Can peripheral perfusion index predict disease mortality in COVID-19 patients in the emergency department

Background Coronavirus disease 2019 (COVID-19) pneumonia remains a major public health concern. The prognostic efficacy of Peripheral Perfusion Index (PPI) has been researched in different pathologies such as trauma and sepsis. We hypothesized that PPI may serve as predictor of mortality in hospitalized patients with COVID-19 infection. This study aimed to describe the association between PPI at admission and COVID-19 mortality, a new mortality prediction tool. Methods This retrospective, observational study was conducted at a tertiary care center in Turkey. Adult patients diagnosed with COVID-19 infection were enrolled in this study between Februrary 15, 2022 to April 15, 2023. Patient demographic and clinical data including vital signs, laboratory parameters and PPI on admission were collected from an electronic database. PPI was measured using Philips G30E patient monitor system. The primary outcome was in-hospital mortality. Results In total, 200 patients with COVID-19 infection were included and 42 (21 %) in-hospital deaths were identified. For all parameters of study, age, oxygen saturation, respiratory rate, PPI, urea, creatinine, White Blood Cell (WBC), and High-sensitive cardiac Troponin T (hs-cTnT) values were significantly different between survivors vs non-survivors. hs-cTnT >21,25 pg/mL[HR:2.823 (95 % CI:1.211–6583)], PPI <2,15 [HR:2485 (95 % CI:1.194–5.175)], Oxygen saturation <87 % [HR:2258 (95 % CI:1.191–4.282)], and WBC >9680 x103/ml [HR:2.124 (95 % CI:1.083–4.163)] were independent predictors of in-hospital mortality. Conclusions This study identified the factors affecting in-hospital mortality among COVID-19 patients. Importantly, besides many parameter, PPI at admission was significantly associated with COVID-19 mortality and could be a feasible marker in emergency department to identify high risk patients.


Introduction
Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is associated with substantial morbidity and mortality.As of November 19, 2023, over 772 million confirmed cases and over six million deaths have been reported globally [1].Even though the World Health Organization has declared the pandemic to be over, the infection still persists, with reported increases in case and Intensive Care Unit (ICU) admission numbers, particularly in America and Europe [1].
Predicting undesirable outcomes is therefore crucial, even on admission to an Emergency Department (ED).In this context, urgent risk-stratification in the ED can lead to significantly reduced risks and better patient outcomes.
The extreme inflammatory response that occurs in severe cases of COVID-19, also known as a cytokine storm, causes worst outcomes such as sepsis, shock, or even death [2].In this context, it is vital to detect this inflammatory response, which occurs in severe COVID-19 cases, at an early stage.For this purpose, many risk-stratification tools use clinical, physiological, and radiologic parameters have been investigated for prognostic effectiveness in COVID-19 cases [3,4].Although risk stratification tools such as Shock Index, Diastolic Shock Index, Age Shock Index, Modified Shock Index are widely accepted and used markers for mortality, there is a need for new prognostic markers [5].
In this respect, Peripheral Perfusion Index (PPI) has emerged as a promising prognostic marker, particularly in septic patients, due to its characteristics of demonstrating changes in peripheral perfusion in the early stages, being noninvasive, objective, and reproducible [6].The PPI is a valuable marker for assessing peripheral perfusion.PPI measures the ratio of the pulsatile to non-pulsatile components in peripheral blood flow and is primarily influenced by two key factors: cardiac output and the balance between the sympathetic and parasympathetic nervous systems.PPI tends to decrease in conditions of sympathetic dominance and/or reduced cardiac output, making it a reliable predictor of patient outcomes [7].It has been suggested that PPI could be used almost as a vital sign even in the pre-COVID-19 era [8].However, there is limited data in the literature regarding the use of such a valuable parameter in COVID-19 cases [9,10].
This study aimed to investigate the role of the PPI value assessed in the ED in predicting mortality in hospitalized COVID-19 cases.

Study design and settings
This retrospective, single-center, cross sectional, and observational study performed at the ED of university-affiliated training and research hospital in Mugla, Turkey.The hospital, comprising over 500 beds, registers an annual influx of approximately 100,000 adult visits to the ED.Ethical approval for the study was obtained from the local ethics committee (Reference Number: 230007-43).This study was conducted according to the Principles of the Declaration of Helsinki of 1975, as revised in 1983 and the Strengthening the Reporting of Observational Studies in Epidemiology (http://www.strobe-statement.org)guidelines.The need to obtain written consent informs was waived because of the retrospective nature of the study.

