Abstract
This study investigated whether serum prealbumin level is associated with mortality in heatstroke patients. A retrospective study of 102 heatstroke patients admitted to the ICU in the Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University from June 2010 to November 2017 was performed they were divided into normal serum prealbumin group (n=79) and low prealbumin group (n=23) according to the difference in PA expression. The clinical data, laboratory inspection data, invasive positive pressure ventilation (IPPV), co-infection, shock and length of ICU stay during the ICU were compared between the two groups of patients. The study endpoints, deaths at admission, were recorded, and the survival curve plotted. Cox regression analysis was performed based on the clinical data of patients, and ROC curve plotted based on Cox multivariate independent prognostic indicators. There were significant differences in clinical variables PLT, ALT, AST, TBIL, ALB, TCH, LDH, TNI, BNP, creatinine, PT, APTT, FBG and D-dimer (P<0.05). The incidence of infection, shock and IPPV was significantly lower in normal serum prealbumin group of patients than those in low prealbumin group (P<0.05). There was a statistically significant difference in short-term survival rate between the groups of patients (χ2=29.101, P<0.001). Prognostic factors for heatstroke patients were IPPV, heart rate, WBC count, PLT count, ALB, PA, TBIL, LDH, CPK, Cr, PCT, PT, APTT, D-dimer, co-infection and shock at admission. Independent prognostic-related factors for heatstroke patients were IPPV, PA level, PLT level, ALB level, CPK level and PT level. When prealbumin <17.95 mg/dl was used as the death threshold for predicting at 28 days, the sensitivity was 77.8%, and the specificity was 85.7%. Significantly associated with the prognosis of heatstroke patients, prealbumin level can be used as an important predictive indicator of the disease progression and worse clinical outcomes.
Keywords: prealbumin, heatstroke, mortality
Introduction
Heat stroke has a high incidence and fatality rate (1), which will continue to increase with global warming (2). Results of epidemiological survey show that the average fatality rate of heat stroke is between 10 and 15%. Once developing into severe heat stroke, it is easy to be complicated by multiple organ dysfunction syndrome (MODS), with a fatality rate >40%. Of the survivors, over 30% of patients suffer from long-term nervous system diseases (2,3). The cluster treatment strategy of ‘early rapid cooling, early rapid expansion, early anticoagulation, and active support of organ function’ can effectively prevent multiple organ dysfunctions and shorten the length of ICU stay, so as to improve prognosis (4). Heatstroke is further divided into typical heatstroke and exertional heatstroke according to whether there are fatigue factors in thermal exposure process (5). Due to its high fatality rate, possibly determining relevant factors for the prognosis of patients reduce the long-term effects on their outcomes.
Prealbumin (PA), synthesized by the liver, the level of which is related to protein metabolism, can sensitively reflect the body's nutrition and liver synthesis function, associated with the inflammatory response in the body (6). Widely used in clinical practice, it has been confirmed to predict the mortality and re-hospitalization rate in patients with acute heart failure (7), mortality in patients with acute kidney injury (8,9), mortality in patients with peritoneal dialysis (10) and mortality in burn patients (11). It can also predict infectious complications after gastric operation (12) and the prognosis of patients in the medical department ICU (13).
At present, the determination of prognostic risk factors for heatstroke patients worldwide, mainly based on the analysis of the relationship between multiple univariates and outcomes in clinical data, ignores the interactive relationship among multiple univariates. In addition, whether prealbumin can predict the prognosis of heatstroke has not been studied. Therefore, in this study, a multivariate analysis of the relationship between clinical parameters and disease outcomes of heatstroke patient at admission in the past 10 years was performed, to determine whether the serum prealbumin level was associated with short-term (28 days) mortality, so as to provide a theoretical basis for guiding their rational treatment and improving the prognosis.
Patients and methods
Subjects
A retrospective study of 118 heatstroke patients admitted to the ICU in the Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University (Changzhou, China), from June 2010 to November 2017, was performed. Except for 1 case of age <18 years, 2 cases of chronic hepatic insufficiency, 2 cases of chronic renal insufficiency and 11 cases with incomplete data caused by dying or discharging within 24 h of admission, 102 patients were eventually included. Among them, 61 male patients and 41 female patients, aged from 20 to 89 years, were divided into normal serum prealbumin group (n=79) and low prealbumin group (n=23) according to the difference in PA expression. In normal serum prealbumin group, there were 50 male patients and 29 female patients, aged from 20 to 73 years, and in low prealbumin group, 11 male patients and 12 female patients, aged from 25 to 89 years. All patients were diagnosed in accordance with China's Diagnostic Standards and Treatment Principles of Occupational Heatstroke (GB11508-89). Exclusion criteria were: Patients aged <18 years, knowing liver and kidney dysfunction and with incomplete data caused by death or discharging within 24 h of admission, were excluded.
This study was approved by the Ethics Committee of the Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University. Patients and their family members were informed and consented to all treatments and examinations.
Grouping scheme
A COBAS8000 analyzer (Roche Diagnostics, Indianapolis, IN, USA) was used to determine serum prealbumin content by immunoturbidimetry. Patients were divided into normal serum prealbumin group (>15 mg/dl) and low prealbumin group (≤15 mg/dl) according to serum prealbumin level.
