Capillary refill time (CRT) is a simple and rapid test that provides a useful indication of the adequacy of tissue perfusion…or is it? This article will highlight many of the limitations and variations that need to be considered when undertaking this assessment. Much of the evidence is dated but human physiology has not changed that much over the last 20 years so it reasonable to accept that those findings remain current today.
CRT is defined as the time taken for a distal capillary bed to regain its colour after pressure has been applied to cause blanching (Pickard et al, 2011). Traditionally, CRT has been taught as a tool for rapid cardiovascular assessment and an indication of the adequacy of tissue perfusion (often after traumatic injury). The physiological principles that underpin peripheral perfusion are complex and affected by many different factors. Although the test may be simple, its interpretation may not be.
Normal CRT is proposed to be less than 2 seconds (typically at the finger), with any time longer than this deemed as prolonged or abnormal. However, there is considerable debate regarding the accuracy of the 2-second cut-off time. Lewin and Maconochie (2007) suggest that 2 seconds is an arbitrary figure rather than an accurate one derived from experimental data, which calls into question the entire measurement. It also seems unlikely that a single cut-off point can usefully be applied to the entire population as there are multiple factors that can affect outcomes. CRT can be affected by age, sex, ambient and body temperature, environment and other factors such as the use of inotropes. Anderson et al (2008) undertook a study with 1000 people, concluding that its suitability as a reliable clinical test is doubtful. A recent systematic review and meta-analysis also questioned its prognostic value (Jaquet-Lagrèze et al, 2023).
Factors affecting CRT measurement
Age
Age affects CRT measurement in a variety of ways that will be discussed in this section.
Neonates
The upper limit of normal for CRT in neonates has been suggested by one study to be 3 seconds irrespective of sex, gestation, weight, size for gestational age, or whether located in an incubator, radiant warmer, or crib (Strozik et al, 1993). Another study questions the use of CRT at all in the first 5 days of life, owing to the wide range of ‘normal’ values (Raju et al, 1999).
Children
In children, an upper limit of normal (in which 95% of the normal population lies) of 2 seconds has been reported (Fleming et al, 2015a). Schriger and Baraff (1988) found that in children up to 12 years old, the median CRT was 0.8 seconds, with an upper limit of normal of 1.9 seconds.
Adults
The range of findings in adults is subject to wider variation. Shriger and Baraff (1988) offer the following:
- Adult men: median 1.0 second (95% normal 1.6 seconds)
- Adult women: median 1.2 seconds (95% normal 2.9 seconds)
- Over 62 years of age: median 1.5 seconds (95% normal 4.5 seconds).
In addition to the above figures, it also appears that there is an average increase of 3.3% in CRT per decade of age, although men have a CRT that is typically 7% lower than women, which will need to be factored in (Anderson et al, 2008). If we accept this evidence, the upper limit of normal would need to be increased to 2.9 seconds for adult women, and up to 4.5 seconds for the older patient to capture 95% of the population.
Temperature
Ambient, skin, and core temperature affect CRT measurement in adults and children. For example, children in a warm environment (mean: 25.7°C) had CRT of less than 2 seconds but only 31% had a similar measurement in a cold environment (mean: 19.4°C) (Gorelick et al, 1993). In adults, CRT has been shown to decrease by 1.2% for every degree Celsius increase in ambient temperature (Anderson et al, 2008).
Local skin temperature also affects CRT in both adults and children. In one study of adults, each 1°C reduction in skin temperature led to a 0.21-second increase in CRT. Additionally, a statistically significant relationship has been shown between CRT and core temperature; CRT is 5% shorter, on average, for each 1°C increase in tympanic temperature (Anderson et al, 2008). The same relationships exist for newborn babies, whose CRT decreased as ambient, skin, and axillary temperature increased (Raju et al, 1999).
Ambient light
Poor light conditions can adversely affect the accuracy of CRT assessment. One study found a statistically significant difference between capillary refill assessments in light versus dark environments (Brown et al, 1994). In this study, 309 participants at an emergency services show were asked to assess each other's CRT. In daylight conditions (approximately 4000 lux), CRT was reported as normal in >94% of healthy participants compared with only <32% of the same participants in dark conditions (moonlight or streetlamp, approximately 3 lux).
Pressure application
Several studies have recommended a number of different approaches over the years. Paramedic practice seems to have adopted a 5-second duration, though with little guidance on the location or amount of pressure to be applied. The amount of pressure is difficult to gauge and this has affected a number of studies where terms such as ‘light’ or ‘moderate’ pressure have been used but with little explanation. ‘Light’ generally relates to a pressure that is just sufficient to cause blanching but ‘moderate’ is less well defined (presumably more than light but not sufficient to cause pain).
A more recent study has added some knowledge to the debate and though the study was too small to be conclusive, it may be useful when gauging time and duration of pressure. Kawaguchi et al (2019) found significant differences in CRT between pressing strengths of 1 and 3 N, but no significant differences among 3, 5, and 7 N. They suggest that pressures of between 3 and 7 N provide the optimal strength range for CRT measurement. They helpfully define a strength of 3 N as that which is required to lift an object weighing 300 g (e.g. two smartphones) using fingers. They also assessed optimal time over which to apply pressure and recommend that 2 seconds be the threshold for obtaining stable CRT measurements for all subjects.
If CRT is to be assessed, apply 3–7 N (the force required to pick up 2 smartphones) over at least 2 seconds.
Skin colour
Darker skin colour can make accurate clinical evaluation of CRT challenging. It has been argued that it is not possible to achieve a reliable CRT in darker-skinned people using a standard test as the presence of melanin prevents the assessor from being able to see the expulsion and return of blood (Matas et al, 2001; Bachour et al, 2023).
Where do I measure?
Traditionally, CRT is measured on the pulp of a digit; however, the sternum and forehead are also areas of common usage. In the assessment of tissue perfusion below the site of an injury, naturally, the site must be distal to the injury for the assessment to be of use. There is little definitive evidence to suggest that one area for measurement is superior to another. However, measurements of CRT taken at the different sites are not comparable (Crook and Taylor, 2013; de Vos-Kerkhoff et al, 2017; Meyer et al, 2024). Therefore, caution needs to be applied.
Is it useful?
In a series of studies into the predictive value of CRT as an indicator of severity of illness or hypovolaemic status, there appears to be little confidence in CRT use, with little validity found. Holcomb et al (2005) and Schriger and Baraff (1988) found that CRT had little specificity and sensitivity in recognising hypovolaemia in trauma field studies and lab-based studies, respectively. However, Mrgan et al (2014) argue that there is a significant association between CRT measured as a continuous variable and short-term mortality in adults.
In children, there is more evidence that the CRT is of use. A systematic review (Fleming et al, 2015b) concluded that CRT is an important ‘red flag’ vital sign for identifying children with serious illness. The authors identified that finding an abnormal CRT increases the likelihood of a serious outcome including death and dehydration, but that a normal CRT does not make a serious outcome less likely. Therefore, a normal CRT should not be used to rule out serious illness in children.
The use of CRT as a measure of haemodynamic status is controversial and there are many limitations. It appears to be of greater use in children than in adults. However, if it is to be used, it must be considered as a part of a holistic assessment process and not in isolation.