Relationship between Capillary Refill Time at Triage and Abnormal Clinical Condition: A Prospective Study
Claudia M. Sansone1, Fabiano Prendin2, Greta Giordano3, Paola Casati4, Anne Destrebecq5, Stefano Terzoni6, *
1 Nurse, Azienda Ospedaliero-Universitaria Maggiore della Carità di Novara, Corso Mazzini 18, 28100 Novara, Italy
2 Nurse, Emergency Department, San Paolo teaching hospital, Via A. di Rudinì, 8 – 20142 Milan, Italy
3 Nurse, ASST Santi Paolo e Carlo, presidio San Paolo, Via A. di Rudinì, 8 – 20142 Milan, Italy
4 Operating theatre, San Raffaele Hospital, Via Olgettina 60, 20132 Milan, Italy
5 Associate professor of Nursing, University of Milan, Via Ovada, 26 – 20142 Milan, Italy
6 PhD, Tutor Nurse, San Paolo bachelor school of Nursing, San Paolo teaching hospital, Via Ovada, 26 – 20142 Milan, Italy
Capillary refill time has been studied in literature as a perfusion indicator. Two pilot studies have proposed possible reference values in healthy adults. No data exist regarding capillary refill time as an indicator of abnormal clinical conditions in adults, which might be of help for triage nurses.
We wanted to assess if any relationships existed, between altered capillary refill time and abnormal clinical conditions in the emergency department. We investigated relations between capillary refill time and vital signs recorded in triage and blood tests, by analyzing the clinical records. Mortality at 24 hours, 7 days and over 14 days was investigated by calling the patients after discharge.
Observational, single-center study on a sample of consecutive patients aged ≥ 18 years in the Emergency Department of a major Milan hospital, from June to October 2014. Multivariate logistic regression was used to investigate the impact of clinical variables on capillary refill time.
1001 patients were enrolled, aged 59 ± 21 (473 aged 65 or more). Longer refill times were found in patients admitted to hospital units after medical consultations in the emergency department compared to those discharged or sent to outpatients. In elderly patients, statistically significant association was found between increased capillary refill time and sepsis (sensitivity 100%, specificity 83.33%, area under the receiver operating characteristics curve 65.95% CI 47-83), oxygen saturation, mean blood pressure, and lactates. In persons aged 45 to 64, altered refill times were associated with abnormal values of glicemia, platelets, and urea.
Capillary refill time can be used by nurses at triage as a complementary parameter to normal vital signs. This is one of the few studies investigating refill time in adult patients.
Keywords: Assessment, Capillary refill time, Clinical conditions, Nursing, Perfusion, Triage.
open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
* Address correspondence to this authors at the San Paolo bachelor school of Nursing, San Paolo teaching hospital, Via Ovada, 26 – 20142 Milan, Italy; Tel: +39 02 8184 4038; Fax: +39 02 8184 4297; E-mail: email@example.com
Capillary refill time (CRT) is defined as “the time needed by a distal body region (e.g. a fingertip) to regain the original color after having been compressed” [1Schriger DL, Baraff L. Defining normal capillary refill: variation with age, sex, and temperature. Ann Emerg Med 1988; 17(9): 932-5. [http://dx.doi.org/10.1016/S0196-0644(88)80675-9] [PMID: 3415066] , 2Pickard A, Karlen W, Ansermino JM. Capillary refill time: is it still a useful clinical sign? Anesth Analg 2011; 113(1): 120-3. [http://dx.doi.org/10.1213/ANE.0b013e31821569f9] [PMID: 21519051] ]. Some authors [3King D, Morton R, Bevan C. How to use capillary refill time. Arch Dis Child-Educ Pract 2014; 99(3): 111-6. [http://dx.doi.org/10.1136/archdischild-2013-305198] -6Schriger DL, Baraff LJ. Capillary refill-is it a useful predictor of hypovolemic states? Ann Emerg Med 1991; 20(6): 601-5. [http://dx.doi.org/10.1016/S0196-0644(05)82375-3] [PMID: 2039096] ] have proposed CRT as an indicator of perfusion in adults: in case of hypovolemic shock, peripheral vasoconstriction is a fundamental protection mechanism, aimed