Observation indicators
Main observation indicators: i) General clinical data and vital signs on admission, including age, sex, body mass index (BMI), systolic blood pressure, heart rate and respiratory rate; ii) major underlying diseases such as hypertension, diabetes mellitus, chronic obstructive pulmonary disease, stroke and tumor and iii) laboratory inspection data, including white blood cell (WBC) count, hemoglobin, platelet (PLT) count, blood sodium, blood potassium, alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), total cholesterol (TCH), lactic dehydrogenase (LDH), phosphocreatine kinase (CPK), forebrain natriuretic peptide (BNP), troponin I (TNI), serum creatinine (Cr), albumin (ALB), prealbumin (PA), C-reactive protein (CRP), procalcitonin (PCT), prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen (FBG) and D-dimer. Blood samples from all patients were obtained within 24 h of admission, used for the determination of the above indicators.
Secondary observation indicators: i) Invasive positive pressure ventilation (IPPV), co-infection, shock, length of ICU stay and outcome during the ICU were compared between two groups of patients; ii) the study endpoints, deaths at admission, were recorded, and the survival curve plotted; iii) Cox regression analysis was performed based on the clinical data of patients and iv) ROC curve was plotted based on Cox multivariate independent prognostic indicators.
Statistical analysis
SPSS 20.0 (IBM Corp., Armonk, NY, USA) software was used for data processing, Shapiro-Wilk for the normality test. The measurement data in accordance with the normal distribution were expressed as mean ± standard deviation (SD), and t-test was used for the comparison between two groups. That not in accordance with the normal distribution were expressed as median (quartile) [M (QL-QU)], and Mann-Whitney U test was used for the comparison between two groups. The χ2 test was used for comparison among count data, Kaplan-Meier survival curves for the analysis of survival rates of two groups and Mantel-Cox was the log-rank test. Univariate Cox regression analysis was performed to determine the association between each potential variable and the mortality in patients, and multivariate Cox regression analysis to identify independent variables predicting the mortality in patients. The receiver operating characteristic (ROC) curve was plotted for statistically significant indicators of multivariate regression analysis. The area under the ROC curve (AUC) was calculated to assess the ability of indicators to predict the mortality. P<0.05 was considered to indicate a statistically significant difference.
Results
Comparison of clinical data of patients
The clinical data of the groups were compared, and it was found that there was no statistically significant difference between them in age, sex, BMI, heart rate, respiratory rate, history of hypertension, history of diabetes mellitus, COPD, history of stroke, history of tumor and number of typical heat stroke (P>0.05), and in NA, K, hemoglobin, WBC CPK, PCT and CRP (P>0.05), but there were significant differences in clinical variables PLT, ALT, AST, TBIL, ALB, TCH, LDH, TNI, BNP, creatinine, PT, APTT, FBG and D-dimer (P<0.05) (Table I).
Table I.
Characteristics of study cohort.
| Characteristics | All (n=102) | Normal serum prealbumin group (n=79) | Low prealbumin group (n=23) | P-value |
|---|---|---|---|---|
| PA (mg/dl), median (IQR) | 19.56±6.21 | 22.05±4.51 | 10.99±2.38 | – |
| Age (years), median (IQR) | 66 (50-77.25) | 66 (52.50–77.50) | 73 (49-77) | 0.916 |
| Male, n (%) | 61 (59.80) | 50 (63.29) | 11 (47.83) | 0.229 |
| BMI (kg/m2), median (IQR) | 24.55 (20.58–26.93) | 24.60 (20.60–26.80) | 22.80 (20.50–26.35) | 0.819 |
| SBP (mmHg) | 120.39±22.93 | 123.86±21.67 | 108.47±23.60 | 0.004 |
| Heart rate (bpm), median (IQR) | 102.5 (80-124.5) | 96 (79-117.5) | 113 (94.5-138) | 0.017 |
| Respiratory rate (bpm), median (IQR) | 21 (18-27.25) | 21 (18-26.50) | 24 (18.50–28.50) | 0.572 |
| Previous disease | ||||
| Hypertension, n (%) | 29 (28.43) | 26 (32.91) | 3 (13.04) | 0.071 |
| Diabetes mellitus, n (%) | 11 (10.78) | 9 (11.39) | 2 (8.69) | 0.714 |
| COPD, n (%) | 5 (4.90) | 3 (3.79) | 2 (8.69) | 0.315 |
| Stroke, n (%) | 8 (7.84) | 7 (8.86) | 1 (4.35) | 0.679 |
| Tumor, n (%) | 4 (3.92) | 2 (2.53) | 2 (8.69) | 0.218 |
| NA (mmol/l), median (IQR) | 137.20 (131.35–140.15) | 137.70 (131.00–140.20) | 136.90 (132.40–138.80) | 0.788 |
| K (mmol/l) | 3.56±0.67 | 3.60±0.62 | 3.41±0.78 | 0.267 |
| Hemoglobin (g/l) | 124.22±17.72 | 124.97±16.21 | 121.60±22.34 | 0.425 |
| WBC (109/l), median (IQR) | 12.84 (9.96–15.99) | 12.17 (9.02–15.26) | 13.88 (12.10–18.19) | 0.070 |
| PLT (109/l), median (IQR) | 121.10 (80-172.25) | 132 (92-179.20) | 87 (51-117.50) | 0.006 |
