Sepsis is a life-threatening syndrome arising from a dysregulated host response to infection; the incidence of sepsis has increased by more than 30% in recent years [
Lipocalin (LCN) family proteins are evolutionarily conserved small proteins (18-40 kDa) [
Considering the potential role of Lcn10 in heart failure, we hypothesized that serum Lcn10 levels may adaptively contribute to the pathogenesis of sepsis in patients who have cardiac dysfunction. As such, the current study investigated serum Lcn10 levels in sepsis patients at the time of admission to characterize the possible correlation between Lcn10 levels and sepsis-induced myocardial dysfunction. We sought to develop a sensitive, precise, and specific biomarker for the assessment of SIMD.
The protocol of the study conformed to the ethical guidelines of the 2008 Helsinki Declaration. Renmin Hospital of Wuhan University guaranteed appropriate ethical and bioethical procedures and certificated this study (no. WDRY2019-K027). After approval of the study, all potential septic patients in ICU at the admission and 20 healthy donors were enrolled. Subjects in the control group were healthy donors who were recruited in the Medical Examination Center. All participants or their authorizer signed informed consent forms before the study. Screening criteria for sepsis performed by the attending physician is suspected infection and
The inclusion criteria were as follows: (1) meeting the diagnostic criteria for Sepsis-3 developed by the American Society of Critical Care Medicine/European Society of Intensive Care Medicine (note: septic shock refers to sepsis patients who required a vasopressor in order to maintain a mean arterial blood pressure greater than 65 mmHg and a serum lactate level higher than 2 mmol/L after sufficient fluid resuscitation) [
In this study, diagnosis of sepsis-induced myocardial dysfunction followed the criteria used in Mayo Clinic which include
On enrollment, the team for this study created standardized case report forms (CRF) to record every patient’s characteristic data. At the time of the ICU admission, clinical data were recorded, including demographic characteristics, past medical history, vital signs, physical examination results, laboratory data, electrocardiogram (ECG), and imaging data. Blood samples were collected for white blood cell count, blood biochemistry, arterial blood gas, blood cultures, and urine cultures at the same time point by the laboratory center of the hospital. The cardiac index (CI) was detected by using Lifegard ICG Hemodynamic Monitors (Analogic, USA). Additionally, we checked the vasopressor usage and hospital length of stay from patients’ medical charts. We calculated Acute Physiology, Age and Chronic Health Evaluation II (APACHE) scores and Sequential Organ Failure Assessment (SOFA) scores after all data were collected [
After entering ICU, we collected patients’ blood samples at the same time as other blood tests before the administration of any treatment. Then, samples were placed at room temperature for 30 min followed by centrifugation at 3000 rpm for 15 min at 4°C. The serum samples were stored at -80°C until further analysis. Lcn10 levels in the collected serum samples were measured in duplicate by ELISA kits from Elabscience Biotechnology Co., Ltd. (Wuhan, China) according to the manufacturer’s instructions. A Multiskan Mk3 microplate reader (Thermo Scientific, USA) was utilized to analyze absorbance at 450 nm of each sample.
Data were analyzed using the Statistical Package for the Social Sciences (ver 19.0) or GraphPad Prism 7 (GraphPad Software), and
To determine diagnostic values, we assessed ROC curves and areas under the receiver operating characteristic curves (AUCs) to discriminate SIMD from sepsis and predict 28-day mortality. The best cut-off value was identified as serum Lcn10 levels present the greatest total of sensitivity and specificity by the Youden index (
Differences between the curves were assessed using the log-rank test. A two-sided significance level of 0.05 was used for statistical inference, and all statistical tests were two-tailed.
This prospective study screened a total of 96 patients who were admitted to the ICU, 75 of them were enrolled in the final study (Figure
Flow diagram and outcome of the study participants. SIMD: sepsis-induced myocardial dysfunction.
Demographic and clinical data of the study population. Values expressed in percentages (%) indicate the proportion of patients within each cohort for each variable. Data are presented as median with interquartile ranges (IQR) where specified. SIMD: sepsis-induced myocardial dysfunction; BMI: body mass index; BP: blood pressure; CRP: c-reactive protein; PCT: procalcitonin; APACHE: Acute Physiology: Age and Chronic Health Evaluation; SOFA: Sequential Organ Failure Assessment; NT-proBNP: N-terminal pro-b-type natriuretic peptide; hs-TnI: high-sensitivity troponin I; CK-MB: creatine kinase-MB; Lcn10: lipocalin 10.
