Hrev_master [page 70] [Emergency Care Journal 2024; 20:12569] Emergency Care Journal 2024 volume 20:12569 Abstract Cool extremities representing impaired skin perfusion are a classical sign of shock. We evaluated associations between the presence and persistence of subjective cool extremities observed by caring nurses and clinical outcomes. We conducted a retrospec- tive observational study in an Intensive Care Unit (ICU) at a uni- versity hospital in Japan. Patients were divided into two groups based on the presence or absence of subjective cool extremities during the first 24 hours of their ICU stay. We compared their char- acteristics and outcomes. In total, 2956 patients were analyzed. Patients with cool extremities were older, had higher severity Acute Physiology and Chronic Health Evaluation (APACHE III) scores, had higher in-hospital mortality (4.1% vs 18%, p<0.001), and had a higher prevalence of acute kidney injury requiring renal replacement therapy (2.1% vs 10%, p<0.001) as compared to patients without nurse-reported cool extremities. Multivariable logistic regression showed cool extremities persisting for ≥12 hours were significantly associated with in-hospital death (adjust- ed Odds Ratio, OR, 1.64) and acute kidney injury requiring renal replacement therapy (adjusted OR 1.82). Patients with subjective cool extremities were more severely ill and had poorer outcomes. Subjective skin temperature assessment may be useful to detect high-risk patients. Introduction Hemodynamic instability is common among critically ill patients and can lead to organ failure and poor outcomes. Hemodynamic monitoring is crucial to detect hemodynamic insta- bility and start interventions to prevent organ damage. Although global hemodynamic measurements, such as blood pressure, heart rate, and cardiac output, are commonly used to assess hemody- namic status, these alone are insufficient for a comprehensive eval- uation of a patient’s circulatory status.1 Recently, impaired micro- circulation has been found to play a critical role in the development of organ dysfunction and adverse outcomes among critically ill patients.2 Therefore, early detection of microcirculation impair- ment is essential in the management of these patients. Signs of impaired skin perfusion, such as mottled skin, pro- longed capillary refill time, and increased center-to-toe tempera- ture gradient, are considered to indicate impaired microcirculation.3 Several previous studies reported associations between signs of impaired skin perfusion and adverse outcomes.4- 7 Subjective skin temperature assessment is the simplest method of evaluating skin perfusion. Several studies reported associations between subjective cool extremities and higher lactate levels4 and adverse outcomes.6,8,9 However, most of these studies used small sample sizes and did not adjust for disease severity using multi- variate analysis. Therefore, it remains unclear whether subjective cool extremities are indicative of severely ill patients or predictors of poor outcomes. This study aimed to evaluate the associations between subjec- Correspondence: Masatoshi Uchida, Department of Emergency and Critical Care Medicine, Dokkyo Medical University, 880 Kitakobayashi, Mibu-machi, Shimotuga-gun, Tochigi 321-0293, Japan. Tel.: +81-282-86-1111. E-mail: m-uchida@dokkyomed.ac.jp Key words: physical examination, skin perfusion, shock, critically ill. Contributions: all authors contributed to the study conception and design. Material preparation, data collection, and analysis were per- formed by MU. The first draft of the manuscript was written by MU, and all authors commented on previous versions. All authors read and approved the final manuscript, and agreed to be held account- able for all aspects of the work. Conflict of interest: the authors declare no potential conflict of interest. Funding: none. Ethics approval: the Institutional Review Board of Dokkyo Medical University Hospital waived the requirement for informed consent because of the observational and retrospective nature of this study (approval number R-43-12J). Availability of data and materials: the corresponding author's derived data supporting this study's findings are available upon rea- sonable request after approval from the Institutional Review Board of Dokkyo Medical University Hospital. Acknowledgments: we thank Audrey Holmes, MA, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript. Received: 15 April 2024. Accepted: 17 June 2024. Early view: 23 July 2024. