthat successful microcirculatory recruitment hadoccurred. More recently, studies have evaluatedsublingual microcirculation during short-term infu-sions of nitroglycerin in septic or nonsepticshock and demonstrated significant improvementsin microcirculation capillary density [29–31]. Arandomized controlled trial in 70 patients withseptic shock failed to show significant differencesin the evolution of the sublingual microcirculationbetween the control and the protocol groups treatedwith a constant dose of nitroglycerin [30]. Althoughthis study precluded the effectiveness of nitrogly-cerin in the sublingual microvascular flow,cutaneous circulation as measured by central-to-toe temperature significantly improved in the nitro-glycerin-treated group compared with the placebogroup. This finding highlights the need foradditional studies to identify the best peripheralvascular bed to target during and after resuscitationtherapy of shock, as well as the type of patientpopulation who may benefit from such an interven-tion. In this line of thought, our group recentlyperformed a crossover study to investigate the effectof intravenous nitroglycerin on clinical parametersin shock patients with abnormal peripheral circu-lation following initial resuscitation [13]. This studyshowed that the use of a stepwise increase in thedose of nitroglycerin (starting 2 mg/h to maximum8 mg/h) normalized the abnormal CRT in all 15patients. These changes occurred in parallel withchanges in Tskin-diff suggesting that the improve-ment in cutaneous microcirculation was likely theresult of increase in cutaneous blood flow. Thenoticeable changes in CRT during nitroglycerininfusion provide evidence that clinical assessmentof peripheral perfusion can be used to titratevasodilator therapy to improve tissue perfusion.However, whether this therapeutic strategy resultsin better clinical outcome still has to be determinedin randomized clinical trials.

Other less specific therapeutic approaches haveshown to have positive effects on peripheral circu-lation. For instance, Futier et al. [32] recently showedthat the administration of a fluid challengeimproved peripheral tissue oxygenation in patientsundergoing major abdominal surgery. Taken onestep further, van Genderen et al. [33

&&

] have exploredthe effects of peripheral perfusion-guided fluidtherapy in patients with septic shock. In a rando-mized, controlled pilot trial, 30 patients were

opyright © 2015 Wolters Kluwer Health, Inc. Una

1070-5295 Copyright � 2015 Wolters Kluwer Health, Inc. All rights rese

randomly assigned to fluid management eithertargeting peripheral perfusion parameters or meanarterial pressure >65 mmHg. The primary studyparameter was the daily fluid balance. In the inter-vention group, patients received less fluids duringthe treatment period, 4227 (�1081 ml) vs. 6069(�1715 ml), and almost 2.5 l less in the followingdays, 7565 (�982 ml) vs. 10 028 (�941 ml). Interest-ingly, patients in the control group stayed longer inthe hospital compared with the intervention groupand had higher odds for organ failure. The con-clusion of the authors was that fluid administrationin the intervention group worked as a safety limitrather than a resuscitation target. Whether theseinterventions are capable of resuscitating differentperipheral vascular beds remains to be determined.Current studies are ongoing to determine the effectsof different interventions on peripheral circulationin critically ill patients.

CONCLUSION

The feasibility and reproducibility of the clinicalassessment of peripheral circulation are substantial,and involved physicians and nurses judge this diag-nostic tool to be easy to use. Perhaps the mostobvious reason why the physical examination willendure is convenience. Reliance on simple methods,such as CRT, skin temperature, and mottling scoremust be emphasized and exploited. To what extentthese abnormalities reflect tissue hypoperfusion inshock remains to be investigated. Nevertheless,most of published reports support our earlier knowl-edge that the vascular bed of peripheral circulationis among the first to deteriorate and the last to berestored after resuscitation. The true clinicalimplications of this strategy may be better definedin the next clinical trials of peripheral perfusion-targeted resuscitation. One next logical step, there-fore, would be incorporating therapeutic strategiesinto resuscitation protocols that aim at normalizing(peripheral) perfusion parameters to investigate theimpact of peripheral perfusion-targeted resuscita-tion in the survival of critically ill patients. Thought-fully integrated with the new technology, theclinical assessment of peripheral perfusion shouldand will continue to be central to intensive careclinical practice.