| ALT (U/l), median (IQR) | 100.50 (51.18–152.95) | 95 (50-139.50) | 127 (95.90-267) | 0.016 |
| AST (U/l), median (IQR) | 140.55 (90.13–213.63) | 118 (83.30–180.70) | 185.30 (141.50-430) | <0.001 |
| TBIL (umol/l), median (IQR) | 17 (10.35–24.10) | 15.60 (10-20.90) | 22.60 (17.60–27.55) | 0.004 |
| ALB (g/l) | 34.08±4.81 | 35.22±4.43 | 30.15±3.96 | <0.001 |
| TCH (mmol/l), median (IQR) | 3.45 (2.79–4.11) | 3.59 (3.05–4.21) | 3.20 (2.60–3.66) | 0.006 |
| LDH (U/l), median (IQR) | 416.80 (316.15–643.40) | 394 (308.45–566.60) | 503.40 (394.65–915.90) | 0.020 |
| CPK (U/l), median (IQR) | 754.55 (290.90-2,059.78) | 688.80 (289.15–1831.45) | 1,109 (495.15-3,081.60) | 0.172 |
| TNI (ng/ml), median (IQR) | 0.15 (0.04–0.58) | 0.10 (0.02–0.36) | 0.57 (0.16–0.84) | <0.001 |
| BNP (pg/ml), median (IQR) | 488 (178-2633) | 394 (170-1,970) | 2480 (255-6505) | 0.037 |
| Creatinine (umol/l), median (IQR) | 93.25 (73.83–127.15) | 87 (72.05–117.90) | 108.40 (88.30–138.10) | 0.036 |
| PCT (ng/ml), median (IQR) | 3.42 (0.90–13.39) | 3.10 (0.81–10.66) | 5.85 (2.79–25.30) | 0.054 |
| CRP (mg/l), median (IQR) | 4.45 (1.38-14) | 4.20 (1.62–12.95) | 8.50 (0.75-25) | 0.680 |
| PT(s), median (IQR) | 15.10 (12.78–18.93) | 14.10 (12.40–16.75) | 24 (17.20–27.65) | <0.001 |
| APTT(s), median (IQR) | 29.55 (25.35–36.58) | 27.80 (24.95–31.60) | 37.50 (34.50–43.85) | <0.001 |
| FBG (g/l), median (IQR) | 2.16 (1.70–2.61) | 2.22 (1.81–2.78) | 1.80 (1.31–2.40) | 0.015 |
| D-dimer (mg/l), median (IQR) | 1.87 (1.03–4.64) | 1.76 (0.76–4.37) | 3.11 (1.75–10.99) | 0.010 |
PA, prealbumin; BMI, body mass index; SBP, systolic blood pressure; COPD, chronic obstructive pulmonary disease; WBC, white blood cell; PLT, platelet; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBIL, total bilirubin; ALB, albumin; TCH, total cholesterol; LDH, lactic dehydrogenase; CPK, creatine phosphate kinase; TNI, troponin; BNP, brain natriuretic peptide; PCT, procalcitonin; CRP, C-reactive protein; PT, prothrombin time; APTT, activated partial thromboplastin time; FBG, fibrinogen.
Comparison of adverse conditions in the course of treatment, length of stay in the hospital, IPPV times and length of ICU stay between two groups of patients
The adverse conditions, length of stay in the hospital and IPPV times in two groups of patients were counted. It was found that the incidence of infection, shock and IPPV was significantly lower in normal serum prealbumin group of patients than those in low prealbumin group, with a statistically significant difference (P<0.05). There was no statistically significant difference between two groups of patients in the length of stay in the hospital and length of ICU stay (P>0.05) (Table II).
Table II.
Adverse conditions in the course of treatment, length of stay in the hospital and length of ICU stay in both groups.
| Index | Normal serum prealbumin group (n=79) | Low prealbumin group (n=23) | P-value |
|---|---|---|---|
| Typical heatstroke, n (%) | 54 (68.36) | 15 (65.22) | 0.803 |
| Infection, n (%) | 45 (56.96) | 17 (73.91) | 0.156 |
| Shock, n (%) | 18 (22.78) | 15 (65.22) | <0.001 |
| IPPV, n (%) | 19 (24.05) | 13 (56.52) | 0.005 |
| Days of ICU (days) | 8 (6-12) | 7 (3-10) | 0.161 |
IPPV, invasive positive pressure ventilation; ICU, intensive care unit.
Length of survival of patients at admission
The length of survival of two groups of patients at admission was calculated. It was found that altogether 8 patients died in normal serum prealbumin group of 79 patients, with the length of survival of 26.177±0.649 days, and 13 patients died in low prealbumin group of 23 patients, with the length of survival of 16.130±2.559 days. By comparison, it was found that there was a statistically significant difference in short-term survival rate between the groups of patients (χ2=29.101, P<0.001) (Fig. 1).
Figure 1.
Kaplan-Meier survival curves according to prealbumin (cut-off 15 mg/dl) at admission. Patients with prealbumin ≤15 mg/dl had significantly poorer survival than those with higher prealbumin.
Cox regression analysis
A univariate Cox regression analysis of the collected data was performed. It was found that prognostic factors for heatstroke patients were IPPV, heart rate, WBC count, PLT count, ALB, PA, TBIL, LDH, CPK, Cr, PCT, PT, APTT, D-dimer and co-infection and shock at admission. Then, a multivariate Cox regression analysis of indicators with differences in univariates was performed. It was found that independent prognostic-related factors for heatstroke patients were IPPV (OR: 7.111, 95% CI: 1.977–25.574, P=0.003), PA level (OR: 1.204, 95% CI: 1.068–1.357, P=0.002), PLT level (OR: 1.010, 95% CI: 1.002–1.019, P=0.017), ALB level (OR: 1.159, 95% CI: 1.021–1.316, P=0.022), CPK level (per 1,000 U/l) (OR: 1.152, 95% CI: 1.052–1.262, P=0.002) and PT level (OR: 1.096, 95% CI: 1.032–1.164, P=0.003) (Tables III and IV).
Table III.