Patients selection
From Feb 15, 2022 to Apr 15, 2023, all consecutive adult patients (aged ≥18 years old) admitted to ED and hospitalized with laboratory-confirmed COVID-19 infection based on positive Reverse-Transcriptase-Polymerase Chain-Reaction (RT-PCR) assay via oropharyngeal or nasopharyngeal swabs, were enrolled.However, certain groups were excluded to maintain the study's integrity.Patients, who had a history of peripheral artery disease, who had sought hospital discharge against medical advice, who had missing data in medical records and who had been referred to other hospitals were excluded.In addition, patients requiring acute interventional procedures and/or with specific diagnoses (acute myocardial infarction, acute aortic syndrome, pulmonary embolism, acute appendicitis, etc.) were excluded from the study, because they were not admitted to the hospital due to COVID-19 infection, even if they were incidentally found to have a positive COVID-19 test.

Data collection
Clinical data included demographic characteristics (age and gender), vital signs (systolic and diastolic blood pressure, heart rate, respiratory rate, fever, and oxygen saturation), PPI, and laboratory parameters at ED admission including complete blood count, renal and liver function tests, serum electrolytes, and High-Sensitive Cardiac Troponin T (hs-cTnT) levels, were extracted from medical records and recorded.
All patients admitted to ED with COVID-19 infection symptoms were evaluated in the COVID-19 area of ED and a trained triage nurse (always available in ED), who was unaware of the study, took vital sign and PPI measurements.The PPI is a new parameter developed for hemodynamic monitoring, derived from the Photoplethysmography (PPG) method, which is traditionally used for pulse oximetry monitoring.PPI is derived from PPG signal and represents the ratio of pulsatile on non-pulsatile light absorbance or reflectance of the PPG signal.PPI determinants are complex and interlinked, involving and reflecting the interaction between peripheral and central haemodynamic characteristics, such as vascular tone and stroke volume.Philips G30E monitor system automatically calculates PPI via pulse oximetry probe and displays it on the screen, thanks to the software it contains.

Outcome measures
The outcome of present study is to determine the prediction power of PPI on in-hospital mortality among hospitalized COVID-19 patients.
M.G.Kaya et al.

Statistical analysis
IBM SPSS v.25.0 package program (SPSS Inc., Chicago, IL) was used to analyze the recorded data.Normality analyses of the data were conducted using the Kolmogorov-Smirnov test.Because of non-normally distribution, all quantitative data were presented as median (minimum-maximum).The categorical variables were expressed as the number of cases (n) with percentages (%).The differences between the groups were investigated using the Mann-Whitney U test.Intragroup comparisons of the categorical variables were made using the chi square test and the Fisher's exact test.Receiver operating characteristic (ROC) analysis was performed to determine the in-hospital mortality predictive power of the PPI and other significantly differiantiated parameters.The optimum cut-off levels were determined using Youden's index (sensitivity + specificity ─ 1).The sensitivity, specificity, and positive and negative predictive values were calculated for the optimum cut-off levels.Univariate and Multivariate Cox proportional hazard models were fitted to study the association between the distributions of risk factors among survival.The results of the analysis were presented in terms of the estimated hazard ratios (HR) with 95 % confidence intervals (95 % CI).Statistical significance was set at p < 0.05.

Results
A total of 263 patients who were admitted to ED and hospitalized with RT-PCR confirmed COVID-19 infection from Feb 15, 2022 to Apr 15, 2023 were identified.Of these 263 patients, 63 excluded.The remaining 200 patients were included (Fig. 1).
For all parameters of study, age, oxygen saturation, respiratory rate, PPI, urea, creatinine, White Blood Cell (WBC), and hs-cTnT values were significantly different between survivors vs non-survivors.We performed ROC analysis to determine the in-hospital mortality predictive power of significantly different parameters mentioned before.The highest AUC value was belong to hs-cTnT, while the lowest was belong to age.Table 2 and Fig. 2 presents the results of ROC analysis in detail.
Univariate Cox regression analysis, showed that oxygen saturation, respiratory rate, PPI, urea, creatinine, WBC, and hs-cTnT were significant factors for in-hospital mortality (Table 3).Consequently, independent prognostic factors associated with in-hospital mortality were figured out by multivariate Cox regression analysis.3.

