Prospective Evaluation of the Peripheral Perfusion Index in Assessing the Organ Dysfunction and Prognosis of Adult Patients With Sepsis in the ICU

Abstract

Background: The peripheral perfusion index (PI) reflects microcirculatory blood flow perfusion and indicates the severity and prognosis of sepsis. Method: The cohort comprised 208 patients admitted to the intensive care unit (ICU) with infection, among which 117 had sepsis. Demographics, medication history, ICU variables, and laboratory indexes were collected. Primary endpoints were in-hospital mortality and 28-day mortality. Secondary endpoints included organ function variables (coagulation function, liver function, renal function, and myocardial injury), lactate concentration, mechanical ventilation time, and length of ICU stay. Univariate and multivariate analyses were conducted to assess the associations between the PI and clinical outcomes. Sensitivity analyses were performed to explore the associations between the PI and organ functions in the sepsis and nonsepsis groups. Result: The PI was negatively associated with in-hospital mortality (odds ratio [OR] 0.29, 95% confidence interval [CI] 0.15 to 0.55), but was not associated with 28-day mortality. The PI was negatively associated with the coagulation markers prothrombin time (PT) (β −0.36, 95% CI −0.59 to 0.13) and activated partial thromboplastin time (APTT) (β −1.08, 95% CI −1.86 to 0.31), and the myocardial injury marker cardiac troponin I (cTnI) (β −2085.48, 95% CI −3892.35 to 278.61) in univariate analysis, and with the PT (β −0.36, 95% CI −0.60 to 0.13) in multivariate analysis. The PI was negatively associated with the lactate concentration (β −0.57, 95% CI −0.95 to 0.19), mechanical ventilation time (β −23.11, 95% CI −36.54 to 9.69), and length of ICU stay (β −1.28, 95% CI −2.01 to 0.55). Sensitivity analyses showed that the PI was significantly associated with coagulation markers (PT and APTT) and a myocardial injury marker (cTnI) in patients with sepsis, suggesting that the associations between the PI and organ function were stronger in the sepsis group than the nonsepsis group. Conclusion: The PI provides new insights for assessing the disease severity, short-term prognosis, and organ function damage in ICU patients with sepsis, laying a theoretical foundation for future research.

Keywords

sepsis peripheral perfusion index microcirculation endothelium organ dysfunction

Introduction

The most recent Sepsis-3 International agreement defines sepsis as a potentially fatal organ malfunction brought on by dysregulation of the host's immune response to infection.¹ With the emphasis on organ dysfunction in sepsis under the new definition, patients with sepsis or those at risk of sepsis can be identified earlier and managed more effectively. Throughout the onset and progression of sepsis, the systemic inflammatory response gradually proceeds to multiple organ failure, with microcirculatory dysfunction thought to play a significant part in this process.^2–5 Damage to endothelial cells, coagulation abnormalities, alterations in vascular permeability, and other factors that contribute to microcirculatory dysfunction are all part of the pathogenic process of sepsis.^6–8

The microcirculation is a terminal vascular system that meets the metabolic needs of the tissues by transferring oxygen, nutrients, and metabolic wastes between the systemic circulation and tissues. When the microcirculation is compromised, endothelial cell damage is frequently present. Endothelial cell damage results in ischemic and hypoxic damage, changes in vascular permeability, inadequate nutrition and oxygenation of organs throughout the body, and eventually functional cell damage and even necrosis. Additionally, aberrant coagulation function causes microthromboses that might exacerbate the organ functional damage.^9,10 Monitoring and assessing the microcirculation state of patients with sepsis can therefore reflect the severity of the condition, assess the prognosis, and serve as a foundation for evidence-based medicine-guided therapy. Numerous studies have noted that the prognosis is significantly influenced by the degree and persistence of microcirculation abnormalities.^11–16 Sepsis is a microcirculatory disease, and changes in the microcirculation are seen in the initial stages of hemodynamically stable sepsis.¹⁷ The peripheral microcirculation frequently changes early in the progression of sepsis and returns to normal later during treatment.^18–20 The most important step to safeguard organ function is early detection of microcirculation changes and initiation of treatment.^18,21,22 However, there is a lag in the clinical application of such treatment because it is challenging to accurately assess endothelial cell damage and microcirculatory dysfunction using standard clinical procedures and serum lactate concentrations.

It is now possible to accurately monitor the microcirculation in patients with sepsis as a result of advancements in microcirculation monitoring technology, including the peripheral perfusion index (PI),²³ skin temperature layer,²⁴ mottling score,²⁵ sidestream dark field,²⁶ capillary refill time (CRT),^27,28 laser doppler,²⁹ near-infrared spectroscopy,³⁰ nailfold microcirculation examination,³¹ and other techniques. Among these monitoring techniques, the PI uses near-infrared technology to assess blood perfusion in the fingertips. The PI is a noninvasive quantitative method for measuring peripheral perfusion that is commonly used in clinical settings to evaluate blood flow in peripheral tissues.³² Originating from pulse oximetry, the PI is a widely utilized method. The PI calculates the ratio of pulsatile blood flow in peripheral tissue to nonpulsatile or static blood. Essentially, this index represents the pulse strength at the monitoring site (usually the finger, toe, or earlobe), which is displayed as a percentage. A higher PI value indicates greater blood flow in the monitored area, suggesting a stronger pulse signal and better peripheral perfusion as a result.^33,34

The detection of variations in arterial blood oxygen levels represents blood perfusion in the microcirculation. In clinical practice, it is simple to monitor the PI continuously and in real time.³⁵ The benefits of the PI are that it is easy to obtain and has great sensitivity. Monitoring the PI enables the early detection of microcirculation blood flow problems and the assessment of disease severity. Changes in the PI are highly predictive of the prognosis in critically sick patients and are directly related to organ function.^36,37 Therefore, research into the associations between changes in microcirculatory function, associated monitoring indicators, and the severity and prognosis of sepsis offers theoretical support and useful advice for raising the standard of care for patients with sepsis.