Predictors of mortality in heat stroke patients, univariate Cox regression analysis.
| Variables | β | SE | Wald | df | OR | 95% CI | P-value |
|---|---|---|---|---|---|---|---|
| Age | 0.008 | 0.014 | 0.335 | 1 | 1.008 | 0.981–1.035 | 0.563 |
| Sex | 0.126 | 0.449 | 0.078 | 1 | 1.134 | 0.470–2.736 | 0.780 |
| BMI | 0.026 | 0.059 | 0.196 | 1 | 1.027 | 0.914–1.153 | 0.658 |
| SBP | 0.013 | 0.010 | 1.689 | 1 | 1.013 | 0.993–1.033 | 0.194 |
| Heart rate | 0.023 | 0.008 | 8.555 | 1 | 1.024 | 1.008–1.040 | 0.003 |
| Respiratory rate | 0.027 | 0.030 | 0.793 | 1 | 1.027 | 0.968–1.090 | 0.373 |
| Hypertension | −0.016 | 0.483 | 0.001 | 1 | 0.984 | 0.382–2.536 | 0.973 |
| Diabetes mellitus | 0.240 | 0.744 | 0.104 | 1 | 1.271 | 0.296–5.459 | 0.747 |
| COPD | 3.084 | 4.285 | 0.518 | 1 | 21.854 | 0.005–96958.873 | 0.472 |
| Stroke | 0.563 | 1.025 | 0.302 | 1 | 1.757 | 0.236–13.091 | 0.582 |
| Tumor | 3.067 | 4.774 | 0.413 | 1 | 21.478 | 0.002–248485.152 | 0.521 |
| NA | 0.025 | 0.020 | 1.648 | 1 | 1.026 | 0.987–1.066 | 0.199 |
| K | −0.396 | 0.328 | 1.462 | 1 | 0.673 | 0.354–1.279 | 0.227 |
| WBC | 0.074 | 0.035 | 4.332 | 1 | 1.076 | 1.004–1.153 | 0.037 |
| Hemoglobin | 0.064 | 0.131 | 0.244 | 1 | 1.067 | 0.826–1.378 | 0.621 |
| per 10 g/l | |||||||
| PLT | 0.017 | 0.005 | 12.599 | 1 | 1.017 | 1.008–1.027 | <0.001 |
| ALT per 100 U/l | 0.091 | 0.062 | 2.163 | 1 | 1.095 | 0.97–1.236 | 0.141 |
| AST per 100 U/l | 0.050 | 0.038 | 1.674 | 1 | 1.051 | 0.975–1.133 | 0.196 |
| TBIL | 0.026 | 0.008 | 9.684 | 1 | 1.026 | 1.010–1.043 | 0.002 |
| ALB | 0.095 | 0.044 | 4.636 | 1 | 1.100 | 1.009–1.200 | 0.031 |
| TCH | −0.412 | 0.272 | 2.288 | 1 | 0.662 | 0.389–1.129 | 0.130 |
| LDH per 100 U/l | 0.129 | 0.043 | 8.920 | 1 | 1.138 | 1.045–1.239 | 0.003 |
| CPK per 1,000 U/l | 0.184 | 0.043 | 18.525 | 1 | 1.202 | 1.105–1.306 | <0.001 |
| TNI | 0.023 | 0.054 | 0.178 | 1 | 1.023 | 0.920–1.139 | 0.673 |
| BNP per 1,000 pg/ml | 0.033 | 0.024 | 1.837 | 1 | 1.033 | 0.986–1.083 | 0.175 |
| Creatinine per | 0.039 | 0.013 | 9.132 | 1 | 1.039 | 1.014–1.066 | 0.003 |
| 10 umol/l | |||||||
| PCT | 0.021 | 0.006 | 11.877 | 1 | 1.022 | 1.009–1.034 | 0.001 |
| CRP per 10 mg/l | 0.038 | 0.067 | 0.332 | 1 | 1.039 | 0.911–1.186 | 0.564 |
| PA | 0.189 | 0.041 | 21.536 | 1 | 1.208 | 1.115–1.308 | <0.001 |
| PT | 0.141 | 0.022 | 39.364 | 1 | 1.151 | 1.102–1.203 | <0.001 |
| APTT | 0.046 | 0.013 | 11.952 | 1 | 1.047 | 1.020–1.075 | 0.001 |
| FBG | −0.467 | 0.306 | 2.327 | 1 | 0.627 | 0.344–1.142 | 0.127 |
| D-dimer | 0.057 | 0.020 | 8.117 | 1 | 1.059 | 1.018–1.102 | 0.004 |
| Type | −0.067 | 0.463 | 0.021 | 1 | 0.936 | 0.378–2.318 | 0.886 |
| Infection | 2.705 | 1.025 | 6.966 | 1 | 14.960 | 2.007–111.533 | 0.008 |
| Shock | 1.838 | 0.484 | 14.401 | 1 | 6.285 | 2.432–16.242 | <0.001 |
| IPPV | 2.855 | 0.626 | 20.823 | 1 | 17.369 | 5.097–59.196 | <0.001 |
BMI, body mass index; SBP, systolic blood pressure; COPD, chronic obstructive pulmonary disease; WBC, white blood cell; PLT, platelet; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBIL, total bilirubin; ALB, albumin; TCH, total cholesterol; LDH, lactic dehydrogenase; CPK, creatine phosphate kinase; TNI, troponin; BNP, brain natriuretic peptide; PCT, procalcitonin; CRP, C-reactive protein; PA, prealbumin; PT, prothrombin time; APTT, activated partial thromboplastin time; FBG, fibrinogen; IPPV, invasive positive pressure ventilation.
Table IV.