Variables | Non-SIMD ( | SIMD ( | |
---|---|---|---|
Age (y), median (IQR) | 62.5 (55.3-68.8) | 58.0 (53.0-67.0) | 0.189 |
Gender, male, | 20 (50%) | 21 (60%) | 0.386 |
BMI (kg/m2), median (IQR) | 22.0 (20.0-25.75) | 20.0 (19.0-22.0) | 0.071 |
Hospital length of stay (days), median (IQR) | 7 (5-12) | 7 (6-11) | 0.905 |
Death before day 28, | 4 (10%) | 11 (31.4%) | 0.021 |
Systolic BP (mmHg), median (IQR) | 97 (86-119) | 90 (80-97) | 0.046 |
Diastolic BP (mmHg), median (IQR) | 54 (44-64) | 52 (45-62) | 0.864 |
Heart rate (beat/min), median (IQR) | 91 (82-110) | 93 (84-112) | 0.381 |
Lactate (mmol/L), median (IQR) | 1.225 (0.890-2.175) | 3.760 (2.330-5.890) | <0.001 |
CRP (mg/L), median (IQR) | 149 (99-203) | 156 (89-244) | 0.737 |
PCT (ng/mL), median (IQR) | 4.76 (1.27-15.30) | 9.43 (2.18-79.34) | 0.359 |
APACHE II score, median (IQR) | 17 (13-22) | 21 (15-26) | 0.015 |
SOFA score, median (IQR) | 7 (5-10) | 11 (9-14) | <0.001 |
NT-proBNP (pg/mL), median (IQR) | 442 (208-2129) | 983 (534-2310) | 0.053 |
hs-TnI (ng/mL), median (IQR) | 0.328 (0.151-1.854) | 1.030 (0.540-2.780) | 0.004 |
CK-MB (ng/mL), median (IQR) | 5.74 (3.19-12.07) | 11.71 (3.08-19.65) | 0.350 |
Myoglobin ( | 154 (89-347) | 207 (144-456) | 0.057 |
Diagnostic value of Lcn10 in sepsis-induced myocardial dysfunction. (a) Serum Lcn10 levels in the healthy donors, SIMD group, and non-SIMD group. (b) Receiver operator characteristic curves of Lcn10 for the diagnosis of SIMD. The AUC of Lcn10 for the diagnosis of SIMD in septic patients was 0.797 (
Given that sepsis patients with myocardial dysfunction exhibited higher levels of circulating Lcn10, we next hypothesized that Lcn10 might be a good prognostic marker of SIMD. To this end, ROC curves were generated for Lcn10 for discriminating SIMD with significant
Sensitivity and specificity of Lcn10 levels for diagnosis of SIMD.
Lcn10 cut-off (ng/mL) | Sensitivity (%) | Specificity (%) |
---|---|---|
>1.469 | 100.0 | 2.5 |
>2.664 | 65.7 | 82.5 |
>3.981 | 2.9 | 100.0 |
By ROC curve analysis, we observed that Lcn10 showed stronger power than other cardiac biomarkers in the prediction of 28-day mortality in septic patients. The area under the ROC curve of Lcn10 was 0.753 (
Predictive performance of Lcn10 in 28-day mortality of SIMD patients. Compared to other commonly used biomarkers, ROC curves showed that Lcn10 is a more powerful predictor for 28-day mortality of SIMD patients. The AUC was 0.753 (
In the present study, we observed elevated serum levels of Lcn10 on admission in sepsis patients with myocardial dysfunction. The mechanism of Lcn10 in SIMD has not yet been illustrated. Scheraga et al. reported that Lcn10 expression could be synergistically upregulated by wound and heat shock proteins, which were characterized as acute-phase proteins after infection [
To our knowledge, this is the first prospective clinical study to assess the diagnostic and prognostic value of serum Lcn10 for SIMD. In the present study, we observed that the incidence and mortality rates were consistent with those of previous studies [
However, our findings in patients were different from previous findings based on RNA sequencing data, which showed a dramatic decrease in Lcn10 RNA levels in the human right ventricle (heart failure) and left ventricle (dilated cardiomyopathy) [
At present, classic cardiac biomarkers (i.e., NT-proBNP, troponin I, and CK-MB) are usually used as auxiliary tools for the diagnosis and prognosis of SIMD [
In addition, the definition of SIMD includes the left ventricular systolic dysfunction, diastolic dysfunction, and right ventricular dysfunction induced by sepsis. Indeed, left ventricular ejection fraction (LVEF), LV end-diastolic volume (LVEDV), and the ratio of
Finally, there are a couple of shortcomings in this study. First, in the absence of a gold standard for diagnosing SIMD, a definition based on the combination of LVEF and hs-TnI or NT-proBNP was developed. As already stated, LVEF only represents the contraction function of the heart. As the understanding of SIMD is limited, the more complex aspects of SIMD remain to be uncovered. Second, due to limited access to medical records, a more detailed analysis of cardiac function could not be performed. Finally, this study was conducted at a single center with a small sample size, and the time of parameter collection was limited. However, the wealth of pilot data in this study could be used as the basis for subsequent multicenter studies.
In conclusion, this clinical observation demonstrates that the level of serum Lcn10 in patients with sepsis complicated with cardiac injury is higher than that in non-SIMD sepsis patients when they are admitted to the hospital. The serum Lcn10 level is positively correlated with the incidence of myocardial dysfunction caused by sepsis. Remarkably, Lcn10 is a more reliable predictor of 28-day mortality than other commonly used biomarkers. Our results strongly suggest the practicability of serum Lcn10 levels as a potential predictor of SIMD. To validate the relationship between the level of serum Lcn10 and sepsis-induced myocardial dysfunction, further prospective investigations are warranted.
The underlying data supporting the results are all from our clinical data.
No conflicting relationship exists for any author.
Lu Wang and Wenjie Xie contributed equally to this paper.