What Is a Normal Capillary Refill Time in a Pediatric Patient?

A normal capillary refill time in a pediatric patient is generally considered to be two seconds or less when measured at the fingertip, though the actual number shifts depending on where on the body you press, the child’s age, and even the temperature of the room. That two-second cutoff is baked into clinical guidelines and early warning scores worldwide, yet the evidence behind it is messier than the clean threshold suggests. Where you test, how hard you press, and what you’re looking for all change the answer in ways that matter at the bedside.

The Two-Second Rule and Where It Comes From

Capillary refill time, or CRT, is measured by pressing on the skin until it blanches white, then counting how long it takes for color to return after you release. A value above two seconds is widely taught as prolonged and is included in pediatric early warning scores and guidelines for identifying serious illness such as sepsis or dehydration.¹ The test is fast, free, and requires no equipment, which is why it remains a staple of pediatric assessment in emergency departments, outpatient clinics, and field triage settings alike.

That said, the two-second threshold was not derived from a single landmark study that definitively mapped out normal values across all ages. It emerged over decades of clinical teaching and was eventually codified into guidelines, sometimes without rigorous validation against patient outcomes. A systematic review of CRT measurement in children confirmed that in children older than seven days of age, the upper limit of normal is roughly two seconds when tested on a finger, but up to four seconds when tested on the chest or foot.² Those are meaningfully different numbers tied to the same child, which is the first sign that “two seconds” isn’t quite the universal answer it sounds like.

Where You Press Changes the Number

CRT measured at the fingertip is consistently faster than CRT measured at the sternum. One study of children found the average fingertip CRT was about one second, while the sternum averaged roughly one and a half seconds, and the two measurements were only weakly correlated with each other.³ In practical terms, you cannot measure CRT on the chest and compare it against a threshold that was established using fingertip data, or vice versa. The two sites are not interchangeable.

This matters because different clinical settings favor different sites. A child whose hands are tucked under a blanket or wrapped in a splint might get tested on the chest. A quick triage in a busy emergency department might default to the fingertip. If the clinician doesn’t mentally adjust the threshold based on location, a perfectly healthy child could register as abnormal on the sternum, or a genuinely underperfused child could seem fine on the fingertip. The systematic review mentioned above found the upper limit of normal stretches to about four seconds on the chest and foot, roughly double the fingertip cutoff.²

Newborns Play by Different Rules

The two-second threshold applies well to infants and older children, but newborns in the first week of life tend to run longer. A study of healthy neonates found the average CRT was about 2.2 seconds, and the authors proposed an upper limit of normal of three seconds rather than two.⁴ Within that population, values did not change meaningfully based on gestational age, sex, or birth weight.⁵

The extremities of newborns are a separate story altogether. CRT measured on the hands and feet of normal newborns averaged over four seconds, with wide ranges extending past eight or nine seconds, and these values did not change significantly in the first 72 hours of life.⁶ A four-second CRT on the finger of an older child would be alarming; on the hand of a day-old baby, it is normal. Clinicians assessing perfusion in newborns therefore tend to test central sites like the sternum or forehead rather than the extremities. One study in newborns found that head and chest CRT measurements clustered more tightly around a normal distribution, while heel measurements were widely scattered and harder to interpret.⁷

Room Temperature Can Wreck the Test

Of all the things that confound CRT measurement, ambient temperature is the most dramatic. A study of healthy children found that the average CRT in a warm room (around 26°C) was under a second, while the same children in a cool room (around 19°C) averaged nearly 2.4 seconds. Every single child registered as normal in the warm setting, but only about a third were classified as normal in the cool one.⁸ That temperature range is not extreme; it falls within what you might encounter moving between a heated exam room and a drafty hallway or triage area. The authors were blunt that this temperature sensitivity raises questions about CRT’s reliability in real-world settings, especially prehospital ones where temperature control is limited.

Research in adult and mixed populations tells a similar story. After cold immersion, CRT values more than doubled compared to baseline.⁹ The implication is straightforward: a child brought in from cold weather, or examined in an air-conditioned ambulance, can have a CRT that looks prolonged purely because of peripheral vasoconstriction from the cold. The same child examined five minutes later after warming up might test perfectly normal. Clinicians who rely heavily on CRT in cold environments risk false alarms.

The ideal testing environment is a room between 20°C and 25°C. The systematic review of CRT in children flagged that ambient temperatures outside this range are associated with longer refill times regardless of the child’s clinical status.² When testing conditions fall outside that window, CRT should be interpreted cautiously and ideally combined with other clinical signs.