Predictors of mortality in heat stroke patients, multivariate Cox regression analysis.
| Variables | β | SE | Wald | df | OR | 95.0% CI | P-value |
|---|---|---|---|---|---|---|---|
| IPPV | 1.962 | 0.653 | 9.022 | 1 | 7.111 | 1.977–25.574 | 0.003 |
| PLT | 0.010 | 0.004 | 5.646 | 1 | 1.010 | 1.002–1.019 | 0.017 |
| ALB | 0.148 | 0.065 | 5.221 | 1 | 1.159 | 1.021–1.316 | 0.022 |
| CPK | 0.142 | 0.046 | 9.319 | 1 | 1.152 | 1.052–1.262 | 0.002 |
| per 1,000 U/l | |||||||
| PA | 0.186 | 0.061 | 9.271 | 1 | 1.204 | 1.068–1.357 | 0.002 |
| PT | 0.091 | 0.031 | 8.811 | 1 | 1.096 | 1.032–1.164 | 0.003 |
IPPV, invasive positive pressure ventilation; PLT, platelet; ALB, albumin; CPK, creatine phosphate kinase; PA, prealbumin; PT, prothrombin time.
ROC curve analysis
ROC curve was plotted to further compare the role of these six indicators in predicting patient mortality. It was found that IPPVAUC=0.842 (95% CI: 0.743–0.941), PAAUC=0.845 (95% CI: 0.760–0.931), PLTAUC=0.780 (95% CI: 0.668–0.892), ALBAUC=0.652 (95% CI: 0.527–0.777), CPKAUC=0.694 (95% CI: 0.558–0.829) and PTAUC=0.862 (95% CI: 0.771–0.953). When prealbumin <17.95 mg/dl was used as the death threshold for predicting at 28 days, the sensitivity was 77.8%, and the specificity was 85.7% (Fig. 2 and Table V).
Figure 2.
Comparison of AUC values among various parameters for survival in heat stroke patients. AUC, area under the receiver operating characteristic curve; IPPV, invasive positive pressure ventilation; CPK, creatine phosphate kinase; PT, prothrombin time; PLT, platelet; ALB, albumin; PA, prealbumin.
Table V.
Comparison of AUC among various parameters for survival in heat stroke patients.
| Variables | AUC | SE | P-value | 95% CI | Sensitivity | Specificity | Cut-off |
|---|---|---|---|---|---|---|---|
| IPPV | 0.842 | 0.051 | <0.001 | 0.743–0.941 | 0.857 | 0.827 | |
| CPK | 0.694 | 0.069 | 0.006 | 0.558–0.829 | 0.571 | 0.778 | 1676.200 |
| PT | 0.862 | 0.046 | <0.001 | 0.771–0.953 | 0.810 | 0.778 | 17.650 |
| PLT | 0.780 | 0.057 | <0.001 | 0.668–0.892 | 0.905 | 0.568 | 125.500 |
| ALB | 0.652 | 0.064 | 0.032 | 0.527–0.777 | 0.619 | 0.691 | 32.450 |
| PA | 0.845 | 0.044 | <0.001 | 0.760–0.931 | 0.857 | 0.778 | 17.950 |
AUC, area under the receiver operating characteristic curve; IPPV, invasive positive pressure ventilation; CPK, creatine phosphate kinase; PT, prothrombin time; PLT, platelet; ALB, albumin; PA, prealbumin.
Discussion
In this study, the relationship between serum prealbumin and mortality in heatstroke patients was shown. The total in-hospital mortality was (20.6%) in this study cohort, lower than that in other reports (14,15). Possible reasons are: First, all heatstroke patients were admitted to the ICU for standard treatment. Secondly, a certain degree of clinical experience was accumulated in the early stage of this group, that is, the cluster treatment strategy of ‘early rapid cooling, early rapid expansion, early anticoagulation, and active support of organ function’ was proposed, effectively reducing the mortality. Finally, some critically ill patients were not included due to their incomplete data, as a result of failure of admission or admission to hospital in <24 h, which affected the calculation of mortality. In this study, prealbumin has been shown to be a predictor of mortality in heatstroke patients. The cumulative survival rate of patients with prealbumin >15 mg/dl is significantly better than that of patients with prealbumin ≤15 mg/dl. The results of multivariate Cox regression analysis have confirmed that low serum prealbumin level on admission is an independent risk factor for the poor prognosis of heatstroke patients. ROC curve analysis has shown that the AUC of prealbumin is greater than that of albumin, with the sensitivity and specificity of it higher than those of albumin, indicating that it is satisfactory for judging the prognosis of heatstroke patients, and that prealbumin may be clinically useful. This is the first clinical study at home and abroad of serum prealbumin level in heatstroke patients.
Composed of the same four subunits, the main function of serum prealbumin is to transport thyroxine and vitamin A, which promotes lymphocyte maturation and immunity enhancement. Due to its small base and shorter half-life (2 days), prealbumin can effectively reflect the synthesis of protein in the body, which has become an internationally recognized indicator of nutritional status. Moreover, as a negative acute phase reaction protein, the expression of serum prealbumin has a significant decrease in tumor, cirrhosis and inflammatory response (16). According to the Nutritional Health Consensus Group, serum prealbumin >15 mg/dl indicates a lower risk of malnutrition (17). Therefore, in this study, 15 mg/dl of serum prealbumin was selected as the cut-off value, 23 patients had ≤15 mg/dl, accounting for 22.5%. In low prealbumin group of patients, the levels of PLT, ALB and TCH were lower, PT and APTT significantly prolonged, and there was no significant difference in PCT and CRP reflecting the inflammatory state.
