Association between ROX index with 28-day mortality in sepsis: A retrospective cohort study

doi:10.21203/rs.3.rs-2634448/v1

Download PDF

Research Article

Association between ROX index with 28-day mortality in sepsis: A retrospective cohort study

https://doi.org/10.21203/rs.3.rs-2634448/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: The ROX index (peripheral oxygen saturation / fraction of inspired oxygen to respiratory rate) can help clinicians assessing patients' conditions. The purpose of this study was to evaluate the effects of time-weighted average ROX (ROX-TWA) index on short-term mortality in sepsis patients.

Methods: In this retrospective study, clinical data of patients diagnosed as sepsis was extracted from the Medical Information Mart for Intensive Care (MIMIC)-III database (version 1.4). The septic patients were further categorized by quartiles of their ROX-TWA levels. The associations between ROX-TWA levels and 28-day mortality were assessed using multivariable Cox proportional hazards model and verified by two-piecewise linear regression model.

Main results: By smooth curve fitting, we find that a non-linear relationship between ROX-TWA index and 28-day mortality in sepsis patients, and inflection point was 8. The risk of developing mortality decreased significantly with an increase in ROX-TWA value until plateaued at a level of 8 (HR 0.75 (0.71, 0.80), P<0.001). Moreover. Subgroup analysis showed, a similar relationship between ROX-TWA index and 28-day mortality in patients using or not using mechanical ventilator.

Conclusion: The relationship between ROX-TWA index and 28-day mortality in sepsis patients is non-linear. ROX-TWA index is an independent risk factor of 28-day mortality in sepsis patients when ROX-TWA is less than 8.

sepsis

ROX index

mortality

MIMIC III

Intensive care

Sepsis is defined as dysregulation of the body to an uncontrolled infection that often leads to life-threatening organ dysfunction(1). It has attracted widespread attention because of its high mortality and incidence rates among hospitalized patients in ICU(2). It has been well accepted in the field that early identification of patients with increasing risks to develop life-threating septic and provide more aggressive interventions can improve the poor prognosis(3). Therefore, how to identify these patients has become increasingly important and multiple scoring systems have been developed including Sequential Organ Failure Assessment (SOFA)(1), the Acute Physiology and Chronic Health Evaluation (APACHE) II or IV(4, 5) and Simplified Acute Physiology Score (SAPS) II or III(6, 7). These evaluations have been extensively used in clinical practice. However, these scoring systems are relatively complex and time consuming. Considering the rapid changing situation of sepsis, a more instant evaluation is needed to better guide the treatment.

The ROX index is calculated based on the ratio of oxygen saturation by the pulse oximetry/fraction of inspired oxygen (SpO2/FiO2) to respiratory rate (RR). Previous study suggested that the increasing ROX index was associated with high flow nasal cannula failure/need for intubation(8). On this basis, Lee et al. study found that the ROX index was an independent prognostic factor for mortality in sepsis(9). However, it is common for patients with sepsis to require invasive ventilation during ICU(10). Ventilators mainly affect gas exchange between the lungs and the outside world, but gas exchange between the pulmonary alveoli and the blood and between the blood and tissue cells is affected by the severity of the disease. Multiple organ dysfunction exacerbates respiratory dysfunction in patients with sepsis, which is clinically manifested by reduced SpO2 and increased respiratory rate and FiO2(11, 12). Patients with prolonged exposure to respiratory dysfunction are at increased risk of death, these pathological changes may explain the correlation between ROX index and mortality in mechanically ventilated patients. Therefore, the ROX index could similarly identify high-risk patients with sepsis in ICU. And we should pay particular attention to the first 24 hours in the ICU, as this is a critical period for patients with sepsis.

In this study, considering that ROX can be time-varying, we chose the time-weighted average (TWA) within first 24 hours in the ICU as the way to evaluate ROX index during ICU management. We aimed to describe the association between the time-weighted average ROX (ROX-TWA) index with short-term mortality in sepsis patients in a relatively large cohort.

Data Source and Participants

The data adopted in this study were obtained from the Medical Information Mart for Intensive Care III (MIMIC III) which is an open access database includes over 50,000 different patients treated at the intensive care unit (ICU) of Beth Israel Deaconess Medical Center (BIDMC) in Boston between 2001 and 2012. After passing required course on the website of National Institutes of Health (NIH) and obtained required certificate (Record ID: 10394271), we accessed the database and obtained the required data for this study. Considering the retrospective nature of this study, informed consent was waived for this study.