at saving blood for vital organs. Also, a rationale exists for investigating this parameter in patients with medical conditions related to inadequate perfusion, e.g. sepsis. The current literature provides very few information about pathologic situations capable of modifying CRT. Based on the available studies, all international algorithms of prehospital trauma care include CRT assessment: the European Resuscitation Council includes CRT as part of the assessment that should be conducted regarding circulation in the ABCDE algorithm (Airways, Breathing, Circulation, Disability, Exposure) [7Monsieurs KG, Nolan JP, Bossaert LL, et al. European Resuscitation Council guidelines for resuscitation 2015. Section 1: executive summary. Resuscitation 2015; 95: 1-80. [http://dx.doi.org/10.1016/j.resuscitation.2015.07.038] [PMID: 26477410] ]. Some indications appear in pediatric studies, as capillary refill time is mentioned in the Surviving Sepsis Campaign section pediatric patients. The guidelines recommend to obtain a capillary refill time of 2 seconds or less as an initial therapeutic endpoint of resuscitation in septic shock [8Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013; 41(2): 580-637. [http://dx.doi.org/10.1097/CCM.0b013e31827e83af] [PMID: 23353941] ]. Mrgan, et al. [9Mrgan M, Rytter D, Brabrand M. Capillary refill time is a predictor of short-term mortality for adult patients admitted to a medical department: an observational cohort study. Emerg Med J 2014; 31(12): 954-8. [http://dx.doi.org/10.1136/emermed-2013-202925] [PMID: 24045049] ] found a statistically significant association between abnormal CRT (as defined by Schrigger and Barraf) and mortality after 24 hours (OR=5.81, 95%CI=[1.36;24.79], p<.001) and 7 days after discharge from the emergency department (OR=4.24, 95%CI=[1.88;9.56], p<.001) although the positive predictive value of 24-hour mortality was as low as 15%. However, the authors provide no information regarding other abnormal conditions possibly related to alterations of CRT.
As regards assessment of CRT, Champion [10Champion HR, Sacco WJ, Carnazzo AJ, Copes W, Fouty WJ. Trauma score. Crit Care Med 1981; 9(9): 672-6. [http://dx.doi.org/10.1097/00003246-198109000-00015] [PMID: 7273818] ] suggested a maximum refill time of 2 seconds, while Schrigger and Barraf [6Schriger DL, Baraff LJ. Capillary refill-is it a useful predictor of hypovolemic states? Ann Emerg Med 1991; 20(6): 601-5. [http://dx.doi.org/10.1016/S0196-0644(05)82375-3] [PMID: 2039096] ] differentiated the cutoff values, based on age and gender, in healthy patients. The refill times they identified as normal do not exceed 1.9 seconds for males and 2.9 for females aged 18 to 64, and 1.8 seconds in both genders aged 65 or more. All the abovementioned values have been studied under controlled conditions of temperature (21°C). Considering the practical difficulties related to decimal values, the authors have also suggested to consider as physiological a refill time of 2 and 3 seconds under 65 years of age, for males and females respectively, as well as 4 seconds in both genders for elderly persons [1Schriger DL, Baraff L. Defining normal capillary refill: variation with age, sex, and temperature. Ann Emerg Med 1988; 17(9): 932-5. [http://dx.doi.org/10.1016/S0196-0644(88)80675-9] [PMID: 3415066] ].
Recently, CRT at triage has been suggested as a parameter of blood perfusion on small samples of adults, without drawing firm conclusions. Therefore, data regarding possible association between CRT and abnormal clinical condition in adults are lacking. These data could be of interest for triage nurses: should relations with specific conditions emerge, CRT might be used as a useful indicator for assessing the degree of urgency of patients. Finally, considering the unsatisfactory predictive value of mortality currently available, as well as the lack of a systematic investigation on large samples in this field, the possibility of consolidating this information with detailed analyses and longer follow-up of patients appears to be of interest.