Cardiac dysfunction is the leading cause of death in patients with heat-related diseases (18). Experimental studies have confirmed that under heat stress conditions, severe damage to myocardial cells is characterized by vacuolar changes and partial necrosis of cells (19). Clinical studies have found that myocardial markers are significantly elevated after heat stress, the elevated levels of which can better predict the severity of myocardial damage and heat stroke (20). The multivariate regression analysis of this study has shown that CPK has certain value in judging the condition and prognosis, consistent with the results of Ye et al (21).
Coagulation dysfunction can occur in the early stage of heatstroke patients, resulting in DIC whose characteristic pathological injury is extensive thrombosis. On the one hand, continuous consumption causes thrombocytopenia and clotting factor deficiency leading to bleeding. On the other hand, extensive thrombosis causes tissue perfusion disorder that is one of the important mechanisms of the occurrence of severe heat stroke MODS (22,23). The multivariate regression analysis of Zhao et al (24) found that DIC is an main risk factor affecting the prognosis of exertional heatstroke patients, with the mortality of patients coexisting of 70.83%. Therefore, coagulation dysfunction is the most severe complication of heatstroke (24). In the blood specimen monitoring of patients with severe heat stroke within 24 h of admission, Pan et al (25) found that PT and APTT significantly prolong, but PLT significantly decreases in death group compared with survival group. The analysis of receiver operating characteristic (ROC) curve has shown that PLT has a predictive value for patients with severe heat stroke (25). Li et al (26) found that prothrombin time on admission, international normalized ratio and fibrinogen can effectively predict the clinical prognosis of heatstroke patients (26). In this study, a multivariate regression analysis was performed on the clinical data of 102 heatstroke patients. The results showed that PT and PLT are independent risk factors affecting the prognosis of patients among all the main clinical parameters collected at admission.
In this study, patients with low serum prealbumin level are found to be correlated with invasive positive pressure ventilation, with invasive mechanical ventilation being an independent risk factor for death in heatstroke patients. Nutritional supplementation in patients with low prealbumin level may shorten extubation time.
Increasing evidence shows that not caused by simple thermal exposure resulting in direct injury, the pathophysiological response of severe heat stroke MODS is a systemic inflammatory response syndrome secondary to thermal injury, which in turn manifests as ‘sepsis’, further triggering MODS (27,28). After heatstroke, the pro-inflammatory and anti-inflammatory factors in the body are out of balance, and then tissue damage caused by pro-inflammatory response and immunosuppression caused by excessive anti-inflammatory response may occur in patients (29). Experimental studies have shown that when the inflammatory response in heatstroke rats is reduced with different methods, the degree of organ damage is reduced, with improved survival rate (30). At present, there are few studies on the changing trends of PCT and CRP, generally considered to be acute phase biomarkers of inflammatory response, in heatstroke patients. Serum PCT monitoring was performed on 68 exertional heatstroke patients within 2 h of admission. Tong et al (31) found that the PCT of non-survivors is higher than that of survivors. After confounding factors adjusted, PCT concentration is also an independent risk factor for mortality, but may not be a good indicator predicting heat stroke accompanied by infection (31). The studies of Hausfater et al show that serum PCT can be induced to release even in the absence of bacterial infection in exertional heatstroke patients (14). Surprisingly, although the CRP level is moderately elevated [4.45 (1.38-14) mg/l], there is no statistically significant difference between survival group and non-survival group [4.2 (1.4, 12.9) vs. 8.7 (1.4, 24.7), P=0.360]. CRP, also an acute phase reaction protein, is positively correlated with the degree of inflammation and tissue damage (32), but often taking time to respond and increase serum concentration after infection or inflammation (33). Studies have also shown that the kinetics of CRP is slower in sepsis and systemic inflammatory response than that in inflammatory mediator (including PCT, leptin, IL-6 and TNF-α) (34,35). Therefore, the decline in CRP sensitivity in the early stage of heatstroke can be explained by these facts. Used as a differential diagnosis in patients with non-infectious high fever at the early stage of admission, CRP saves patients' valuable time and avoids unnecessary diagnostic checks.
In the univariate analysis used to determine main risk factors affecting the prognosis in this group of studies, the results obtained are similar to those worldwide, indicating that these parameters are representative. Multivariate Cox regression analysis is used to remove confounding factors and screen out independent risk factors affecting the prognosis of heatstroke, so as to analyze and evaluate the sensitivity and specificity of each risk factor. Besides, it evaluates the effect of prealbumin on prognosis for the first time, being the highlight of this group of studies, with accurate and reliable results.
There are also some shortcomings in this study. First, study design is observation, so the causal relationship between prealbumin and other markers needs further confirmation. Secondly, the number of cases included in the study, which is relatively small, is derived from a single hospital, with limited representativeness that may cause statistical bias. In addition, most of the cases are mainly typical heat stroke, cannot fully representing other types of heat stroke. Therefore, further long-term multi-center clinical studies with large samples are necessary for confirming factors affecting the prognosis of heatstroke.
In summary, the measurement of prealbumin is a useful tool evaluating the nutritional and inflammatory status of heatstroke patients. Moreover, the lower level of serum prealbumin is independently associated with the increase of death risk.
Acknowledgements
Not applicable.
Funding
This study was supported by Changzhou Medical and Health Guiding Science and Technology Project (WZ201021).
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Authors' contributions
QY and WL collected and interpreted the general information of patients. JY and JJ acquired and analyzed the οbservation indicators. TX and YZ contributed to statistical analysis. All authors read and approved the final manuscript.
Ethics approval and consent to participate
This study was approved by the Ethics Committee of The Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University (Changzhou, China). Patients who participated in this research had complete clinical data. The signed informed consents were obtained from the patients or the guardians.