Acknowledgements

None.

Financial support and sponsorship

This work was supported by the Department of IntensiveCare, Erasmus MC, University Medical Centre, Rotter-dam, the Netherlands.

uthorized reproduction of this article is prohibited.

rved. www.co-criticalcare.com 5

Cop

CE: Namrta; MCC/21306; Total nos of Pages: 6;

MCC 21306

Cardiopulmonary monitoring

Conflicts of interest

There are no conflicts of interest.

REFERENCES AND RECOMMENDEDREADINGPapers of particular interest, published within the annual period of review, havebeen highlighted as:

& of special interest&& of outstanding interest

1. Guzman JA, Dikin MS, Kruse JA. Lingual, splanchnic, and systemic hemody-namic and carbon dioxide tension changes during endotoxic shock andresuscitation. J Appl Physiol 2005; 98:108–113.

2. Guzman JA, Lacoma FJ, Kruse JA. Relationship between systemic oxygensupply dependency and gastric intramucosal PCO2 during progressive he-morrhage. J Trauma 1998; 44:696–700.

3. Shepherd AP, Kiel JW. A model of countercurrent shunting of oxygen in theintestinal villus. Am J Physiol 1992; 262 (4 Pt 2):H1136–H1142.

4. Schlichtig R, Kramer DJ, Pinsky MR. Flow redistribution during progressivehemorrhage is a determinant of critical O2 delivery. J Appl Physiol 1991;70:169–178.

5. Chien LC, Lu KJ, Wo CC, Shoemaker WC. Hemodynamic patterns precedingcirculatory deterioration and death after trauma. J Trauma 2007; 62:928–932.

6. Poeze M, Solberg BC, Greve JW, Ramsay G. Monitoring global volume-related hemodynamic or regional variables after initial resuscitation: what is abetter predictor of outcome in critically ill septic patients? Crit Care Med2005; 33:2494–2500.

7. Dunser MW, Takala J, Brunauer A, Bakker J. Re-thinking resuscitation:leaving blood pressure cosmetics behind and moving forward to permissivehypotension and a tissue perfusion-based approach. Crit Care 2013; 17:326.

8. Lima A, Bakker J. Noninvasive monitoring of peripheral perfusion. IntensiveCare Med 2005; 31:1316–1326.

9. Champion HR, Sacco WJ, Carnazzo AJ, et al. Trauma score. Crit Care Med1981; 9:672–676.

10. Pickard A, Karlen W, Ansermino JM. Capillary refill time: is it still a usefulclinical sign? Anesth Analg 2011; 113:120–123.

11.&

Fleming S, Gill P, Jones C, et al. Validity and reliability of measurement ofcapillary refill time in children: a systematic review. Arch Dis Child 2013;98:265–268.

Although it was performed in a paediatric population, this study has helped shedlight on questions related to the optimum method of measurement of CRT. Thissystematic review should be a reference study for future investigations in the adultpopulation.12. Lima A, van Genderen ME, Klijn E, et al. Peripheral vasoconstriction influences

thenar oxygen saturation as measured by near-infrared spectroscopy. Inten-sive Care Med 2012; 38:606–611.

13. Lima A, van Genderen ME, van Bommel J, et al. Nitroglycerin reverts clinicalmanifestations of poor peripheral perfusion in patients with circulatory shock.Crit Care 2014; 18:R126.

14. Carrillo AE, Cheung SS, Flouris AD. A novel model to predict cutaneous fingerblood flow via finger and rectal temperatures. Microcirculation 2011;18:670–676.

yright © 2015 Wolters Kluwer Health, Inc. Unaut

6 www.co-criticalcare.com

15. Sessler DI. Skin-temperature gradients are a validated measure of fingertipperfusion. Eur J Appl Physiol 2003; 89:401–402.