The purpose of this study was to assess the usefulness of the PI in predicting the short-term outcomes of septic patients and nonseptic patients in the intensive care unit (ICU) to provide clinical data support, direct clinical decision making, and lay the groundwork for the application of the PI in risk assessment and in the development of tailored treatment plans. We believe that the examination of the PI will enable clinicians to stratify patients by risk and predict the likelihood of organ dysfunction.

Materials and Methods

Study Design and Research Population

This single-center prospective study was supported by the National High Level Hospital Clinical Research Funding (2022-PUMCH-A-221). The study was conducted from November 1, 2022 to May 31, 2023 at Peking Union Medical College Hospital, a top tertiary hospital in China. Patients who were aged 18 years or older and diagnosed with infection were included. Patients who were younger than 18 years or refused to provide written informed consent were excluded. A total of 208 adult patients admitted to the ICU with infection were finally included in the study. Determining whether patients are classified into the infected group is based on a comprehensive assessment of clinical and laboratory evidence. The primary criterion is the confirmation of the presence of an infection focus, which is usually accomplished through clinical observation and imaging examinations; secondly, microbiological evidence supporting the infection must be available, including but not limited to cultures of blood, urine, cerebrospinal fluid, or other body fluids, immunological antibody and its titer determination, as well as utilizing next-generation sequencing technology; finally, significant changes in inflammation-related laboratory indicators, such as white blood cell count, C-reactive protein, and procalcitonin, are also important criteria for determination. Only when all the above conditions are met, are patients classified into the infected group. Figure 1 shows the study flow chart. We categorized the study population into nonsepsis and sepsis groups based on the Sepsis-3 third International consensus definitions.¹ All participants were followed up for 28 days. All clinical data and analytic methods were preserved. The raw data are available from the corresponding author upon reasonable request. The study protocol was approved by the local ethics committee (ethical approval number: I-22PJ1072).

Data Collection

We obtained the demographic, medication, and laboratory data from the electronic medical records system. Medication data included the medical history, ICU variables, and outcomes. The medical history included the diagnosis of diabetes mellitus (DM), autoimmune disease, or malignancy. The ICU variables included the use of surgical or nonsurgical treatment, PI, mean arterial pressure, lactate concentration, maximum dose of norepinephrine per minute, and Acute Physiology and Chronic Health Evaluation (APACHE) II score. We also collected the diagnosis, especially whether the patients were diagnosed with sepsis. Laboratory data included routine blood tests, liver function indicators (alanine aminotransferase [ALT], Aspartate Aminotransferase [AST], Total Bilirubin; Dbil: Direct Bilirubin [Tbil], and Direct Bilirubin [Dbil]), renal function indicators (creatinine [Cr], blood urea nitrogen [BUN]), coagulation function indicators (prothrombin time [PT], activated partial thromboplastin time [APTT], D-dimer), heart function indicators (cardiac troponin I [cTnI], N-terminal pro-B-type natriuretic peptide [NT-proBNP]), and inflammation biomarkers (procalcitonin, high-sensitivity C-reactive protein). All of the abovementioned variables were obtained on admission to the ICU. QG and HL extracted the data independently and double-checked the extracted data using a standardized collection form. The data were then reviewed by XW and HZ. In particular, the collection of PI data was sourced from independent observations by 2 experienced clinicians, who monitored the PI of both the left and right hands of patients to obtain the optimal values under the prevailing conditions.

Endpoints

The primary endpoints were in-hospital mortality and 28-day mortality, while the secondary endpoints included organ function, lactate concentration, mechanical ventilation time (MV), length of ICU stay, and duration of hospitalization.

Statistical Analysis

The Kruskal Wallis test was used to check for data normality. Continuous variables that were normally distributed are presented as means ± standard deviations. Categorical variables are presented as number (percentage). The χ² test was used to compare categorical variables. The Mann-Whitney test was used to compare continuous variables that conformed to normal distribution. Continuous variables with non-normal distribution were analyzed with the Kolmogorov-Smirnov test. Generalized estimating equations were used to conduct univariant and multivariant analyses. A 2-tailed P-value of <.05 was considered to indicate a statistically significant difference in all analyses. R software, version 4.2.0 (http://www.R-project.org/), was used for all statistical analyses. All figures were created using GraphPad Prism, version 10.0.

Results

The final cohort comprised a total of 208 patients, among which 117 patients were diagnosed with sepsis. The characteristics of the study cohort are summarized in Table 1. There were 110 men. Forty-four patients were diagnosed with DM and 15 had autoimmune disease, while almost half of the study population (n = 98) were diagnosed with tumors. The mean age of the total cohort was 57.4 years, and there was no significant difference in age between the sepsis and nonsepsis groups. The proportion of males was larger in the sepsis group than the nonsepsis group, and most of the nonsurgical patients in the ICU were in the sepsis group. Similarly, most patients without tumors who were admitted to the ICU had sepsis.