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
References
- 1.Glazer JL. Management of heatstroke and heat exhaustion. Am Fam Physician. 2005;71:2133–2140. [PubMed] [Google Scholar]
- 2.Bouchama A, Knochel JP. Heat stroke. N Engl J Med. 2002;346:1978–1988. doi: 10.1056/NEJMra011089. [DOI] [PubMed] [Google Scholar]
- 3.Hammami MM, Bouchama A, Al-Sedairy S, Shail E, AlOhaly Y, Mohamed GE. Concentrations of soluble tumor necrosis factor and interleukin-6 receptors in heatstroke and heatstress. Crit Care Med. 1997;25:1314–1319. doi: 10.1097/00003246-199708000-00017. [DOI] [PubMed] [Google Scholar]
- 4.Lee EJ, Lee SW, Park JS, Kim SJ, Hong YS. Successful treatment of severe heat stroke with selective therapeutic hypothermia using an automated surface cooling device. Resuscitation. 2013;84:e77–e78. doi: 10.1016/j.resuscitation.2013.02.024. [DOI] [PubMed] [Google Scholar]
- 5.Leon LR. Heat stroke and cytokines. Prog Brain Res. 2007;162:481–524. doi: 10.1016/S0079-6123(06)62024-4. [DOI] [PubMed] [Google Scholar]
- 6.Vandevyver S, Dejager L, Vandenbroucke RE, Libert C. An acute phase protein ready to go therapeutic for sepsis. EMBO Mol Med. 2014;6:2–3. doi: 10.1002/emmm.201303524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Franco J, Formiga F, Trullas JC, Salamanca Bautista P, Conde A, Manzano L, Quirós R, Franco ÁG, Ezquerro AM, Montero-Pérez-Barquero M, RICA investigators group Impact of prealbumin on mortality and hospital readmission in patients with acute heart failure. Eur J Intern Med. 2017;43:36–41. doi: 10.1016/j.ejim.2017.05.009. [DOI] [PubMed] [Google Scholar]
- 8.Xie Q, Zhou Y, Xu Z, Yang Y, Kuang D, You H, Ma S, Hao C, Gu Y, Lin S, et al. The ratio of CRP to prealbumin levels predict mortality in patients with hospital-acquired acute kidney injury. BMC Nephrol. 2011;12:30. doi: 10.1186/1471-2369-12-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Berbel MN, Góes CR, Balbi AL, Ponce D. Nutritional parameters are associated with mortality in acute kidney injury. Clinics (São Paulo) 2014;69:476–482. doi: 10.6061/clinics/2014(07)06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lee KH, Cho JH, Kwon O, Kim SU, Kim RH, Cho YW, Jung HY, Choi JY, Kim CD, Kim YL, et al. Low prealbumin levels are independently associated with higher mortality in patients on peritoneal dialysis. Kidney Res Clin Pract. 2016;35:169–175. doi: 10.1016/j.krcp.2016.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yang HT, Yim H, Cho YS, Kim D, Hur J, Kim JH, Lee BC, Seo DK, Kim HS, Chun W. Prediction of clinical outcomes for massively-burned patients via serum transthyretin levels in the early postburn period. J Trauma Acute Care Surg. 2012;72:999–1005. doi: 10.1097/TA.0b013e3182413bd8. [DOI] [PubMed] [Google Scholar]
- 12.Bae HJ, Lee HJ, Han DS, Suh YS, Lee YH, Lee HS, Cho JJ, Kong SH, Yang HK. Prealbumin levels as a useful marker for predicting infectious complications after gastric surgery. J Gastrointest Surg. 2011;15:2136–2144. doi: 10.1007/s11605-011-1719-z. [DOI] [PubMed] [Google Scholar]
- 13.Li L, Dai L, Wang X, Wang Y, Zhou L, Chen M, Wang H. Predictive value of the C-reactive protein-to-prealbumin ratio in medical ICU patients. Biomarkers Med. 2017;11:329–337. doi: 10.2217/bmm-2016-0266. [DOI] [PubMed] [Google Scholar]
- 14.Hausfater P, Hurtado M, Pease S, Juillien G, Lvovschi VE, Salehabadi S, Lidove O, Wolff M, Bernard M, Chollet-Martin S, et al. Is procalcitonin a marker of critical illness in heatstroke? Intensive Care Med. 2008;34:1377–1383. doi: 10.1007/s00134-008-1083-y. [DOI] [PubMed] [Google Scholar]
- 15.Bouchama A, Dehbi M, Mohamed G, Matthies F, Shoukri M, Menne B. Prognostic factors in heat wave related deaths: A meta-analysis. Arch Intern Med. 2007;167:2170–2176. doi: 10.1001/archinte.167.20.ira70009. [DOI] [PubMed] [Google Scholar]
- 16.Davis CJ, Sowa D, Keim KS, Kinnare K, Peterson S. The use of prealbumin and C-reactive protein for monitoring nutrition support in adult patients receiving enteral nutrition in an urban medical center. J Parenter Enteral Nutr. 2012;36:197–204. doi: 10.1177/0148607111413896. [DOI] [PubMed] [Google Scholar]
- 17.Prealbumin in Nutritional Care Consensus Group Measurement of visceral protein status in assessing protein and energy malnutrition: Standard of care. Nutrition. 1995;11:169–171. [PubMed] [Google Scholar]
- 18.Wang X, Yuan B, Dong W, Yang B, Yang Y, Lin X, Gong G. Humid heat exposure induced oxidative stress and apoptosis in cardiomyocytes through the angiotensin II signaling pathway. Heart Vessels. 2015;30:396–405. doi: 10.1007/s00380-014-0523-6. [DOI] [PubMed] [Google Scholar]