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright: the Author(s), 2024 Licensee PAGEPress, Italy Emergency Care Journal 2024; 20:12569 doi:10.4081/ecj.2024.12569 Publisher's note: all claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher. Persistent cool extremities as an independent predictor of adverse clinical outcomes among critically ill patients: a single-center retrospective observational study Masatoshi Uchida,1 Migaku Kikuchi,1 Eisei Hoshiyama,2 Koji Wake1 1Department of Emergency and Critical Care Medicine, Dokkyo Medical University, Tochigi; 2Department of Neurology, Dokkyo Medical University, Tochigi, Japan Non -co mmerc ial us e o nly tive cool extremities and patient characteristics, disease severity, and clinical outcomes among non-selected Intensive Care Unit (ICU) patients. We hypothesized that among these patients, i) sub- jective cool extremities could be associated with disease severity, and ii) subjective cool extremities and their persistence could be associated with adverse clinical outcomes. Materials and Methods Study design This retrospective observational study was conducted in a 12- bed general ICU at a university hospital in Japan from April 2016 to December 2019. The Institutional Review Board waived the requirement for informed consent because of the observational and retrospective nature of this study (approval number R-43-12J). Participants All adult patients (aged ≥16 years) were included. For patients readmitted to the ICU during the same hospitalization, only the initial admission was included. We excluded patients without assessment of cool extremities during the first 24 hours from ICU admission. Data collection The information collected for all patients admitted to the ICU from their electrical medical records included: age, gender, height, weight, admission classification (elective surgery, emergency sur- gery, non-operative), admission type (elective or emergency), admission source (general ward, emergency room, operation room, others), hospitalization date, ICU admission date, chronic organ insufficiency (congestive heart failure, respiratory failure, liver cir- rhosis, use of immunosuppressants, chronic hemodialysis, lym- phoma, acute leukemia, cancer with metastases), diagnosis at ICU admission, and vital signs and laboratory data within 24 hours after ICU admission. We also recorded patients’ severity scores, includ- ing the Acute Physiology and Chronic Health Evaluation (APACHE) II, APACHE III, and Simplified Acute Physiology Score (SAPS) II, along with the treatment received during their ICU stay (mechanical ventilation use, vasopressor use, and Renal Replacement Therapy, [RRT], use) and prognosis, including in- ICU death and in-hospital death. Exposure The primary exposure in this study was subjective cool extrem- ities within 24 hours after ICU admission and the duration of cool extremities. The presence of cool extremities was based on the ICU chart recorded by the nurse providing care. In the studied ICU, the caring nurse routinely assesses and documents the presence of sub- jective cool extremities. The timing of this assessment is not prede- fined and is based on the caring nurse’s discretion. For this study, the duration of cool extremities was defined as follows. Among patients who exhibited at least one episode of cool extremities within 24 hours after ICU admission, the duration of the cool extremities episode was calculated as the time between the onset and resolution of cool extremities. If cool extremities appeared and disappeared more than once within 24 hours after ICU admission, the duration was calculated as the sum of these episodes. In cases where the res- olution of cool extremities was not documented within 24 hours after ICU admission or by the time of ICU discharge, the cool extremities were considered to have persisted until 24 hours after ICU admission or ICU discharge (Figure 1). For patients that did not have cool extremities, the duration was defined as zero. The duration of cool extremities was categorized into two groups: “transient” if the duration was less than 12 hours and “persistent” if the duration was ≥12 hours. This threshold was based on a previous study.10 Article Figure 1. Examples of the duration of cool extremities. Black circles represent cool extremities and white circles represent no cool extrem- ities at that time point. Patient A: Cool extremities appeared and disappeared within first 24 hours. The duration was defined as interval between this sign appearing and disappearing (a). Patient B: Cool extremities appeared and disappeared multiple times within 24 hours. The duration was defined as the sum of intervals between each episode (b + c). Patient C: Cool extremities appeared but did not disappear within first 24 hours. The duration was defined as interval of time point when this sign appeared and 24 hours after ICU admission (d). Patient D: Cool extremities appeared multiple times within 24 hours but disappeared and were not observed until discharge for a second episode. The duration of the second episode was defined as the interval between when the sign appeared and discharge (f), and the total duration was defined as sum of these periods (e + f). [Emergency Care Journal 2024; 20:12569] [page 71] Non -co mmerc ial us e o nly Outcomes The primary outcome was in-hospital mortality. The secondary outcome was Acute Kidney Injury (AKI) that required RRT, which was defined as RRT use during the ICU stay for patients not on chronic dialysis. Covariates Covariates for adjustment were selected based on previously reported outcome predictors among critically ill patients or clinical importance, and included: age, gender, comorbidities, days before ICU admission after hospitalization,11 cardiopulmonary resuscita- tion before ICU admission, reason for ICU admission,12,13 diagno- sis for ICU admission (cardiovascular disease vs sepsis vs others), APACHE III score, hyperlactatemia (more than 2 mmol/L) within 24 hours after ICU admission, mechanical ventilation within 24 hours after ICU admission, and use of vasoactive agents within 24 hours after ICU admission. Statistical analyses Patients were first divided into those who had cool extremities within 24 hours after ICU admission and those who did not. Patient characteristics, severity of illness, treatment during ICU stay, and outcomes were compared between the two groups. Categorical variables were reported as counts and percentages, and compar- isons between the two groups were made using Fisher’s exact tests and chi-square tests as appropriate. Continuous variables were reported as median and Interquartile Range (IQR) and compared using a Mann–Whitney U-test. Locally Estimated Scatterplot Smoothing (LOESS) curves with 95% Confidence Intervals (CI) were used to illustrate the relationships between the duration of cool extremities and in-hospital mortality and prevalence of AKI requiring RRT. Multivariable logistic regression analyses were used to analyze associations between cool extremities and out- comes, adjusted for potential confounding factors, and the OR and 95% CI were calculated. In the multivariable logistic regression models, we treated cool extremities as a categorical value with three levels: i) no cool extremities, ii) transient cool extremities (duration <12 hours), and iii) persistent cool extremities (duration ≥12 hours). A sensitivity analysis was also conducted in which the duration of cool extremities was entered into the model as a con- tinuous variable. All statistical tests were two-tailed, with p-values <0.05 considered statistically significant. All analyses were per- formed using R version 4.1.3 (R Foundation for Statistical Computing, Vienna, Austria). Results Patients’ characteristics and outcomes by the pres- ence of subjective cool extremities During the study period, 3278 cases were admitted to the ICU and 2956 patients were analyzed (Figure 2). Twenty-two patients were excluded because of no assessment of cool extremities within 24 hours after ICU admission. Of these patients, 14 patients had died within 6 hours after ICU admission. Among the analyzed patients, there was a median of six assess- ments of cool extremities within 24 hours after ICU admission. Patients’ characteristics, severity of illness, treatment during their ICU stay, and clinical outcomes