Table 1.

Characteristics of the Study Population.

	Total (n = 208)	Nonsepsis (n = 91)	Sepsis (n = 117)	P-value
Gender				<.001
Male	110(52.9%)	29(31.9%)	69(59.0%)
Female	98(47.1%)	62(68.1%)	48(41.0%)
Age	57.4 ± 16.6	55.3 ± 19.0	59.0 ± 14.3	.106
DM				.669
Yes	44(21.2%)	18(19.8%)	26(22.2%)
No	164(78.8%)	73(80.2%)	91(77.8%)
AID				.188
Yes	15(7.2%)	9(9.9%)	6(5.1%)
No	193(92.8%)	82(90.1%)	111(94.9%)
Tumor				<.001
Yes	98(47.1%)	61(67.0%)	37(31.6%)
No	110(52.9%)	30(33.0%)	80(68.4%)
Surgery				<.001
Yes	177(85.1%)	86(94.5%)	91(77.8%)
No	31(14.9%)	5(5.5%)	26(22.2%)

Abbreviations: AID, autoimmune disease; DM, diabetes mellitus.

Table 2 summarizes the ICU variables and outcomes of the study cohort. Both the PI and mean arterial pressure did not significantly differ between the sepsis and nonsepsis groups. However, the lactate concentration, maximum dose of norepinephrine, and APACHE II score were significantly higher in the sepsis group than the nonsepsis group. The in-hospital and 28-day mortality were both significantly higher in the sepsis group than the nonsepsis group. Regarding the secondary outcomes, the sepsis group had a longer mean duration of hospitalization, longer ICU stay, and longer MV than the nonsepsis group.

Table 2.

ICU Variables and Outcomes of the Study Population.

	Total (n = 208)	Nonsepsis (n = 91)	Sepsis (n = 117)	P-value
ICU parameters
PI	1.9 ± 1.0	2.0 ± 1.1	1.8 ± 1.0	.167
MAP	96.6 ± 19.4	97.9 ± 18.0	95.5 ± 20.4	.374
Lac	2.8 ± 3.0	2.1 ± 1.9	3.3 ± 3.5	.003
NE	0.2 ± 0.4	0.1 ± 0.1	0.3 ± 0.5	<.001
APACHE II	14.6 ± 6.4	12.4 ± 4.2	16.4 ± 7.2	<.001
Outcomes
LOS	18.3 ± 13.6	16.3 ± 8.4	19.8 ± 16.4	.042
ICU stay	5.3 ± 6.6	2.6 ± 1.3	7.5 ± 8.1	<.001
MV	48.6 ± 115.7	10.5 ± 22.7	78.2 ± 146.4	<.001
In-hospital Mortality				.004
Yes	10(4.8%)	0	10(8.5%)
No	198(95.2%)	91(100%)	107(91.5%)
28-day mortality				<.001
Yes	27(13.0%)	1(1.1%)	26(22.2%)
No	181(87.0%)	90(98.9%)	91(77.8%)

Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II score; ICU, intensive care unit; Lac, lactic acid; LOS, length of stay; MAP, mean arterial pressure; MV, mechanical ventilation; NE, norepinephrine; PI, peripheral perfusion index .

Figure 2 shows the differences in organ function between the sepsis and nonsepsis groups. Compared with the nonsepsis group, the sepsis group had a prolonged PT and APTT, higher Cr and BUN concentrations, and higher cTnI and NT-proBNP concentrations. Table 3 shows the correlations between the PI and outcomes. In the single-factor model, the PI was a protective factor for mortality. A higher PI was negatively associated with a lower lactate concentration, shorter ICU stay, and shorter MV. These associations remained statistically significant after adjusting for age, sex, DM, autoimmune disease, tumors, surgery, and APACHE II score.

Table 3.

Regression Results for the Associations Between the Peripheral Perfusion Index and Primary Outcomes in Single-Factor and Multifactor Models.

	Single-factor model			Multifactor model
	β/OR	95%CI low	95%CI up	β/OR	95%CI low	95%CI up
Lac	−0.57	−0.95	−0.19	−0.43	−0.79	−0.06
LOS	−1.42	−3.04	0.20	−0.93	−2.69	0.83
ICU stay	−1.28	−2.01	−0.55	−0.87	−1.48	−0.27
MV	−23.11	−36.54	−9.69	−14.72	−25.79	−3.66
Mortality	0.29	0.15	0.55	0.21	0.07	0.59
28-day mortality	0.73	0.46	1.14	0.83	0.52	1.35

β is stated for Lac, LOS, ICU stay, and MV. OR is stated for mortality and 28-day mortality. The multifactor model was adjusted for age, sex, diabetes mellitus, autoimmune disease, tumors, surgery, and Acute Physiology and Chronic Health Evaluation II score. Bold, statistically significant.

Abbreviations: CI, confidence interval; ICU, intensive care unit; Lac, lactate; LOS, length of stay; MV, mechanical ventilation; OR, odds ratio.