The study presented in this paper aims to investigate possible relations between alterations of CRT in adult patients admitted to an Emergency Department (ED), and abnormal clinical conditions, as well as mortality after 24 hours, 1 week, and two weeks after discharge from the ED.
2. MATERIALS AND METHODOLOGY
A prospective, observational, single center study was conducted. We recruited a non-randomized sample of consecutive patients, aged 18 or more, who presented to the emergency department of the San Paolo teaching hospital in Milan, Italy, between June 1 and October 20, 2014. All patients with codes of emergency or urgency were included. Pediatric patients, newborns, and non-urgent cases were excluded. During the triage process in the ER, capillary refill time was assessed in all patients, in the same conditions of temperature (air conditioning room, 21°C temperature as described in the literature) and with the same clock [9Mrgan M, Rytter D, Brabrand M. Capillary refill time is a predictor of short-term mortality for adult patients admitted to a medical department: an observational cohort study. Emerg Med J 2014; 31(12): 954-8. [http://dx.doi.org/10.1136/emermed-2013-202925] [PMID: 24045049] -19Beckow J. The truth about capillary refill. JEMS 2005; 30(1): 14. [PMID: 15662338] ] to avoid any potential bias. The criteria for obtaining the refill time data included compressing the fingertip of the second finger (left or right hand was irrelevant, according to the literature [1Schriger DL, Baraff L. Defining normal capillary refill: variation with age, sex, and temperature. Ann Emerg Med 1988; 17(9): 932-5. [http://dx.doi.org/10.1016/S0196-0644(88)80675-9] [PMID: 3415066] ]), releasing, and measuring the time required to regain the initial color. A visual analogic scale (VAS) was used for measuring pain; all medical diagnoses, as well as the results of all blood tests, were recorded. Standard reference ranges were used for all tests and vital signs, according to the literature [11Longo DL, Kasper DL, Jameson JL, Fauci AS, Hauser SL, Loscalzo J. Harrison’s principles of internal medicine. 18th ed. New York: McGraw-Hill 2012., 12Whitworth JA. 2003 World Health Organization (WHO)/International Society of Hypertension (ISH) statement on management of hypertension. J Hypertens 2003; 21(11): 1983-92. [http://dx.doi.org/10.1097/00004872-200311000-00002] [PMID: 14597836] ]. Several cutoff values were taken into consideration for this study. At first, all analyses were performed according to the criteria defined by Schrigger and Barraf [1Schriger DL, Baraff L. Defining normal capillary refill: variation with age, sex, and temperature. Ann Emerg Med 1988; 17(9): 932-5. [http://dx.doi.org/10.1016/S0196-0644(88)80675-9] [PMID: 3415066] ], thus accepting 2 seconds as a cutoff for all patients. Considering that most of the available literature refers to healthy patients, while those in pathologic conditions could have longer refill times, the analysis was repeated after setting the cutoff to 3 seconds for all patients, and 4 seconds for elder persons respectively. We chose to avoid decimal cutoffs, since this would generate potential biases during the measurement: correct assessment of the refill time might actually require multiple attempts in order to obtain reliable data, thus requiring repeated pressure of the capillary bed of the fingertip, which can be misleading. A capillary refill time of 2 seconds was choses as a cutoff. All analyses were repeated considering 3 seconds and 4 seconds as alternative cutoff values. Differences between CRT values in patients with different priority codes at nursing triage were described. Mortality was assessed through telephone contact after 24 hours, 1 week, and 2 weeks from discharge. In case the patients were transferred to other hospital units, mortality was assessed through their medical records. A written form with the same questions for all patients was used in case of telephone contact.