are reported in Table 1. The medi- an age was 70 years (IQR: 61-76 years), 35% of patients were female, the median APACHE III score was 58 (IQR: 44-74), 52% required mechanical ventilation, 43% received vasoactive agents, and 9.2% required RRT. The rate of in-hospital mortality was 8.8%, and that of AKI requiring RRT was 4.8%. Article Figure 2. Patient flow diagram. [page 72] [Emergency Care Journal 2024; 20:12569] Non -co mmerc ial us e o nly In total, 1018 patients had cool extremities within 24 hours after ICU admission. Patients with cool extremities were older (69 [61-75] years vs 71 [63-77] years, p<0.001) and more likely to be on chronic dialysis (3.9% vs 11%, p<0.001) than those without cool extremities. Cardiovascular disease (33% vs 57%, p<0.001) and infection (7.0% vs. 18%, p<0.001) were more common diag- noses at ICU admission among patients with cool extremities than among those without cool extremities. These patients also had higher severity scores on admission (APACHE III score: 52 [41- 65] vs 71 [55-89], p<0.001; SAPS II score: 28 [21-36] vs 41 [31- 55], p<0.001), more frequently required mechanical ventilation (41% vs 72%, p<0.001), vasopressors (32% vs 65%, p<0.001), and RRT (4.6% vs 18%, p<0.002), and had higher in-hospital mortality (4.1% vs 18%, p<0.001) and prevalence of AKI requiring RRT (2.1% vs 10%, p<0.001). Clinical outcomes by the duration of subjective cool extremities The LOESS curve showed positive relationships between the duration of cool extremities and the proportions of AKI requiring RRT and in-hospital mortality (Figure 3). Multivariable logistic regression analysis revealed that persistent cool extremities were significantly associated with in-hospital death (adjusted OR 1.64, 95% CI 1.08-2.47, p=0.018) and AKI requiring RRT (adjusted OR 1.82, 95% CI 1.11-3.01, p=0.018) (Table 2). The sensitivity analy- sis showed that the duration of cool extremities was significantly associated with in-hospital mortality and AKI requiring RRT, with the odds of in-hospital death increasing by 2% and that for AKI requiring RRT by 4% for each hour increase in duration (Table 3). Article [Emergency Care Journal 2024; 20:12569] [page 73] Table 1. Patients’ characteristics and outcomes (N=2956). Variables Overall, N=29561 Cool extremities, n=10181 No cool extremities, n=19381 p-value2 Age (years), median (IQR) 70 (61, 76) 71 (63, 77) 69 (61, 75) <0.001 Gender, n (%) 0.4 Female 1048 (35) 351 (34) 697 (36) Male 1908 (65) 667 (66) 1241 (64) Height (cm), median (IQR) 161 (153, 168) 160 (153, 168) 162 (154, 168) 0.4 Weight (kg), median (IQR) 58 (50, 67) 56 (49, 65) 59 (51, 67) <0.001 Comorbidities, n (%) Chronic heart failure 8 (0.3) 2 (0.2) 6 (0.3) 0.7 Chronic respiratory failure 24 (0.8) 11 (1.1) 13 (0.7) 0.2 Liver cirrhosis 48 (1.6) 10 (1.0) 38 (2.0) 0.045 Use of immunosuppressants 266 (9.0) 88 (8.6) 178 (9.2) 0.6 On chronic hemodialysis 189 (6.4) 114 (11) 75 (3.9) <0.001 Lymphoma 30 (1.0) 12 (1.2) 18 (0.9) 0.5 Acute leukemia 7 (0.2) 2 (0.2) 5 (0.3) >0.9 Cancer with metastases 174 (5.9) 39 (3.8) 135 (7.0) <0.001 Days before ICU admission after 4 (2, 7) 4 (0, 8) 4 (2, 7) <0.001 hospitalization, median (IQR) Cardiopulmonary resuscitation before 89 (3.0) 78 (7.7) 11 (0.6) <0.001 ICU admission, n (%) Reason for ICU admission, n (%) <0.001 Transfer from ward 368 (12) 232 (23) 136 (7.0) Transfer from emergency room 285 (9.6) 150 (15) 135 (7.0) Elective surgery 1867 (63) 430 (42) 1437 (74) Urgent surgery 330 (11) 144 (14) 186 (9.6) Other 106 (3.6) 62 (6.1) 44 (2.3) Diagnosis at ICU admission, n (%) Cardiovascular diseases 1220 (41) 577 (57) 643 (33) <0.001 Infectious diseases 316 (11) 180 (18) 136 (7.0) <0.001 Others 1472 (50) 293 (29) 1179 (61) <0.001 APACHE III score, median (IQR) 58 (44, 74) 71 (55, 89) 52 (41, 65) <0.001 SAPS II score, median (IQR) 32 (23, 43) 41 (31, 55) 28 (21, 36) <0.001 Hyperlactatemia3, n (%) (N=2920) 1638 (56) 641 (64) 997 (52) <0.001 Mechanical ventilator use, n (%) 1531 (52) 729 (72) 802 (41) <0.001 Vasoactive agents use, n (%) 1278 (43) 663 (65) 615 (32) <0.001 RRT use, n (%) 271 (9.2) 181 (18) 90 (4.6) <0.001 AKI requiring RRT4, n (%) (N=2767) 134 (4.8) 94 (10) 40 (2.1) <0.001 In-hospital death, n (%) (N=2955) 261 (8.8) 181 (18) 80 (4.1) <0.001 ICU, Intensive Care Unit; IQR, Interquartile Range; AKI, Acute Kidney Injury; RRT, Renal Replacement Therapy. 