The correlations between the PI and organ function are shown in Table 4. The PI was negatively associated with the PT, APTT, and cTnI in the single-factor model. However, in the multifactor model, the PI was only associated with the PT. The results also suggested that the PI tended to be negatively associated with the Cr, BUN, and NT-proBNP. To further explore the relationships between the PI and organ function, we conducted sensitivity analyses. Table 5 and Figure 3 show that the association between the PI and organ function was stronger in the sepsis group than the nonsepsis group. The PI was significantly associated with the PT, APTT, and cTnI in the sepsis group. In contrast, the PI was only significantly associated with the PT in the nonsepsis group.

Table 4.

Regression Results for the Associations Between the Peripheral Perfusion Index and Organ Function in Both Single-Factor and Multifactor Models.

	Single-factor model			Multifactor model
	β	95%CI low	95%CI up	β	95%CI low	95%CI up
PLT	1.76	−9.29	12.82	2.27	−9.40	13.93
PT	−0.36	−0.59	−0.13	−0.36	−0.60	−0.13
APTT	−1.08	−1.86	−0.31	−0.83	−1.68	0.02
D-dimer	0.26	−0.86	1.39	0.60	−0.39	1.60
ALT	7.96	−6.67	22.58	9.00	−7.77	25.78
Cr	−5.29	−16.59	6.01	2.94	−8.07	13.96
BUN	−0.63	−1.41	0.16	−0.33	−0.99	0.34
cTnI	−2085.48	−3892.35	−278.61	−1324.05	−2831.94	183.83
NT-proBNP	−787.48	−1660.13	85.17	−238.86	−998.83	511.12

The multifactor model was adjusted for age, sex, diabetes mellitus, autoimmune disease, tumor, surgery, and Acute Physiology and Chronic Health Evaluation II score. Bold, statistically significant.

Abbreviations: ALT, alanine aminotransferase; APTT, activated partial thromboplastin time; BUN, blood urea nitrogen; CI, confidence interval; Cr, creatinine; cTnI, cardiac troponin I; NT-proBNP, N-terminal pro-B-type natriuretic peptide; PLT, platelet count; PT, prothrombin time.

Table 5.

Sensitivity Analysis of the Associations Between the Peripheral Perfusion Index and Organ Function in the Nonsepsis and Sepsis Groups.

	Nonsepsis			Sepsis
	β	95%CI low	95%CI up	β	95%CI low	95%CI up
PLT	0.74	−13.16	11.68	3.94	−13.63	21.52
PT	−0.22	−0.42	−0.01	−0.49	−0.87	−0.10
APTT	−0.75	−1.74	0.24	−1.37	−2.51	−0.22
D-dimer	0.03	−0.82	0.89	0.46	−1.50	2.43
ALT	3.66	−4.86	12.19	11.68	−14.60	39.97
Cr	−2.63	−11.26	6.00	−7.59	−27.20	12.01
BUN	−0.10	−0.54	0.35	−1.09	−2.47	0.29
cTnI	−2762.57	−6750.71	1225.56	−1675.86	−3293.93	−57.80
NT-proBNP	13.51	−58.35	85.38	−1450.05	−3009.04	108.94

Bold, statistically significant.

Discussion

Peripheral perfusion monitoring aims to detect early indications of tissue hypoperfusion. Previous studies reported significant correlations between PI values and microcirculatory function measures, which were found to be particularly evident in indicators such as skin temperature gradient, CRT, and lactate levels.^23,38,39 Alterations in these indicators are reflective of disturbances in microcirculatory function. Furthermore, the PI can indicate the equilibrium of cardiac output, the balance of the sympathetic and parasympathetic nervous systems, peripheral microcirculation activity, blood flow, vascular tone, and microcirculatory function. This foundational understanding confirms the effective use of the PI for assessing prognosis and stratifying severity in patients. Thus, peripheral perfusion monitoring can be used to categorize the disease severity and more accurately assess the prognosis.^11,36,37,40 The purpose of the present study was to use this noninvasive assessment technique to predict mortality and organ function impairment in critically ill patients. To fully assess the clinical value of the PI, we performed multi-organ function analysis to examine the relationship between the PI and numerous clinical indicators, including inflammatory markers, prognostic scores, coagulation function, liver function, kidney function, and myocardial injury.

Prognosis and Risk Stratification

The present study showed that the in-hospital mortality of patients with infection and sepsis was significantly negatively associated with the PI. Accordingly, and in line with the findings of earlier studies,^36,40,41 a higher PI denotes a better prognosis and a reduced risk of death. This discovery has significant clinical implications. The microcirculatory perfusion, which can indicate the disease severity, can be monitored by assessing the PI. The present findings support the usefulness of the PI in assessing the prognoses of critically ill patients.

Although we found a clear correlation between PI values and in-hospital mortality, the correlation between PI values and 28-day mortality did not reach significance. We believe that this may have occurred because 28-day mortality is influenced by multiple factors that are beyond the scope of measurements during the acute hospitalization period. These factors could include the heterogeneity in underlying health conditions of the patient sample in the study, postdischarge medical care, patient adherence to treatment, and other health-related behaviors not captured by the PI. Additionally, because sepsis is an acutely developing disease, its impact might more directly reflect a shorter-term outcome. Meanwhile, other complications and health issues arising during this process could play a more significant role.