The inclusion criteria mainly included:(1) patients with a diagnosis of sepsis defined by the Sepsis-3 definition(The infected or suspected infected population was determined according to the International Classification of Diseases 9th Edition (ICD-9) of MIMIC-III, then patients with an Sequential Organ Failure Assessment (SOFA) score of ≥ 2 points was selected from the population. Infections is defined as follows: a positive microbiological culture, and ICD-9 diagnostic codes including terms “sepsis” or “septic”, “bacteremia”, “infection”, “meningitis”, “peritonitis”, “pneumonia”) (2) If there were multiple admission to ICU from the same patient, only the first admission was collected. (3) aged ≥ 18 years and ICU stays more than 24h. (4) Patients with accurate respiratory rate, SpO2, and FiO2 within 24 hours of each patient' initial admission to ICU. (Fig. 1).

Data Extraction

Demographic parameter measured upon ICU admission including age and gender were extracted from the database. Sequential Organ Failure Assessment score (SOFA) and Simplified Acute Physiology Score (SAPS) calculated upon ICU admission were obtained along with the histories of receiving vasopressors, mechanical ventilator (MV) and renal replacement therapy (RRT) within 24 hours. Comorbidities such as malignancy, pneumonia, respiratory failure, chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), renal failure, and liver disease were also collected. Laboratory results such as arterial serum parameters, blood gas analysis, and complete blood test were documented. Furthermore, physical examinations such as body temperature, blood pressure, heart rate, respiratory rate, SpO2, FiO2 were collected (Table 1).

Table 1: Baseline characteristics of participants

Variables	ROX-TWA quartiles				P-value
Variables	Q1(<8.80) (n=1777)	Q2(8.80-9.80) (n=1871)	Q3(9.90-11.80) (n=1790)	Q4(>11.80) (n=1882)	P-value
	Q1(<8.80) (n=1777)	Q2(8.80-9.80) (n=1871)	Q3(9.90-11.80) (n=1790)	Q4(>11.80) (n=1882)
Age (years)	62.42 (16.38)	64.57 (15.57)	65.27 (15.47)	64.23 (15.44)	<0.001
Gender, n (%)					0.001
Male	1061 (59.71%)	1099 (58.74%)	1062 (59.33%)	1012 (53.77%)
Female	716 (40.29%)	772 (41.26%)	728 (40.67%)	870 (46.23%)
SOFA	7.44(3.66)	6.25 (3.01)	5.77 (2.89)	5.19 (2.73)	<0.001
SAPS II	45.07 (15.59)	42.31 (13.98)	41.93 (13.77)	40.30 (12.96)	<0.001
HR, beats/min	77.82 (16.30)	73.85 (15.09)	71.73 (14.40)	70.35 (14.47)	<0.001
MAP, mmHg	55.28 (11.43)	56.90 (10.70)	56.55 (11.44)	57.97 (11.58)	<0.001
Temperature, ◦C	36.14 (0.87)	36.13 (0.86)	36.07 (0.86)	36.01 (0.84)	<0.001
TWA-RR, beats/min	23.25 (3.43)	20.67 (2.67)	18.90 (2.39)	16.34 (2.25)	<0.001
TWA-SpO2, %	95.27 (2.81)	96.26 (1.75)	96.61 (1.67)	97.06 (1.74)	<0.001
TWA-FiO2, %	64.36 (11.47)	51.53 (6.89)	46.68 (5.63)	42.12 (5.03)	<0.001
Platelet (K/µl)	210.17(112.32)	207.16(98.16)	207.56 (100.65)	203.43 (100.94)	0.226
WBC (K/µl)	13.24 (7.29)	13.25 (6.55)	13.00 (6.37)	12.32 (5.98)	<0.001
PH	7.34 (0.11)	7.36 (0.09)	7.36 (0.10)	7.37 (0.09)	<0.001
Bicarbonate (mmol/L)	22.49 (5.16)	23.05 (4.69)	23.25 (4.76)	23.18 (4.97)	<0.001
PCO2(mmHg)	43.85 (12.80)	43.02 (11.49)	42.39 (11.02)	41.26 (11.26)	<0.001
Creatinine (mg/dl)	1.10 (0.80-1.63)	1.00 (0.70-1.50)	1.00 (0.70-1.40)	0.90 (0.70-1.40)	<0.001
PO2(mmHg)	98.00(71.25-169.00)	147.00(88.75-275.00)	178.00(106.00-299.00)	191.00(116.00-317.00)	<0.001
Lactate(mmol/L)	2.00 (1.30-3.40)	2.00 (1.30-3.10)	1.90 (1.30-2.90)	1.80 (1.20-2.90)	<0.001
Anion gap (mmol/L)	15.03 (4.16)	14.26 (3.94)	13.93 (3.90)	13.94 (4.11)	<0.001
Mechanical ventilator, n (%)	1417 (79.74%)	1674 (89.47%)	1651 (92.23%)	1733 (92.08%)	<0.001
Vasopressor, n (%)	895 (50.37%)	899 (48.05%)	867 (48.44%)	782 (41.55%)	<0.001
Renal replacement therapy, n (%)	86 (4.84%)	73 (3.90%)	73 (4.08%)	94 (4.99%)	0.278
CHF, n (%)	639 (35.96%)	653 (34.90%)	589 (32.91%)	498 (26.46%)	<0.001
Renal failure, n (%)	187 (10.52%)	214 (11.44%)	215 (12.01%)	237 (12.59%)	0.248
Liver disease, n (%)	138 (7.77%)	115 (6.15%)	120 (6.70%)	144 (7.65%)	0.166
Malignancy, n (%)	381 (21.44%)	311 (16.62%)	259 (14.47%)	308 (16.37%)	<0.001
COPD, n (%)	304(17.11%)	357(19.08%)	312(17.43%)	276(14.67%)	0.004
Respiratory failure, n (%)	1307(73.55%)	1128(60.29%)	961(53.69%)	819(43.52%)	<0.001
Pneumonia, n (%)	1042(58.64%)	883(47.19%)	731(40.84%)	604(32.09%)	<0.001