1Data presented as median (IQR) or n (%). 2Wilcoxon rank sum test; Pearson’s chi-squared test; Fisher’s exact test. 3Lactate more than 2 mmol/L within 24 hours after ICU admission. 4RRT use during the ICU stay for patients not on chronic dialysis. Non -co mmerc ial us e o nly Discussion Key findings This retrospective observational study evaluated the clinical significance of subjective skin temperature assessment. Subjective cool extremities were recorded for 34.4% of critically ill patients admitted in the studied ICU. Patients who had cool extremities were older, had higher severity scores, required more invasive treatment, and more frequently had adverse clinical outcomes than those without cool extremities. The proportion of adverse clinical outcomes increased with the duration of cool extremities. Persistent cool extremities with 12 or more hours duration was associated with in-hospital mortality and AKI requiring RRT, inde- pendent of potential confounders. Relationship to previous studies Peripheral hypoperfusion is the clinical manifestation of circu- latory failure, and several skin hypoperfusion signs have been examined in relation to laboratory values and clinical outcomes. Skin mottling refers to a purpuric discoloration of the skin caused by hypoperfusion of the skin.14 It was previously reported that skin mottling was associated with lower skin microcirculatory satura- tion,15 higher lactate level,5 and mortality.7,16 In terms of capillary refill time, the time required to recolor the tip of a finger was reported to be associated with postoperative complications,17 high- er lactate,18 and mortality.18 Both the presence and persistence of these skin hypoperfusion signs have been associated with mortali- ty.16 These skin hypoperfusion signs can be observed noninvasive- ly, but this has the disadvantages of requiring training for evalua- tion and having capillary refill time cutoff values that vary from study to study.17,18 Although subjective assessment of cool extremities is highly examiner-dependent and can be found in conditions that differ from peripheral hypoperfusion, such as hypothermia and peripher- al artery diseases, several previous studies reported similar results to our study. Kaplan et al. investigated associations between sub- jective extremity skin temperature and hypoperfusion among sur- gical ICU patients.4 That study found patients who had cool extremities had higher serum lactate and lower cardiac output than those without cool extremities. Lima et al. investigated the rela- tionships between subjective abnormal peripheral perfusion (increase in capillary refill time or subjective cool extremities) and hyperlactatemia and Sequential Organ Failure Assessment (SOFA) score improvement among initial resuscitated ICU patients.6 They reported that patients with subjective abnormal peripheral perfu- sion were more likely to have hyperlactatemia and had lower SOFA score improvements than those without abnormal peripheral perfusion. Wiersema et al. reported that subjective cool extremities within 24 hours after ICU admission was associated with AKI dur- ing the ICU stay among patients who were acutely admitted to the ICU.8 Lin et al. reported that post-resuscitated patients with severe sepsis and septic shock who had cool extremities had higher APACHE II scores and higher mortality than other patients.9 These findings suggested that despite assessment being subjective, cool extremities had clinical significance for the assessment of periph- eral hypoperfusion and patient illness severity. The duration and adverse clinical outcomes have been reported for other peripheral hypoperfusion signs.1,19,20 Therefore, we inves- tigated the relationship between the duration of cool extremities and outcomes and found a dose-dependent association between the duration of cool extremities and adverse clinical outcomes. Furthermore, cool extremities that persisted for 12 hours or more were associated with adverse clinical outcomes, even after adjust- ing for potential confounders. There are various possible reasons for this finding. Previous studies showed that persistent peripheral hypoperfusion was associated with poor outcomes despite normal- ization of global