When Two Clinicians Disagree on the Same Child

CRT seems like a simple, objective test, but studies on interobserver agreement paint a less reassuring picture. When two observers measured CRT in healthy volunteers, the limits of agreement between them were nearly two seconds, whether they used a stopwatch or video analysis.¹⁰ For a test where the entire normal-to-abnormal threshold is two seconds, a two-second margin of disagreement between clinicians is a problem.

A more recent experimental study specifically looking at naked-eye CRT assessment in children found poor agreement across all professional groups tested, with limits of agreement ranging from about 1.2 seconds to 2.0 seconds.¹¹ Even video-based CRT assessment, where multiple raters reviewed the same recorded clips, showed only moderate agreement, and the biggest source of variability was the individual child being assessed rather than the rater, suggesting that some children just produce ambiguous color-return patterns.¹²

Part of the issue is that people press with different force and for different durations. One study collecting data on clinician pressing force during CRT measurement recorded an average of about 13 newtons, but individual variation around that average was substantial.¹³ Pressing too lightly may not fully blanch the nailbed; pressing too hard or too long can prolong the refill. There is no universally standardized pressing protocol. The systematic review found that longer pressing times were associated with longer CRT values, which means two clinicians using different techniques will produce different readings on the same child at the same moment.²

Skin Color, Lighting, and Visibility Challenges

CRT is fundamentally a visual test: you watch for color to return to blanched skin. That raises obvious questions about whether the test works equally well across different skin tones and lighting conditions. When researchers used a pulse oximeter-based device to measure CRT and compared it to clinician assessment, the difference between device and clinician readings did not vary significantly across fair, medium, and dark skin.¹⁴ That’s somewhat reassuring for device-assisted measurement, but it says less about whether unaided clinician assessment is equally accurate across skin tones.

A scoping review looking specifically at clinical deterioration assessment in children with dark-colored skin found no published studies reporting on CRT assessment in that population. The authors flagged CRT as one of several assessment tools with specific information gaps for children with darker skin.¹⁵ The absence of research is itself the finding here. Clinicians working with diverse patient populations should be aware that the evidence base for CRT was built predominantly in lighter-skinned populations, and the applicability of visual CRT assessment to children with darker skin tones is simply understudied.

Lighting compounds the problem. A study testing CRT assessment under different conditions found that in daylight, observers detected normal CRT in about 94% of participants. In low light conditions mimicking a dimly lit ambulance or nighttime field setting, two-thirds of assessments resulted in the observer being unable to detect capillary refill at all.¹⁶ This is not a subtle effect. In poor lighting, CRT is effectively useless for most patients.

What a Prolonged CRT Tells You Clinically

Despite all the caveats about measurement, a genuinely prolonged CRT still carries real diagnostic weight. It is one of the few bedside signs that can help detect significant dehydration in children. A systematic review and meta-analysis of CRT’s ability to detect serious illness in children found that for dehydration of 5% or greater (the threshold where children typically need intravenous fluids), a prolonged CRT substantially increased the probability that the child was significantly dehydrated. Across studies of children with gastrointestinal symptoms in emergency departments, the specificity was high, ranging from 88% to 94%.¹⁷ In other words, a prolonged CRT in a vomiting, diarrheal child is a fairly reliable red flag. Interestingly, this pattern broke down for dehydration caused by diabetic ketoacidosis, where CRT was far less useful because the mechanism of fluid loss is different.

Digital measurement tools improved accuracy further. One study using an imaging technique to quantify CRT found the digitally measured version had a positive likelihood ratio of nearly 12 for detecting 5% or greater dehydration, roughly triple the value of conventional visual CRT assessment.¹⁸ The takeaway is that CRT has genuine diagnostic power for dehydration, but the precision of the measurement matters enormously.

In sepsis, CRT reflects what is happening in the smallest blood vessels. Children with sepsis who had prolonged CRT showed fewer small capillaries recruited and more damage to the glycocalyx, the protective lining of blood vessels. When CRT remained prolonged at 24 hours after admission, children had worse outcomes and longer hospital stays compared to those whose CRT had normalized.¹⁹ CRT normalization has become a therapeutic target in some pediatric sepsis management protocols, essentially serving as a bedside proxy for whether the microcirculation is recovering.