Notes: TWA, time-weighted average; SOFA, sequential organ failure assessment; SAPS II, simplified acute physiology score; HR, heart rate; MAP, mean arterial pressure; RR, respiratory rate; SpO2, peripheral oxygen saturation; FiO2, fraction of inspired oxygen; WBC, white blood cell; PH, potential of hydrogen; PCO2, partial pressure of carbon dioxide; PO2, partial pressure of oxygen; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease.

Measurement of ROX Over Time

To evaluate the ROX over time, we calculated the 24-hour time-weighted average (TWA) of SpO2, FiO2 and RR. Based on the above data, the ROX-TWA index was computed using the following formula: (SpO2-TWA) / (FiO2-TWA× RR-TWA).

Outcomes

28-day mortality were the primary endpoint of the study.

Statistical Analysis

Categorical variables were presented as percentage or frequency as appropriate and continuous variables were expressed as mean with standard deviation or Median (Q1-Q4 inter-quartile range). Categorical variables were compared using chi-square test or Fisher’s exact test and continuous variables were tested by one-way analysis of variance (ANOVA) or Kruskal–Wallis test. For the larger missing data, dummy variables were used to indicate missing covariate values. Meanwhile, some covariates are adjusted through non-linear relationships. Hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between ROX-TWA values with mortality were calculated with Cox proportional hazards regression models. Non-adjusting model, partially-adjusting model and fully-adjusting model are used to present the analysis results. Moreover, the median of each ROX-TWA quartile in the model was invoked as a continuous variable to test for linear trend.

The non-linear relationship between ROX-TWA values and mortality of sepsis was determined by generalized additive models (GAM). Based on the smoothing curve, the two-piecewise linear regression model was established, and further calculated the threshold effect of ROX-TWA values on 28-day mortality, with adjustment for potential confounders. When the ratio between 28-day mortality and ROX values was evident on a smoothed curve, the recursive method automatically calculated the inflection point. We selected log-likelihood ratio test to verify the one-line linear model with the two-piecewise linear regression model. Furthermore, this study investigated the relationship between ROX-TWA using or not using ventilator use and 28-day mortality by smoothing curves. Also, Continuous variables were converted into categorical variables based on corresponding clinical cut points, and subgroup analysis was performed using the stratified Cox regression model. Then, likelihood ratio test was used after effect correction test for subgroup indexes.

All data were processed using EmpowerStats (www.empowerstats.com,X&Y solutions, inc. Boston, Massachusetts) and statistical packages R (http://www.R-project.org; The R Foundation; version 3.6.3).

Participants

A total of 7,320 patients fulfilling the inclusion criteria were eventually included in this study. The disease characteristics of all participants were summarized in Table 1. All patients were further assigned into four different groups based on their ROX-TWA index. Compared with other three groups (Q1-Q4), patients with highest ROX-TWA were associated with a significantly lower sofa, heart rate and WBC count. The Kaplan-Meier curve revealed that the survival rate was notably higher in group 4 than in group 1 at 28 days of follow-up (log-rank P < 0.001, Fig. 2).