hemodynamic measurements (e.g., blood pres- sure, cardiac output) as a result of initial resuscitation.1 In the mod- ern ICU context, it is assumed that many patients with shock could have been initially resuscitated within 6 hours because of the influ- ence of early goal-directed therapy21 and surviving sepsis cam- Article Table 2. Multivariable logistic regression analysis for the associations between cool extremities and clinical outcomes. Variables In-hospital death AKI requiring RRT N Adjusted OR 95% CI p-value N Adjusted OR 95% CI p-value Cool extremities 2,919 2,735 No cool extremities — — — — Transient cool extremities1 1 0.65, 1.53 >0.9 0.9 0.51, 1.56 0.7 Persistent cool extremities2 1.64 1.08, 2.47 0.018 1.82 1.11, 3.01 0.018 AKI, Acute Kidney Injury; RRT, Renal Replacement Therapy; OR, Odds Ratio; CI, Confidence Interval; ICU, Intensive Care Unit. 1Cool extremities duration less than 12 hours. 2Cool extremities duration 12 hours or more. Multivariable logistic regression model adjusted for age, gender, chronic heart failure, chronic respiratory failure, liver cirrhosis, use of immunosuppressants, on chronic hemodialysis, lymphoma, acute leukemia, can- cer with metastases, days before ICU admission after hospitalization, cardiopulmonary resuscitation before ICU admission, reason for ICU admission, diagnosis for ICU admission (cardiovascular disease vs. sepsis vs. others), APACHE III score, hyperlactatemia (more than 2 mmol/L) within 24 hours after ICU admission, mechanical ventilation within 24 hours after ICU admission, and vasoactive agents use within 24 hours after ICU admission. Table 3. Multivariable logistic regression analysis for the associations between cool extremities duration as a continuous variable and clin- ical outcomes.Variab Variables In-hospital death AKI requiring RRT N Adjusted OR 95% CI p-value N Adjusted OR 95% CI p-value Cool extremities duration (hours) 2,919 1.02 1.00, 1.04 0.026 2,735 1.04 1.02, 1.07 0.001 AKI, Acute Kidney Injury; RRT, Renal Replacement Therapy; OR, Odds Ratio; CI, Confidence Interval; ICU, Intensive Care Unit. Multivariable logistic regression model adjusted for age, gender, chronic heart failure, chronic respiratory failure, liver cirrhosis, use of immunosuppressants, on chronic hemodialysis, lymphoma, acute leukemia, cancer with metastases, days before ICU admission after hospitalization, cardiopulmonary resuscitation before ICU admission, the reason for ICU admission, diagnosis for ICU admission (cardiovascular disease vs. sepsis vs. others), APACHE III score, hyperlactatemia (more than 2 mmol/L) within 24 hours after ICU admission, mechanical ventilation within 24 hours after ICU admission, and vasoactive agents use within 24 hours after ICU admission. [page 74] [Emergency Care Journal 2024; 20:12569] Non -co mmerc ial us e o nly paign guidelines.22 Therefore, many patients who had cool extrem- ities for more than 12 hours may have had persistent peripheral hypoperfusion after initial resuscitation. In contrast, cool extremities that persisted for less than 12 hours were not associated with poor outcomes. These patients may have had circulatory failure but were resuscitated adequately or had another reason for cool extremities that was not peripheral hypoperfusion, such as transient hypothermia. Cool extremities due to peripheral hypoperfusion may be distinguished from other “low-risk” cool extremities by focusing on the duration. Moreover, the relationships between cool extremities and adverse outcomes did not change when we adjusted for hyperlactatemia, which is a well-known sign of peripheral hypoperfusion. This finding may be attributed to the fact that not all hyperlactatemia is due to tissue hypoperfusion.22 This result suggested that regardless of lactate measurement, assessing skin perfusion is important for under- standing a patient’s circulatory status. Significance and implications In this study, we revealed that subjective cool extremities were associated with patient illness severity, and persistent cool extrem- ities were an independent predictor of adverse clinical outcomes. These results suggested that physical examination findings play an important role in assessing patient illness severity and determining the need for additional treatment. However, physical