CRT and What It Does Not Tell You

Clinicians sometimes treat CRT as a window into how hard the heart is pumping, but the relationship between CRT and traditional hemodynamic measures is weak. A study of children in pediatric intensive care found that CRT correlated poorly with cardiac output, blood pressure, and other hemodynamic variables, particularly in children recovering from cardiac surgery.²⁰ In general (non-surgical) ICU patients, children with prolonged CRT did have lower stroke volume, but the correlation was still not strong enough to predict specific hemodynamic values from a CRT reading alone.

This makes physiological sense. CRT reflects what is happening in the tiny capillary beds at the skin surface, which is governed not just by how much blood the heart is pushing out, but by local vascular tone, autonomic nervous system responses, skin temperature, and the integrity of the endothelial lining. A child on vasopressors, for example, might have adequate cardiac output but prolonged CRT because the medications are constricting peripheral blood vessels. CRT is best understood as a test of microvascular perfusion and reactivity, not as a measure of the whole cardiovascular system.²¹

Automated Measurement and the Push for Precision

Given the reliability problems with naked-eye CRT assessment, researchers have been developing automated alternatives. These include smartphone-based imaging tools, pulse oximeter attachments, and dedicated devices that standardize both the pressing force and the optical detection of color return. The motivation is clear: if the main weakness of CRT is inconsistency in how different people perform and read the test, removing the human element from the measurement itself should help.

Early results from device-based CRT measurement have been promising. The imaging technique that achieved a likelihood ratio nearly three times higher than conventional visual assessment for detecting dehydration suggests that the biological signal in CRT is strong; it is the measurement noise that lets the test down.¹⁸ Work on automated finger compression devices has begun establishing reference values for clinician pressing force, laying the groundwork for devices that apply a standardized amount of pressure.¹³

There is also interest in integrating CRT into point-of-care devices that combine it with other noninvasive measurements. One study tested a model combining CRT data with other bedside parameters for dehydration assessment and found moderate discriminating ability even when blood pressure was excluded from the model, which could be relevant for home or telehealth settings where standard clinical equipment is unavailable.²² The vision, still somewhat aspirational, is a parent pointing a smartphone at a child’s finger and getting a reliable perfusion reading that a pediatrician could act on remotely. The technology is not there yet for routine clinical use, but the trajectory is toward making CRT a quantified, standardized measurement rather than a subjective impression.

1
PubMed Central. Capillary refill time in sick children: a clinical guide for general practice
2
Archives of Disease in Childhood. Validity and reliability of measurement of capillary refill time in children: a systematic review
3
PubMed. The agreement of fingertip and sternum capillary refill time in children
4
PubMed Central. Determinants of Capillary Refill Time in Healthy Neonates
5
PubMed Central. Capillary refilling time in newborn babies: normal values
6
PubMed. Capillary refill time in the hands and feet of normal newborn infants
7
Acta Paediatrica. Capillary refilling time in newborns–optimal pressing time, sites of testing and normal values
8
PubMed. Effect of ambient temperature on capillary refill in healthy children
9
PubMed. Defining normal capillary refill: variation with age, sex, and temperature
10
PubMed Central. Lack of agreement between different observers and methods in the measurement of capillary refill time in healthy volunteers: an observational study
11
PubMed Central. Can we trust naked eye assessments of the capillary refill test in children? An experimental study
12
Pediatric Emergency Care. Comparison of Bedside and Video-Based Capillary Refill Time Assessment in Children
13
PubMed. Evaluation of Automated Finger Compression for Capillary Refill Time Measurement in Pediatrics
14
Circulation. Abstract 276: Impact of Skin Color on Accuracy of Capillary Refill Time Measurement by Pulse Oximeter
15
PubMed Central. Assessing Clinical Deterioration in Children With Dark-Coloured Skin: A Scoping Review
16
PubMed. Adverse lighting condition effects on the assessment of capillary refill
17
PLOS ONE. The Diagnostic Value of Capillary Refill Time for Detecting Serious Illness in Children: A Systematic Review and Meta-Analysis
18
Pediatrics. A Novel Imaging Technique to Measure Capillary-Refill Time: Improving Diagnostic Accuracy for Dehydration in Young Children With Gastroenteritis
19
PubMed Central. The association between prolonged capillary refill time and microcirculation changes in children with sepsis
20
PubMed Central. Capillary refill and core-peripheral temperature gap as indicators of haemodynamic status in paediatric intensive care patients
21
Europe PMC. The intricate relationship between capillary refill time and systemic hemodynamics in septic shock
22
Pediatric Emergency Care. Point-of-Care Noninvasive Technology for Pediatric Dehydration Assessment