Association of ROX values with mortality

Table 2 revealed a significant association between ROX-TWA values and mortality rate in the non-adjusted model (P < 0.0001). In detail, patients in Q4 had a 59% reduced risk of death compared to patient in Q1 [HR 0.41 (95% CI 0.35, 0.48)]. After adjusting for potential variables factors, the hazard rations were 0.60(0.52, 0.70), 0.68(0.58, 0.80) and 0.60(0.51, 0.72) for ROX quartiles 2–4, respectively (P for trend < 0.0001).

Table 2

Association between ROX-TWA and 28-day mortality in different models
	Non-adjusted	Adjust I	Adjust II
ROX-TWA	0.88 (0.86, 0.90) < 0.0001	0.87 (0.85, 0.89) < 0.0001	0.93 (0.91, 0.95) < 0.0001
ROX-TWA quintiles
Q1(< 8.80)	1.0	1.0	1.0
Q2(8.80–9.80)	0.52 (0.46, 0.60) < 0.0001	0.50 (0.44, 0.57) < 0.0001	0.60 (0.52, 0.70) < 0.0001
Q3(9.90–11.80)	0.49 (0.43, 0.56) < 0.0001	0.46 (0.40, 0.53) < 0.0001	0.68 (0.58, 0.80) < 0.0001
Q4(> 11.80)	0.41 (0.35, 0.47) < 0.0001	0.40 (0.34, 0.46) < 0.0001	0.60 (0.51, 0.72) < 0.0001
P for trend	< 0.0001	< 0.0001	<0.0001
Notes: Adjust I model adjusted for age, gender; adjust II model adjusted for age, gender, SOFA score, SAPS II score, heart rate, mean arterial pressure, temperature, platelet, white blood cell, PH, bicarbonate, creatinine, anion gap, PCO2, PO2, lactate, Mechanical ventilator use, Vasopressor use, Renal replacement therapy use, malignancy, respiratory failure, COPD, pneumonia, congestive heart failure, renal failure and liver disease.

Plateau Effect of ROX on Mortality

To evaluate whether a linear relationship between ROX-TWA and mortality exist, we adopted the smooth curve fitting. After adjusting potential confounders, the nonlinear relationship between ROX-TWA values and mortality was observed, and the resulting curve demonstrated a critical point with two-stage changes (Fig. 3). When the ROX-TWA value was smaller than the tipping point, ROX-TWA value and mortality were negatively correlated. However, when the ROX-TWA value was greater than the tipping point, no significant association was observed. In Table 3, we further analyzed the threshold effect using curve fitting model. Since the P vale for the log likelihood ratio test was less than 0.05, we decided to use two-piecewise binary logistic regression to fit the correlation between ROX-TWA values and mortality. The final results revealed that the value of 8 was the inflection point of ROX-TWA values. When the ROX-TWA values was smaller than 8, the mortality decreases with the increase of ROX-TWA (HR = 0.75, 95%CI 0.71–0.80, P < 0.0001). When the ROX-TWA values was greater than 8, the increase of ROX-TWA values was not associated changes in mortality rate with a hazard ratio of 1.00(0.97, 1.03) (Table 3).

Table 3

Results of ROX-TWA and 28-day mortality two-piecewise linear regression.
Outcome	HR (95% CI)	P value
One-line linear regression model	0.93(0.91,0.95)	< 0.0001
Two-piecewise linear regression model
ROX-TWA < 8	0.75(0.71,0.80)	< 0.0001
ROX-TWA ≥ 8	1.00 (0.97, 1.03)	0.8160
Log-likelihood ratio test	< 0.001
Effect: 28-day mortality; cause: ROX values. Adjusted for age, gender, SOFA score SAPS II score, heart rate, mean arterial pressure, temperature, platelet, white blood cell, PH, bicarbonate, creatinine, anion gap, PCO2, PO2, lactate, Mechanical ventilator use, Vasopressor use, Renal replacement therapy use, malignancy, Respiratory failure, COPD, pneumonia, congestive heart failure, renal failure and liver disease.

Subgroup Analyses

We demonstrated by smoothing curves (Fig. 4) that Trends in ROX-TWA and 28-day mortality are consistent, regardless of whether using or not using mechanical ventilator. The results of the interactions and stratified analyses (Fig. 5) showed that the association between ROX-TWA and 28-day mortality was consistent in different subgroups (age, gender, mechanical ventilator use, malignancy, pneumonia, respiratory failure, COPD, congestive heart failure, renal failure and liver disease).