Moreover, this situation could potentially result from the PI being less successful in predicting medium-term and long-term mortality than it is in predicting short-term mortality. In-hospital mortality more closely represents the short-term prognosis, whereas the 28-day mortality more accurately reflects the medium-term prognosis. The PI is used for immediate monitoring and is thus linked to short-term mortality but is less related to medium-term and long-term mortality. However, this also demonstrates that the PI is more sensitive, and that quick assessment is important, underscoring the necessity of dynamic evaluation. Another possible reason for the lack of association between the PI and 28-day mortality may be that the association between the PI and the mortality risk is not linear, possibly due to the existence of a dose-effect relationship, and that the impact on the mortality risk only becomes significant when the PI drops by a certain extent; as 28-day mortality events are relatively rare, the statistical power was insufficient to detect the effect. The third possible reason for the lack of association between the PI and 28-day mortality is that the PI only reflects changes in microcirculatory perfusion, which is not the only determinant of death.

The severity of disease at the time of admission is better determined by the in-hospital mortality, whereas the 28-day mortality is affected by numerous variables after discharge. Therefore, we think that the PI is more suited to assessing the severity of the current illness and the short-term prognosis; however, further research is needed to confirm the predictive value of the PI for 28-day mortality.

Assessment of Coagulation Function

The current study showed that the PI was useful in predicting organ function. The PT and APTT were negatively associated with the PI, and the significance of these associations was especially high in patients with sepsis. This implies that the PI can represent the coagulation function, with a greater PI indicating less impairment of coagulation. This serves as an additional resource for assessing the disease severity. As the endothelium of the blood vessels is damaged by inflammation, ischemia, hypoxia, and other factors when the PI is low, it is possible that this endothelial damage is the cause of the coagulation abnormality. The release of acetylheparin sulfate and the subsequent rise in heparin-like substances in the blood circulation is known as endogenous endothelial heparinization or endogenous heparin-like syndrome, which leads to further coagulation dysfunction.^42,43,44 In addition, endothelial dysfunction and disruption of the glycocalyx barrier in sepsis can result in organ failure, decreased transcapillary exchange, and microvascular thrombosis.⁹ Another important factor that contributes to coagulation abnormalities in patients with sepsis is the ongoing depletion of coagulation factors caused by macrothrombosis.⁴⁵ The PT and APTT may be prolonged because of any of the aforementioned processes. The sensitivity of the PI as an index of the coagulation function was greater in the sepsis group than the nonsepsis group; this indicates that the inflammatory strike was greater in patients with sepsis than in those without sepsis. This inflammatory strike mechanism has a large impact on coagulation disorders in patients with sepsis.

Endothelial damage and microcirculatory dysfunction are closely related to anomalies of the coagulation process in patients with sepsis. Previous research has shown that a number of processes following endothelial injury can result in coagulation malfunction, microthrombosis, and capillary leakage, which can induce systemic multiple organ dysfunction and aggravate the disease.⁴⁶ If these deviations from normal microcirculatory function occur early in the disease, an assessment of the PI will indicate these changes. Early therapeutic intervention may then lessen the disease severity and thus improve the prognosis. However, further research is required to confirm this suggestion.

Monitoring and Evaluation of Organ Function

Particularly in the sepsis group, the PI was negatively associated with the cTnI and NT-proBNP concentrations. These findings suggest that the PI may help with the evaluation of cardiac damage. In hemodynamic analysis, the PI and cardiac output are closely associated, although the autonomic nerve function also affects the PI. A low PI indicates increased cardiac strain, increased oxygen demand, and insufficient oxygen supply, which increases the likelihood of myocardial damage. A previous study revealed that sympathetic nerve excitation has a negative impact on the body, particularly in patients with sepsis.⁴⁷ Furthermore, an overactive sympathetic nervous system may increase the cardiac afterload, which may also contribute to myocardial injury.⁴⁸ A low PI indicates the occurrence of microembolic events, which can cause microvascular obstruction and exacerbate myocardial ischemic injury.⁴⁹

Our study did not show that the PI can reliably predict changes in the ALT, Cr, or BUN concentrations as indicators of liver or renal function. In the sepsis group, the PI and BUN tended to be negatively associated, but without statistical significance. However, the PI was only evaluated once, at the time of ICU admission, which may not accurately reflect the dynamic changes in renal function that occur over the course of the disease. Furthermore, the mechanism of renal injury in sepsis is complicated, involving not just renal tubular destruction, inflammation, and perfusion. The present findings offer a potential starting point for using the PI to assess renal function. A prospective trial that collects data at several timepoints over the course of the disease is warranted to investigate the effectiveness of the PI for monitoring changes in the renal function.

Other Outcome Measures

We discovered significant associations between the PI and the lactate concentration, MV, and ICU stay. In the clinical setting, the lactate concentration is a crucial indicator of perfusion, and the association between the PI and the lactate concentration supports the validity of this relationship and the mechanism by which the PI changes in response to changes in perfusion. The relationships between the PI and the MV and the ICU stay are more sensitive to the immediate outcomes, which helps to improve clinical treatment.

The PI has clinical benefits and may be used to assess the state and short-term prognosis of patients in the ICU, especially those with sepsis. The PI has proven to be a reliable indicator of the short-term prognosis and mortality risk, and a risk assessment index. Although the PI had no association with the 28-day mortality, it still offers useful clinical data, and therapeutic intervention may improve the prognosis. Additionally, the PI has prognostic value in assessing the functioning state of vital organs such the heart, liver, kidney, and blood coagulation in critical care patients. The PI can be used as an additional index for monitoring the functional condition of these organs, despite the fact that the link between the functional impairment of different organs and the PI varies. However, a further study with a large sample size is warranted to confirm the effectiveness of the PI in predicting the long-term prognosis, liver function, and kidney function. It is important to note that the PI cannot predict whether a patient will live or die, but can only predict the disease severity. The PI may be used as an auxiliary index to assess the prognosis, but should not be relied upon exclusively.