16. House JR, Tipton MJ. Using skin temperature gradients or skin heat fluxmeasurements to determine thresholds of vasoconstriction and vasodilata-tion. Eur J Appl Physiol 2002; 88:141–145.

17. Lima A, van Bommel J, Sikorska K, et al. The relation of near-infrared spectro-scopy with changes in peripheral circulation in critically ill patients. Crit CareMed 2011; 39:1649–1654.

18. Schriger DL, Baraff L. Defining normal capillary refill: variation with age, sex,and temperature. Ann Emerg Med 1988; 17:932–935.

19. Anderson B, Kelly AM, Kerr D, et al. Impact of patient and environmentalfactors on capillary refill time in adults. Am J Emerg Med 2008; 26:62–65.

20. Hernandez G, Pedreros C, Veas E, et al. Evolution of peripheral vs metabolicperfusion parameters during septic shock resuscitation. A clinical-physiologicstudy. J Crit Care 2012; 27:283–288.

21. Lima A, Jansen TC, van Bommel J, et al. The prognostic value of the subjectiveassessment of peripheral perfusion in critically ill patients. Crit Care Med2009; 37:934–938.

22. van Genderen ME, Lima A, Akkerhuis M, et al. Persistent peripheral andmicrocirculatory perfusion alterations after out-of-hospital cardiac arrest areassociated with poor survival. Crit Care Med 2012; 40:2287–2294.

23. van Genderen ME, Paauwe J, de Jonge J, et al. Clinical assessment ofperipheral perfusion to predict postoperative complications after major ab-dominal surgery early: a prospective observational study in adults. Crit Care2014; 18:R114.

24.&&

Ait-Oufella H, Bige N, Boelle PY, et al. Capillary refill time exploration duringseptic shock. Intensive Care Med 2014; 40:958–964.

The first study to demonstrate the prognostic value of CRT concomitantly mea-sured in two different places in patients with septic shock.25. Ait-Oufella H, Lemoinne S, Boelle PY, et al. Mottling score predicts survival in

septic shock. Intensive Care Med 2011; 37:801–807.26. Ait-Oufella H, Joffre J, Boelle PY, et al. Knee area tissue oxygen saturation is

predictive of 14-day mortality in septic shock. Intensive Care Med 2012;38:976–983.

27.&

Coudroy R, Jamet A, Frat JP, et al. Incidence and impact of skin mottling overthe knee and its duration on outcome in critically ill patients. Intensive CareMed 2015; 41:452–459.

This study is interesting and valuable as it includes a large cohort of 791 critically illpatients.28. Lima A. Circulatory shock and peripheral circulatory failure: a historical

perspective. Neth J Crit Care 2014; 18:14–18.29. Spronk PE, Ince C, Gardien MJ, et al. Nitroglycerin in septic shock after

intravascular volume resuscitation. Lancet 2002; 360:1395–1396.30. Boerma EC, Koopmans M, Konijn A, et al. Effects of nitroglycerin on sub-

lingual microcirculatory blood flow in patients with severe sepsis/septic shockafter a strict resuscitation protocol: a double-blind randomized placebocontrolled trial. Crit Care Med 2010; 38:93–100.

31. den Uil CA, Caliskan K, Lagrand WK, et al. Dose-dependent benefit ofnitroglycerin on microcirculation of patients with severe heart failure. IntensiveCare Med 2009; 35:1893–1899.

32. Futier E, Christophe S, Robin E, et al. Use of near-infrared spectroscopyduring a vascular occlusion test to assess the microcirculatory responseduring fluid challenge. Crit Care 2011; 15:R214.

33.&&

van Genderen M, Noel E, van der Valk RJ, et al. Early peripheral perfusion-guided fluid therapy in patients with septic shock. Am J Respir Crit Care Med2015; 191:477–480.

This is the first clinical trial to demonstrate the impact of fluid therapy targeting CRTduring resuscitation of patients with septic shock.

horized reproduction of this article is prohibited.

Volume 21 � Number 00 � Month 2015