examination findings are subjective and cannot be used to make decisions on their own. An individual’s physical examination findings may lead to more objective examinations, such as a multi-person evaluation, bedside ultrasonography, and blood tests, which may provide a more accurate picture of the patient’s condition. In recent years, point-of-care testing and the development of new testing equip- ment have made it possible to obtain objective data immediately. Therefore, subjective information (e.g., physical examination) tends to be overlooked. A previous survey found that half of the participating physicians reported physical examinations had limit- ed value for ICU patients, and more than half of attending physi- cians and fellows reported that they only saw patients occasionally.24 We believe physical examinations provide clues to determine whether a patient should receive more invasive and expensive procedures. Strengths and limitations To our knowledge, this is the largest study that evaluated sub- jective skin temperature assessment and clinical outcomes among ICU patients. However, our study had several limitations. First, the presence or absence of cool extremities in this study was based on subjective assessment, and the objective accuracy was not evaluat- ed. Although the examiner might have rated more severely ill patients as having cool extremities, persistent cool extremities were associated with adverse outcomes after adjusting for severity score, serum lactate level, mechanical ventilation use, and vasoac- tive agents use. Therefore, the results considered any bias resulting from vital signs, laboratory data, and treatment received. Moreover, subjective assessment of skin temperature has been reported to be correlated with objective peripheral hypoperfusion signs.6 Although the lowest body temperature in the cool extremi- ties group was significantly lower than that in the group without cool extremities, the effect of body temperature on in-hospital mortality was adjusted for in the APACHE III score. Therefore, this did not affect our finding that subjective cool extremities were an independent predictor of poor prognosis independent of severity score. The fact that even a subjective, untrained assessment may be an independent prognostic factor is of great clinical value in that routine clinical records may be useful to predict patients’ out- comes. Second, we only focused on the first 24 hours after ICU admission. Therefore, cool extremities that appeared after 24 hours Article [Emergency Care Journal 2024; 20:12569] [page 75] Figure 3. Association between duration of cool extremities and in-hospital mortality and prevalence of Acute Kidney Injury (AKI) requir- ing Renal Replacement Therapy (RRT). The gray bands represent the 95% confidence interval. Non -co mmerc ial us e o nly were not evaluated. As most critically ill patients are most severely ill immediately after the start of ICU treatment, we believed that assessing the significance of physical findings at that time had great clinical value. Third, the duration of cool extremities was not precise because the assessment of cool extremities was intermit- tent. To address this issue, the duration was divided into two groups for our multivariable analysis. Most of the analyzed patients had at least five physical examinations within 24 hours of admission to the ICU. Therefore, this classification had acceptable accuracy in assessing the duration of cool extremities. Fourth, because this was a single-center study, the findings might not apply to other centers because of differences in patient backgrounds and examiners’ examination skills. The reproducibility of our results should be confirmed in a multicenter study. Conclusions Patients with subjective cool extremities had higher severity scores, required more invasive treatment, and had more frequent adverse clinical outcomes than those without cool extremities. Cool extremities that persisted for 12 or more hours were associat- ed with in-hospital mortality and AKI requiring RRT. Subjective skin temperature assessment may be useful as a quick, no-cost, and noninvasive tool to help predict high-risk patients and could be used as a “trigger” for additional, time-consuming, and expensive invasive testing. References 1. Sakr Y, Dubois M-J, De Backer D, et al. Persistent microcircu- latory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med 2004;32:1825-31. 