Limitations

This study has several limitations. First, there was no dynamic monitoring of the change in the PI over the course of the disease, as the study only assessed the PI once at the time of admission to the ICU. Second, as this was an observational study, the causal relationship between the PI and the prognosis cannot be established. Third, this study did not differentiate between various sepsis subtypes or variations in underlying conditions (eg, surgical and medical septic patients), which may impact the interpretation of the findings. Finally, the usefulness of the PI in nonseptic patients might require further validation in a larger population.

Perspectives

This study offers recommendations for further research and implies that the PI has potential in the assessment of patients with sepsis in the ICU. Microcirculation monitoring has great potential in the field of sepsis research and plays a crucial role in the pathophysiology of the condition. When seen macroscopically, the microcirculation directly connects the macrocirculation to tissue perfusion. To convey the flow of blood and oxygen to tissues, the hemodynamics of the macrocirculation must first pass via the microcirculation. Microcirculation also has a similar role in the microenvironment to endothelial damage and mitochondrial cellular damage. Because endothelial dysfunction impairs the microcirculation, it also causes mitochondrial oxidative stress damage. The central role of microcirculation in the development of sepsis has been acknowledged theoretically. New monitoring techniques like the PI are continually advancing due to ongoing advancements in peripheral perfusion monitoring technologies. In the future, targeted therapy to improve the microcirculation in sepsis is likely to be developed.

In the future, we plan to perform a prospective study to collect dynamic PI data throughout the course of the disease and use a multicenter design to increase the sample size and strengthen the conclusions. To investigate personalized PI monitoring measures, we will also perform subgroup analyses. A thorough evaluation system may be established by combining the PI with other variables. Finally, the inclusion of PI monitoring in clinical research may demonstrate how the PI affects the prognosis of patients with sepsis.

Conclusion

The PI is a useful tool for assessing the disease severity and prognosis of patients with sepsis in the ICU. The mortality rate and the PI are inversely associated. Additionally, the coagulation function and the degree of cardiac injury tended to have a negative association with the PI, indicating that the PI may reflect functional damage to critical organs during the course of the disease and offer a new way to assess organ failure. Additionally, we found that the MV increased as the PI decreased, suggesting that PI monitoring helps predict the response to treatment.

In conclusion, the PI is a novel index for the noninvasive assessment of disease severity and other clinical prognostic indicators of patients with sepsis in the ICU, with its significance in predicting organ function validated in this study. This research provides evidence and theoretical support for the future clinical application of PI in guiding sepsis treatment. Furthermore, it lays the foundation for further exploration of the relationship between sepsis-induced organ dysfunction and peripheral microcirculation.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the National High Level Hospital Clinical Research Funding (2022-PUMCH-A-221) and the Wu Jieping Medical Foundation (320.6750.2022-02-13).

ORCID iDs

Qirui Guo

Guangjian Wang

References

Singer

Deutschman

Seymour

, et al. The third International consensus definitions for sepsis and septic shock (sepsis-3). JAMA. 2016;315(8):801. doi:https://doi.org/10.1001/jama.2016.0287

Kattan

Ibarra-Estrada

Ospina-Tascón

Hernández

. Perspectives on peripheral perfusion assessment. Curr Opin Crit Care. 2023;29(3):208-214. doi:https://doi.org/10.1097/MCC.0000000000001038

De Backer

Ricottilli

Ospina-Tascón

. Septic shock: a microcirculation disease. Curr Opin Anaesthesiol. 2021;34(2):85-91. doi:https://doi.org/10.1097/ACO.0000000000000957

Hernandez

Bruhn

Ince

. Microcirculation in Sepsis: New Perspectives.

Yajnik

Maarouf

. Sepsis and the microcirculation: the impact on outcomes. Curr Opin Anaesthesiol. 2022;35(2):230-235. doi:https://doi.org/10.1097/ACO.0000000000001098

Tarbell

Cancel

. The glycocalyx and its significance in human medicine. J Intern Med. 2016;280(1):97-113. doi:https://doi.org/10.1111/joim.12465

Levi

Van Der Poll

. Endothelial injury in sepsis. Intensive Care Med. 2013;39(10):1839-1842. doi:https://doi.org/10.1007/s00134-013-3054-1

Cohen

Vincent

Adhikari

NKJ

, et al. Sepsis: a roadmap for future research. Lancet Infect Dis. 2015;15(5):581-614. doi:https://doi.org/10.1016/S1473-3099(15)70112-X

Lupu

Kinasewitz

Dormer

. The role of endothelial shear stress on haemodynamics, inflammation, coagulation and glycocalyx during sepsis. J Cell Mol Med. 2020;24(21):12258-12271. doi:https://doi.org/10.1111/jcmm.15895

10.

Walborn

Rondina

Mosier

Fareed

Hoppensteadt

. Endothelial dysfunction is associated with mortality and severity of coagulopathy in patients with sepsis and disseminated intravascular coagulation. Clin Appl Thromb. 2019;25:107602961985216. doi:https://doi.org/10.1177/1076029619852163

11.

Sakr

Dubois

De Backer

Creteur

Vincent

. Persistent microcirculatory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med. 2004;32(9):1825-1831. doi:https://doi.org/10.1097/01.CCM.0000138558.16257.3F

12.