Discussion
Our study focused on the utility of PPI as a predictive parameter of mortality in patients hospitalized with COVID-19 in ED.We found that lower PPI values is associated with increased risk of mortality.A cutoff PPI value of 2,15 predicted mortality with a sensitivity of 73,8 % and a specificity of 63,3 %.The PPI, as a noninvasive, easily measurable, subjective/quantitative parameter of peripheral perfusion, has previously been investigated as a prognostic marker in the evaluation of many diseases in the ED [7].
Although it has been demonstrated to be valuable in predicting both the severity of the disease and cardiovascular complications in COVID-19 patients, the majority of these studies have been conducted in intensive care and/or in patient units.To the best of our knowledge, this is the first study that demonstrated the successful utilization of PPI in predicting mortality among COVID-19 patients in the ED.
Considering the critical need for early identification, prompt intervention, and accurate risk assessment, it becomes paramount for emergency physicians to distinguish patients at high mortality risk among those severely afflicted with COVID-19.Hence, it's not unexpected to witness a surge in research dedicated to developing straightforward and convenient prediction tools.These tools aim to aid emergency physicians in swiftly and accurately identifying patients at elevated risk of mortality.For this reason, it is not surprising to see a growing number of studies focusing on simple and convenient prediction tools that can help emergency care physicians to more rapidly and effectively recognize patients at high risk of mortality.In a recent study from Turkey published by Korkut et al.PPI is an easy to apply and useful parameter in the ED in determening the severity of COVID-19 patients [11].Another study conducted by Akdur et al., in our country, which has methodological similarities to our study, demonstrated that low PPI (<1.5) values are associated with mortality in COVID-19 patients.Although the patient populations in both studies are similar, there are differences in the study parameters and outcomes.While Akdur and colleagues examined parameters such as NEWS score, Shock Index, and Computed Tomography-Severity Score in addition to PPI, our study focused specifically on PPI.Additionally, while their study examines 14-and 90-day mortality as outcomes, our study focuses on in-hospital mortality [4].Although the investigated time-to-death and the resulting cut-off values may differ, the findings of our study are consistent with the literature.A decreased PPI value (<2,15) demonstrated a significant association with in-hospital mortality (HR 2.485; 95 % CI 1.194-5.175;p: 0.015).In the prognostic evaluation of COVID-19 patients in ED, PPI appears to be as effective as many known parameters such as age, respiratory rate, oxygen saturation, kidney function tests and hs-cTnT.
Our study focused on the impact of PPI on mortality; however, vital signs, laboratory parameters, and demographic data are also part of our research.Recent findings have underscored the significance of several biochemical tests, either independently or in combination, in predicting COVID-19-related mortality.In a meta-regression analysis published by Zhang JJY et al., observed that higher leukocyte counts, elevated levels of ALT, AST, lactate dehydrogenase (LDH) and raised procalcitonin levels were note-worthy predictors of admission to intensive care unit.Further, the researchers found that elevated LDH and high leukocyte counts were significantly associated with COVID-19 led mortality [12].In a comprehensive meta-analysis conducted by Izcovich et al., similar to our study, 49 parameters such as increasing age, elevated myocardial injury markers, increased WBC, rising plasma creatinine, and blood urea nitrogen levels were identified as high/moderate predictors of poor prognosis [13a].In parallel to these results in our study, elevated hs-cTnT levels, increased WBC counts, and raised plasma creatinine and urea levels were significantly different in deceased patients.Furthermore, a higher hs-cTnT and WBC values demonstrated a significant association with in-hospital mortality (HR 2.823, 95 % CI 1.211-6.583;HR 2.124, 95 % CI 1.083-4.163;respectively).In previous studies, vital signs were reported to be associated with mortality in patients with COVID-19.A clinical study from US reported that heart rate (higher), respiratory rate (higher), and oxygen saturation (lower) were associated with COVID-19 mortality [14].Another study from Turkey also demonsrated that low levels of oxygen saturation was associated with 30-day mortality (OR 0,69; p:0.052) [15].Similarly to the literature, our study also found that low levels of oxygen saturation were associated with in-hospital mortality.(HR 2.258, 95 % CI 1.191-4.282,p:0.013).
Our study has some limitations.First, this was a singlecenter, retrospective study with a small sample size, and therefore, the findings of the study cannot be generalized.Second, we only focused on patients with hospitalized COVID-19 patients; COVID-19 patients discharged from ED were not analyzed.Third, we did not assess medications that affect vital signs such as beta-blockers, and arrhythmias and co-morbidities that affect mortality were not identified.Fourth, vital signs were measured at a single time point, which may not accurately reflect disease dynamics.Finally, before its application in clinical practice, the PPI should undergo external validation in a large-scale studies.

Conclusions
This study demonstrated that the initial PPI is a significant prognostic factor in patients with COVID-19 infection.In clinical settings it may contribute to patient triage as much succesfully as traditional vital signs and laboratory parameters.PPI can be easily used and can be applied for risk stratification of in hospital mortality of in COVID-19 patients.

Fig. 2 .
Fig. 2. ROC Analysis Showing the Relationship Between Survivors vs Non-survivors.Receiver operating characteristic curves for the hs-cTnT, Urea, Creatine, Respiratory Rate, WBC, and Age (A); the Peripheral Perfusion Index, and Oxygen Saturation (B) predicting in-hospital mortality in COVID-19 patients.* hs-cTnT: High-sensitivity cardiac troponin T test, WBC: White Blood Cell.

Table 1
The summary of demographic, laboratory and clinical data of patients.

Table 2
Diagnostic Effectiveness of Parameters in the Differentiation of Between Survivors vs Non-survivors.