2.1. Statistical Analysis
Wilcoxon’s signed-rank test was used for comparing median values (e.g. median capillary refill times). The values of all blood tests and vital parameters were dychotomized (altered/non-altered) according to the abovementioned criteria, and chi-square tests for association were performed, in order to assess possible relations with capillary refill time. Since patients are complex systems, and many variables are related to each other, a multivariate stepwise logistic regression model was used, to assess the role of all clinical indicators in the overall presentation of our patients. In such model, alterations of CRT were studied in relationship with vital signs and blood tests. Finally, a sensitivity and specificity analysis was conducted, to determine whether CRT values could help distinguishing generic infections from sepsis. In this analysis, a ROC curve (Receiver Operating Characteristics) was used. 95% confidence intervals were calculated in all inferential analyses; the results were considered statistically significant in presence of p-values below .05. For each analysis, the actual number of patients was taken into account (see Table 2) as not all patients underwent the same tests. All data were collected through the FirstAid software, officially in use in the Emergency Department, and analyzed with STATA® 11 for Windows. We complied with the rules of the local Ethical committee; the hospital management approved the study, which was conducted in respect of the principles stated in the Declaration of Helsinki [20Declaration of Helsinki: ethical principles for medical research involving human subjects 2015.]. Patients’ data were treated anonymously, as all subjects were identified by the unique code provided by the hospital software. All data were treated according to the national Italian law on privacy protection.
1001 consecutive patients were enrolled (482 males, 48.15%, and 519 females, 51.85%), aged 59 ± 21 (473 aged 65 or more). Table 1 summarizes the categories of medical diagnoses of the patients.
Table 1 Medical diagnoses.
As a first finding, significantly higher refill times were found in patients admitted to hospital units after medical consultations in the ER (n=566, Me=4, IQR= [3King D, Morton R, Bevan C. How to use capillary refill time. Arch Dis Child-Educ Pract 2014; 99(3): 111-6. [http://dx.doi.org/10.1136/archdischild-2013-305198] , 5Lewin J, Maconochie I. Capillary refill time in adults. Emerg Med J 2008; 25(6): 325-6. [http://dx.doi.org/10.1136/emj.2007.055244] [PMID: 18499809] ]. 19 refused to be admitted, 1 was transferred to another hospital) if compared to those discharged (n=221, Me=3, IQR= [2Pickard A, Karlen W, Ansermino JM. Capillary refill time: is it still a useful clinical sign? Anesth Analg 2011; 113(1): 120-3. [http://dx.doi.org/10.1213/ANE.0b013e31821569f9] [PMID: 21519051] , 4Champion HR, Sacco WJ, Carnazzo AJ, Copes W, Fouty WJ. Trauma score. Crit Care Med 1981; 9(9): 672-6. [http://dx.doi.org/10.1097/00003246-198109000-00015] [PMID: 7273818] ]) or sent to outpatients for further medical activities (n=194, Me=3, IQR= [2Pickard A, Karlen W, Ansermino JM. Capillary refill time: is it still a useful clinical sign? Anesth Analg 2011; 113(1): 120-3. [http://dx.doi.org/10.1213/ANE.0b013e31821569f9] [PMID: 21519051] , 4Champion HR, Sacco WJ, Carnazzo AJ, Copes W, Fouty WJ. Trauma score. Crit Care Med 1981; 9(9): 672-6. [http://dx.doi.org/10.1097/00003246-198109000-00015] [PMID: 7273818] ]), p<.001 for all comparisons. 413 patients (41.25%) had capillary refill times above 2 seconds. Of these, 12 had the highest priority code, 214 had intermediate priority codes and 187 had low priority. Of note, the number of patients aged 65 or more increased with the severity of clinical conditions, if compared to the number of younger persons (65 vs 122 among low-priority codes, 133 vs 81 among intermediate priority codes, and 10 vs 2 among high priority patients). Statistically significant associations were found between single variables and altered refill times. Considering that some parameters can be altered mostly in the elderly, or at least in the middle-aged, we chose to divide patients into three separate ranges, based on their age (18-44, 45-64, and 65 or more), as defined in the literature [11Longo DL, Kasper DL, Jameson JL, Fauci AS, Hauser SL, Loscalzo J. Harrison’s principles of internal medicine. 18th ed. New York: McGraw-Hill 2012.].