Mouncey

Osborn

Power

, et al. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med. 2015;372(14):1301-1311. doi:https://doi.org/10.1056/NEJMoa1500896

13.

De Backer

Donadello

Sakr

, et al. Microcirculatory alterations in patients with severe sepsis: impact of time of assessment and relationship with outcome. Crit Care Med. 2013;41(3):791-799. doi:https://doi.org/10.1097/CCM.0b013e3182742e8b

14.

Hernández

Ospina-Tascón

Damiani

, et al. Effect of a resuscitation strategy targeting peripheral perfusion status vs serum lactate levels on 28-day mortality among patients with septic shock: the ANDROMEDA-SHOCK randomized clinical trial. JAMA. 2019;321(7):654. doi:https://doi.org/10.1001/jama.2019.0071

15.

Zampieri

Damiani

Bakker

, et al. Effects of a resuscitation strategy targeting peripheral perfusion status versus serum lactate levels among patients with septic shock. A Bayesian reanalysis of the ANDROMEDA-SHOCK trial. Am J Respir Crit Care Med. 2020;201(4):423-429. doi:https://doi.org/10.1164/rccm.201905-0968OC

16.

Kattan

Ospina-Tascón

Teboulet

, et al. Systematic assessment of fluid responsiveness during early septic shock resuscitation: secondary analysis of the ANDROMEDA-SHOCK trial. Crit Care. 2020;24(1):23. doi:https://doi.org/10.1186/s13054-020-2732-y

17.

Lam

Tyml

Martin

Sibbald

. Microvascular perfusion is impaired in a rat model of normotensive sepsis. J Clin Invest. 1994;94(5):2077-2083. doi:https://doi.org/10.1172/JCI117562

18.

Trzeciak

Dellinger

Parrillo

, et al. Early microcirculatory perfusion derangements in patients with severe sepsis and septic shock: relationship to hemodynamics, oxygen transport, and survival. Ann Emerg Med. 2007;49(1):88-98.e2. doi:https://doi.org/10.1016/j.annemergmed.2006.08.021

19.

Lima

Bakker

. Clinical assessment of peripheral circulation. Curr Opin Crit Care. 2015;21(3):226-231. doi:https://doi.org/10.1097/MCC.0000000000000194

20.

Hasanin

Mukhtar

Nassar

. Perfusion indices revisited. J Intensive Care. 2017;5(1):24. doi:https://doi.org/10.1186/s40560-017-0220-5

21.

De Backer

Durand

. Monitoring the microcirculation in critically ill patients. Best Pract Res Clin Anaesthesiol. 2014;28(4):441-451. doi:https://doi.org/10.1016/j.bpa.2014.09.005.

22.

Ince

De Backer

Mayeux

. Microvascular dysfunction in the critically ill. Crit Care Clin. 2020;36(2):323-331. doi:https://doi.org/10.1016/j.ccc.2019.11.003

23.

Pinto Lima

Beelen

Bakker

. Use of a peripheral perfusion index derived from the pulse oximetry signal as a noninvasive indicator of perfusion. Crit Care Med. 2002;30(6):1210-1213. doi:https://doi.org/10.1097/00003246-200206000-00006.

24.

Bourcier

Pichereau

Boelle

, et al. Toe-to-room temperature gradient correlates with tissue perfusion and predicts outcome in selected critically ill patients with severe infections. Ann Intensive Care. 2016;6(1):63. doi:https://doi.org/10.1186/s13613-016-0164-2

25.

Ait-Oufella

Bourcier

Alves

, et al. Alteration of skin perfusion in mottling area during septic shock. Ann Intensive Care. 2013;3(1):31. doi:https://doi.org/10.1186/2110-5820-3-31

26.

Boerma

Koopmans

Konijn

, et al. Effects of nitroglycerin on sublingual microcirculatory blood flow in patients with severe sepsis/septic shock after a strict resuscitation protocol: a double-blind randomized placebo controlled trial. Crit Care Med. 2010;38(1):93-100. doi:https://doi.org/10.1097/CCM.0b013e3181b02fc1

27.

Castro

Kattan

Ferri

, et al. Effects of capillary refill time-vs. lactate-targeted fluid resuscitation on regional, microcirculatory and hypoxia-related perfusion parameters in septic shock: a randomized controlled trial. Ann Intensive Care. 2020;10(1):150. doi:https://doi.org/10.1186/s13613-020-00767-4

28.

Dubée

Hariri

Joffre

, et al. Peripheral tissue hypoperfusion predicts post intubation hemodynamic instability. Ann Intensive Care. 2022;12(1):68. doi:https://doi.org/10.1186/s13613-022-01043-3

29.

Guven

Dijkstra

Kuijper

, et al. Comparison of laser speckle contrast imaging with laser Doppler perfusion imaging for tissue perfusion measurement. Microcirculation. 2023;30(1). doi:https://doi.org/10.1111/micc.12795

30.

Skarda

Mulier

Myers

Taylor

Beilman

. Dynamic near-infrared spectroscopy measurements in patients with severe sepsis. Shock. 2007;27(4):348-353. doi:https://doi.org/10.1097/01.shk.0000239779.25775.e4

31.

El Miedany

Ismail

Wadie

Hassan

. Nailfold capillaroscopy: tips and challenges. Clin Rheumatol. 2022;41(12):3629-3640. doi:https://doi.org/10.1007/s10067-022-06354-1

32.