2. Dubin A, Henriquez E, Hernández G. Monitoring peripheral perfusion and microcirculation. Curr Opin Crit Care 2018;24:173-80. 3. De Backer D, Ospina-Tascon G, Salgado D, et al. Monitoring the microcirculation in the critically ill patient: current meth- ods and future approaches. Intensive Care Med 2010;36:1813- 25. 4. Kaplan LJ, McPartland K, Santora TA, et al. Start with a sub- jective assessment of skin temperature to identify hypoperfu- sion in intensive care unit patients. J Trauma 2001;50:620-7. 5. Ait-Oufella H, Lemoinne S, Boelle PY, et al. Mottling score predicts survival in septic shock. Intensive Care Med 2011;37:801-7. 6. Lima A, Jansen TC, van Bommel J, et al. The prognostic value of the subjective assessment of peripheral perfusion in critical- ly ill patients. Crit Care Med 2009;37:934-8. 7. Dumas G, Lavillegrand J-R, Joffre J, et al. Mottling score is a strong predictor of 14-day mortality in septic patients whatever vasopressor doses and other tissue perfusion parameters. Crit Care 2019;23:211. 8. Wiersema R, Koeze J, Eck RJ, et al. Clinical examination find- ings as predictors of acute kidney injury in critically ill patients. Acta Anaesthesiol Scand 2020;64:69-74. 9. Lin SJ-S, Cheng Y-Y, Chang C-H, et al. Traditional Chinese medicine diagnosis “yang-xu zheng”: significant prognostic predictor for patients with severe sepsis and septic shock. Evid Based Complement Alternat Med 2013;2013:759748. 10. Henning RJ, Wiener F, Valdes S, et al. Measurement of toe temperature for assessing the severity of acute circulatory fail- ure. Surg Gynecol Obstet 1979;149:1-7. 11. Goldhill DR, McNarry AF, Hadjianastassiou VG, et al. The longer patients are in hospital before Intensive Care admission the higher their mortality. Intensive Care Med 2004;30:1908- 13. 12. Onwochei DN, Fabes J, Walker D, et al. Critical care after major surgery: a systematic review of risk factors for unplanned admission. Anaesthesia 2020;75:e62-74. 13. Harris S, Singer M, Sanderson C, et al. Impact on mortality of prompt admission to critical care for deteriorating ward patients: an instrumental variable analysis using critical care bed strain. Intensive Care Med 2018;44:606-15. 14. Contou D, de Prost N. Skin mottling. N Engl J Med 2016;375:2187. 15. Kazune S, Caica A, Volceka K, et al. Relationship of mottling score, skin microcirculatory perfusion indices and biomarkers of endothelial dysfunction in patients with septic shock: an observational study. Crit Care 2019;23:311. 16. Coudroy R, Jamet A, Frat J-P, et al. Incidence and impact of skin mottling over the knee and its duration on outcome in crit- ically ill patients. Intensive Care Med 2015;41:452-9. 17. van Genderen ME, Paauwe J, de Jonge J, et al. Clinical assess- ment of peripheral perfusion to predict postoperative compli- cations after major abdominal surgery early: a prospective observational study in adults. Crit Care 2014;18:R114. 18. Ait-Oufella H, Bige N, Boelle PY, et al. Capillary refill time exploration during septic shock. Intensive Care Med 2014;40:958-64. 19. Hernandez G, Pedreros C, Veas E, et al. Evolution of peripher- al vs metabolic perfusion parameters during septic shock resuscitation. A clinical-physiologic study. J Crit Care 2012;27:283-8. 20. Puskarich MA, Trzeciak S, Shapiro NI, et al. Whole blood lac- tate kinetics in patients undergoing quantitative resuscitation for severe sepsis and septic shock. Chest 2013;143:1548-53. 21. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed ther- apy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368-77. 22. Dellinger 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;41:580- 637. 23. Kushimoto S, Akaishi S, Sato T, et al. Lactate, a useful marker for disease mortality and severity but an unreliable marker of tissue hypoxia/hypoperfusion in critically ill patients. Acute Med Surg 2016;3:293-7. 24. Rodrigo V, Guillamet CV, Rishi MA, et al. Physical examina- tion in the Intensive Care Unit: opinions of physicians at three teaching hospitals. Southwest J Pulm Crit Care 2015;10:34-43. Article [page 76] [Emergency Care Journal 2024; 20:12569] Non -co mmerc ial us e o nly