Table 2 shows the results of the analysis and reports the number of patients for each blood parameter, since not all patients underwent the same exams. Values are presented as odds ratios and 95% confidence intervals. Statistically significant results are marked in bold. Only significant p-values are reported: in all other cases, the analysis reported p-values above 0.05.
Table 2 Associations between altered refill times and blood exams.
Statistically significant associations were found between altered capillary refill time and diastolic hypertension in younger patients (aged 18-44). In middle-aged persons (45-64), CRT was associated with hyperglycemia, thrombocytopenia, and high levels of urea and alanine transaminase.
A clinically relevant finding emerged in elderly patients, in which CRT alterations showed statistically significant and strong associations with blood oxygen saturation <90%, diastolic hypotension, mean blood pressure, and lactates. According to the literature, such parameters can be found in patients with sepsis. In the multivariate logistic regression model, the associations were confirmed, as shown in Table 3.
Table 3 Results of the multivariate logistic model.
Actually, only seven patients were diagnosed with sepsis in our sample, all aged 65 or more. All seven had capillary refill time alterations; based on this finding, we decided to further investigate this aspect, and we conducted a sensitivity analysis to determine whether CRT could be able to distinguish between generic infections and sepsis. Sensitivity was 100%, as CRT was elevated in all septic patients; specificity was 83.33%, with an area under the ROC (Receiver Operating Characteristics) curve of .65, 95%CI=[.47-.83]. Although not fully satisfactory, these latter findings account for further data collection and investigation in future research, especially considering the interesting data regarding specificity.
We were able to retrieve the data of all 1001 patients, through telephone contact or medical records. 25 patients enrolled for this study died, out of 1001 (2.49%). They had been admitted to the emergency department for trauma (n=7), sepsis (n=3), hemorrhage (n=3), central nervous system disorders (n=3), and respiratory (n=2), gastroenteric (n=2), nephrological (n=1) or cardiological (n=1) conditions. The remaining 3 had generic diagnoses, which required further medical assessment (e.g. “acute pain”).
Out of those 25, 4 died in 24 hours, 2 in a week, and 19 after two weeks or more. We sought to find associations between capillary refill time alterations and mortality; Table 4 summarizes the results.
Table 4 Associations between capillary refill time and mortality.
No statistically significant relationship was found between altered CRT values and mortality, regardless of age.
The findings of the present study show an association between prolonged CRT in elder adults and several clinical indicators of septic conditions. Such relationships appear to be strong, as confirmed by odds ratios, and stay evident in both bivariate and multivariate analysis. Sensitivity analysis yielded clear distinction between patients diagnosed with sepsis and persons with generic infections (e.g. urinary tract infections). These findings appear to be the first in literature, and deserve further investigation. In persons aged 45 to 64, prolonged CRT was associated with hyperglycemia, as well as alterations in the amount of alanine transaminase and urea; these two latter findings might be of interest and need to be investigated in-depth. As to our knowledge, no papers in the existing literature have investigated this relationship in adults so far. In particular, hyperglycemia has only been investigated in infants [21Jucá CA, Rey LC, Martins CV. Comparison between normal saline and a polyelectrolyte solution for fluid resuscitation in severely dehydrated infants with acute diarrhoea. Ann Trop Paediatr 2005; 25(4): 253-60. [http://dx.doi.org/10.1179/146532805X72395] [PMID: 16297299] ]. In patients aged 18 to 44, an association was found between CRT and diastolic hypertension. Overall, from the point of view of pathophysiology, the results of this investigation appear to be congruent with the characteristics and age ranges of diseases onset. The small number of patients diagnosed with sepsis prevents from drawing firm conclusions; however, the association of indicators of sepsis with prolonged CRT strongly suggest that refill time can actually be studied as an indicator of sepsis in the general elder population. Some authors, such as Brunauer et al. [22Brunauer A, Koköfer A, Bataar O, et al. Changes in peripheral perfusion relate to visceral organ perfusion in early septic shock: A pilot study. J Crit Care 2016; 35: 105-9. [http://dx.doi.org/10.1016/j.jcrc.2016.05.007] [PMID: 27481743] ] have recently suggested that CRT is associated with pulsatility index of the liver and intestines; pulsatility index is a sonographic surrogate of vascular tone, meant to be assessed in visceral organs in patients with early septic shock. Our findings support such results and point out the importance of assessing CRT in patients with suspected sepsis, as suggested by Postelnicu et al. [23Postelnicu R, Evans L. Monitoring of the physical exam in sepsis. Curr Opin Crit Care 2017; 23(3): 232-6. Epub ahead of print [http://dx.doi.org/10.1097/MCC.0000000000000403] [PMID: 28306558] ].