Grigorasi

Corlade-Andrei

Ciumanghel

, et al. Monitoring Perfusion Index in patients presenting to the emergency department due to drug use. J Pers Med. 2023;13(2):372. doi:https://doi.org/10.3390/jpm13020372

33.

Spiliopoulos

Georgiadou

Karahaliou

, et al. Feasibility and clinical value of intraprocedural Doppler ultrasonography blood flow parameters during peripheral endovascular procedures for limb salvage: a pilot study. J Endovasc Ther. 2023;8:152660282311798. Published online June. doi:https://doi.org/10.1177/15266028231179838.

34.

Strauss

Fodor

Anker

, et al. The perfusion index as a noninvasive method for monitoring revascularized and replanted digits. Clin Hemorheol Microcirc. 2024;86(1-2):143-152. doi:https://doi.org/10.3233/CH-238120

35.

Coutrot

Dudoignon

Joachim

Gayat

Vallée

Dépret

. Perfusion index: physical principles, physiological meanings and clinical implications in anaesthesia and critical care. Anaesth Crit Care Pain Med. 2021;40(6):100964. doi:https://doi.org/10.1016/j.accpm.2021.100964

36.

Liu

Long

Wang

. The peripheral perfusion index and transcutaneous oxygen challenge test are predictive of mortality in septic patients after resuscitation. Crit Care. 2013;17(3):R116. doi:https://doi.org/10.1186/cc12788

37.

Mongkolpun

Orbegozo

Cordeiro

CPR

Franco

CJCS

Vincent

Creteur

. Alterations in skin blood flow at the fingertip are related to mortality in patients with circulatory shock. Crit Care Med. 2020;48(4):443-450. doi:https://doi.org/10.1097/CCM.0000000000004177

38.

Singla

Mahajan

. An observational study to correlate arterial lactate level and peripheral perfusion index in context of tissue perfusion in sepsis. Int J Res Med Sci. 2023;11(5):1665-1670. doi:https://doi.org/10.18203/2320-6012.ijrms20231333.

39.

Iizuka

Yoshinaga

Nakatomi

Takahashi

Yoshida

Sanui

. A low peripheral perfusion index can accurately detect prolonged capillary refill time during general anesthesia: a prospective observational study. Saudi J Anaesth. 2023;17(1):33. doi:https://doi.org/10.4103/sja.sja_634_22

40.

Agerskov

Thusholdt

ANW

Holm-Sørensen

, et al. Association of the intraoperative peripheral perfusion index with postoperative morbidity and mortality in acute surgical patients: a retrospective observational multicentre cohort study. Br J Anaesth. 2021;127(3):396-404. doi:https://doi.org/10.1016/j.bja.2021.06.004.

41.

Kuzhikkombil Mani

Ghosh

. Perfusion Index variations in children with septic shock: Single-Center Observational Cohort Study in India. Pediatr Crit Care Med. 2023;25(1):47-53(1):47–53. doi:https://doi.org/10.1097/PCC.0000000000003348

42.

Johansson

Henriksen

Stensballe

, et al. Traumatic endotheliopathy: a prospective observational study of 424 severely injured patients. Ann Surg. 2017;265(3):597-603. doi:https://doi.org/10.1097/SLA.0000000000001751

43.

Premkumar

Bihari

Saxena

, et al. Heparin-like effect associated with risk of bleeding, sepsis, and death in patients with severe alcohol-associated hepatitis. Clin Gastroenterol Hepatol. 2020;18(2):486-495.e3. doi:https://doi.org/10.1016/j.cgh.2019.04.057

44.

Hadid

Kafri

Al-Katib

. Coagulation and anticoagulation in COVID-19. Blood Rev. 2021;47:100761. doi:https://doi.org/10.1016/j.blre.2020.100761

45.

Schouten

Wiersinga

Levi

Van Der Poll

. Inflammation, endothelium, and coagulation in sepsis. J Leukoc Biol. 2008;83(3):536-545. doi:https://doi.org/10.1189/jlb.0607373

46.

Gigante

Levy

Van Gorp

, et al. Management of patients on antithrombotic therapy with severe infections: a joint clinical consensus statement of the ESC Working Group on thrombosis, the ESC Working Group on atherosclerosis and vascular biology, and the International Society on thrombosis and haemostasis. Eur Heart J. 2023;44(32):3040-3058. doi:https://doi.org/10.1093/eurheartj/ehad388

47.

Wang

Zhang

Liu

Zhang

. Effect of Norepinephrine on Peripheral Perfusion Index and its association with the prognosis of patients with sepsis. J Intensive Care Med. 2024;39(1):21–27. doi:https://doi.org/10.1177/08850666231187333

48.

Huang

Liu

Zhang

Ding

Wang

. On behalf of the Critical Care Ultrasound Study Group (CCUSG). Focus on host/organ unregulated response: a common cause of critical illness. Chin Med J (Engl). 2023;136(1):108-110. doi:https://doi.org/10.1097/CM9.0000000000002374

49.

Nakatani

Ohta-Ogo

Nishio

, et al. Microthrombosis as a cause of fulminant myocarditis-like presentation with COVID-19 proven by endomyocardial biopsy. Cardiovasc Pathol. 2022;60:107435. doi:https://doi.org/10.1016/j.carpath.2022.107435