- 19.Quinn CM, Duran RM, Audet GN, Charkoudian N, Leon LR. Cardiovascular and thermoregulatory biomarkers of heat stroke severity in a conscious rat model. J Appl Physiol (1985) 2014;117:971–978. doi: 10.1152/japplphysiol.00365.2014. [DOI] [PubMed] [Google Scholar]
- 20.Hausfater P, Doumenc B, Chopin S, Le Manach Y, Santin A, Dautheville S, Patzak A, Hericord P, Mégarbane B, Andronikof M, et al. Elevation of cardiac troponin I during non-exertional heat-related illnesses in the context of a heatwave. Crit Care. 2010;14:R99. doi: 10.1186/cc9034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ye J, Mo W, Chen Y, Yang A. An analysis of laboratory results of parameters of organ function in patients with heat stroke. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2015;27:658–661. doi: 10.3760/cma.j.issn.2095-4352.2015.08.008. (In Chinese) [DOI] [PubMed] [Google Scholar]
- 22.Jilma B, Derhaschnig U. Disseminated intravascular coagulation in heat stroke: A hot topic. Crit Care Med. 2012;40:1370–1372. doi: 10.1097/CCM.0b013e31823d785d. [DOI] [PubMed] [Google Scholar]
- 23.Leon LR, Helwig BG. Heat stroke: Role of the systemic inflammatory response. J Appl Physiol (1985) 2010;109:1980–1988. doi: 10.1152/japplphysiol.00301.2010. [DOI] [PubMed] [Google Scholar]
- 24.Zhao JJ, Zhou JJ, Hu J, Zhou FH, Kang HJ, Liu H, Pan L, Song Q. Analysis of risk factors affecting prognosis of exertional heat stroke. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2013;25:515–518. doi: 10.3760/cma.j.issn.2095-4352.2013.09.002. (In Chinese) [DOI] [PubMed] [Google Scholar]
- 25.Pan ZG, Shao Y, Liu YN, Gu ZT, Zhang XQ, Xu YQ, Su L. Relationship between early coagulability parameters at admission and outcome in patients with severe heatstroke. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2013;25:725–728. doi: 10.3760/cma.j.issn.2095-4352.2013.12.007. (In Chinese) [DOI] [PubMed] [Google Scholar]
- 26.Li Z, Wang Z, Tu X, Lu W, Fang Y, Gao J, Yu S, He R, Shi K, Chu H. Analysis of indicators for clinical prognosis of the patients with heat stroke. Zhonghua Jizhen Yixue Zazhi. 2016;25:1058–1061. (In Chinese) [Google Scholar]
- 27.Liu Z, Sun X, Tang J, Tang Y, Tong H, Wen Q, Liu Y, Su L. Intestinal inflammation and tissue injury in response to heat stress and cooling treatment in mice. Mol Med Rep. 2011;4:437–443. doi: 10.3892/mmr.2011.461. [DOI] [PubMed] [Google Scholar]
- 28.Leon LR, Blaha MD, DuBose DA. Time course of cytokine, corticosterone, and tissue injury responses in mice during heat strain recovery. J Appl Physiol (1985) 2006;100:1400–1409. doi: 10.1152/japplphysiol.01040.2005. [DOI] [PubMed] [Google Scholar]
- 29.Lim CL, Mackinnon LT. The roles of exercise-induced immune system disturbances in the pathology of heat stroke: The dual pathway model of heat stroke. Sports Med. 2006;36:39–64. doi: 10.2165/00007256-200636010-00004. [DOI] [PubMed] [Google Scholar]
- 30.Hsu CC, Niu CS, Lin MT. Decrease of heatstroke-induced multiorgan dysfunction by whole body cooling in streptozotocin-induced diabetic rats. Chin J Physiol. 2009;52:47–55. doi: 10.4077/CJP.2009.AMH010. [DOI] [PubMed] [Google Scholar]
- 31.Tong HS, Liu YS, Wen Q, Tang YQ, Yuan FF, Su L. Serum procalcitonin predicting mortality in exertional heatstroke. Emerg Med J. 2012;29:113–117. doi: 10.1136/emj.2010.107680. [DOI] [PubMed] [Google Scholar]
- 32.Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, Rumley A, Lowe GD, Pepys MB, Gudnason V. C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med. 2004;350:1387–1397. doi: 10.1056/NEJMoa032804. [DOI] [PubMed] [Google Scholar]
- 33.Martin C, Boisson C, Haccoun M, Thomachot L, Mege JL. Patterns of cytokine evolution (tumor necrosis factor-alpha and interleukin-6) after septic shock, hemorrhagic shock, and severe trauma. Crit Care Med. 1997;25:1813–1819. doi: 10.1097/00003246-199711000-00018. [DOI] [PubMed] [Google Scholar]
- 34.Pourakbari B, Mamishi S, Zafari J, Khairkhah H, Ashtiani MH, Abedini M, Afsharpaiman S, Rad SS. Evaluation of procalcitonin and neopterin level in serum of patients with acute bacterial infection. Braz J Infect Dis. 2010;14:252–255. doi: 10.1016/S1413-8670(10)70052-0. [DOI] [PubMed] [Google Scholar]
- 35.Yousef AA, Amr YM, Suliman GA. The diagnostic value of serum leptin monitoring and its correlation with tumor necrosis factor-alpha in critically ill patients: A prospective observational study. Crit Care. 2010;14:R33. doi: 10.1186/cc8911. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.