As regards mortality, in this sample of patients no significant associations with CRT were found. Other authors have studied CRT in different populations, finding a relationship between refill time and 14-day mortality after resuscitation in patients with septic shock [24Ait-Oufella H, Bige N, Boelle PY, et al. Capillary refill time exploration during septic shock. Intensive Care Med 2014; 40(7): 958-64. [http://dx.doi.org/10.1007/s00134-014-3326-4] [PMID: 24811942] ]. However, no studies appear to be available in literature regarding patients with our inclusion criteria.
This was a single-center, non-randomized study, conducted in a general hospital. The number of patients with sepsis only allows preliminary conclusions on the real capacity of CRT to distinguish between generic infections and sepsis. Further investigations are needed to strengthen this part of the study. In our sample, many different types of problems were diagnosed; therefore, not all patients underwent the same blood tests. This limitation is partially amended by the overall large number of persons we studied, which allows drawing preliminary conclusions from our data.
Capillary refill time has shown association with several conditions, which change over the lifespan. Older patients have a broader range of abnormal condition, which present parallel to CRT alterations.
The potential utility of CRT as a clinical indicator is also supported by the fact that the highest CRT values were found in the patients admitted to hospital units after consultation in the ED. This suggests that CRT alteration become more evident in case of severe and urgent problems.
This is one of the few studied to have investigated the potential role of CRT at triage as a clinical indicator in adults at triage. Other authors have studied the same topic in pediatric patients, but data regarding adult patients were lacking. The present investigation has identified several categories of health problems; in the future, it would be useful to understand if CRT can improve sensitivity and specificity of the current diagnostic methods for such conditions. If so, CRT would become a very interesting indicator for both triage nurses and physicians in the emergency department. Further investigation on large samples of patients with specific diseases is needed to compare the accuracy of diagnoses made with or without taking capillary refill time into account.
ETHICS APPROVAL AND CONSENT TO PARTICIPATE
Rules of the local Ethical Committee and the Declaration of Helsinki were considered during this study.
HUMAN AND ANIMAL RIGHTS
No Animals/Humans were used for studies that are base of this research.
CONSENT FOR PUBLICATION
CONFLICT OF INTEREST
The authors (editor) declare no conflict of interest, financial or otherwise.
Mrgan M, Rytter D, Brabrand M. Capillary refill time is a predictor of short-term mortality for adult patients admitted to a medical department: an observational cohort study. Emerg Med J 2014; 31(12): 954-8. [http://dx.doi.org/10.1136/emermed-2013-202925] [PMID: 24045049]
Karlen A, Daniels J, Kwizera A, Ibingira C, Dumont G, Ansermino JM. Validation of Capillary Refill Time Measurements Using Photo-Plethysmogram from a Portable Device for Effective Triage in Children. Global Humanitarian Technology Conference (GHTC). [http://dx.doi.org/10.1109/GHTC.2011.19]
Beckow J. The truth about capillary refill. JEMS 2005; 30(1): 14. [PMID: 15662338]
Declaration of Helsinki: ethical principles for medical research involving human subjects 2015.
Jucá CA, Rey LC, Martins CV. Comparison between normal saline and a polyelectrolyte solution for fluid resuscitation in severely dehydrated infants with acute diarrhoea. Ann Trop Paediatr 2005; 25(4): 253-60. [http://dx.doi.org/10.1179/146532805X72395] [PMID: 16297299]