In this retrospective study of sepsis patients who participated in MIMIC-III 2001–2012, the ROX-TWA index was independently and negatively associated with 28-day mortality after adjustment other potential covariates. Moreover, we found a nonlinear relationship between the ROX-TWA index and 28-day mortality. The risk of short-term death decreased by 15% for each 1 increment in ROX-TWA. However, after the ROX-TWA values were greater than 8 (saturation effect), the declining trend of death risk as ROX-TWA values increased disappeared. Furthermore, our study demonstrated that the relationship between ROX-TWA and 28-day mortality was similar in sepsis patients using or not using MV by subgroup analysis.

The respiratory parameters included in the ROX index are SpO2, FiO2 and RR, which can comprehensively evaluate the respiratory status of patients. A recent study has shown that respiratory rate may be a predictor of mortality risk in patients with sepsis in the ICU(13).The prognostic value of the ROX index has been well studied in many diseases. Roca et al. first described the ROX index and prosed that ROX value > 4.88 was associated with a reduced risk of invasive mechanical ventilation(14). In addition, when ROX index ≥ 9.2, it was also used to guide the weaning of HFOT(High-flow nasal oxygen therapy)(15). Furthermore, ARDS patients defined by PaO2/FIO2 rate or SpO2/FIO2 rate have very alike clinical characteristics and outcomes(16). During the coronavirus disease pandemic, the ROX index showed a good efficacy to examine the severity of patient(17, 18). Similar with our study, Lee et al. examined the ROX index in septic patients and discovered that a ROX index < 10 was independent risk factors for 28-day mortality(9). Multiple organ dysfunction, such as cardiac and pulmonary dysfunction, leads to abnormal ventilation ratios, tissue ischemia and hypoxia in sepsis patients, and finally to reduced SpO2 and increased respiratory rate. These pathological changes may explain the correlation between the ROX index and mortality in sepsis patients. The difference between plateau points of ROX index might be due to different definitions of sepsis and changes in the concept of treatment(1, 19–21). Nevertheless, both studies verified that the ROX index could be a reliable and useful tool in the clinical setting to spontaneously identify patients with increased risk.

To my knowledge, the study by Lee et al. is the first to validate the association between ROX and sepsis death, but the study has some potential flaws. First, their study did not measure dynamic changes in the ROX index. The ROX index probably improved with patient treatment, so a single measurement does not reflect the true relationship between ROX and prognosis of sepsis patients. This may be a bias. Secondly, participants included intubated patients in their study, but since the ROX was developed for patients in spontaneous ventilation, they did not consider the impact of positive ventilation may influence the results. On this basis, our study further validated the ROX index as an independent prognostic factor for 28-day mortality in patients with sepsis. We took into account the time factor in the calculation and obtained ROX values by time-weighted averaging. On the one hand, time-weighted averages may reduce bias caused by human and instrumental errors and also avoid outliers. On the other hand, the patient's condition worsened or improved, his respiratory rate and blood oxygen saturation will also change. Accordingly, repeated or continuous measurements will be more reflective of the true situation and more helpful for the management of these patients. In addition, we found a similar relationship between ROX and mortality in patients using or not using MV. In a retrospective study of a 50,000-strong sepsis population in Korea from September 2019 to April 2020, 28-day mortality was always significantly higher in the low-level ROX than in the high-level ROX whether using or not using MV (P < 0.05)(22). This study further supports the reliability of our conclusions. More importantly, we not only adjusted for the confounding factors that may affect the prognosis of sepsis reported in the previous literature, but also further explored the nonlinear relationship between ROX and mortality. To the best of our knowledge, this is the first report of the saturation effect of ROX and 28-day mortality, while providing an exact inflection point.

Assessing severity of the patient is very important for the planned intervention and prognosis of the disease. our study showed the prognostic value of ROX index among septic patients, and the increase of its specific range is directly related to the reduction of the risk of death. At present, clinicians often use multiple indicators and prognostic score systems such as lactate, Sofa and APACHE II when the treatment of sepsis. Compared to lactate, Sofa and APACHE II in the clinical practice, The ROX has fewer variables, and is easier to collect, which is conducive to early and rapid judgment of disease treatment. On the other hand, the ROX index can be obtained spontaneously through the ECG monitor without additional examination, and can be repeatedly after employment of the intervention. Therefore, monitoring the ROX index may help find critically ill patients and perform early treatment.

Compared with Lee studies, we have a larger sample size and better balance baseline confounding factors. In addition, ROX can be time-varying, Single data may be lacks analysis of baseline levels and continuous fluctuations, so we calculated time-weighted average to better assess the reliability of the relationship between ROX values and 28-day mortality among septic patient. However, this study has some limitations. First, all of our patients are septic patient, our results may not be suitable for patients with other diseases. Second, even though the sample size is large, it is a retrospective study obtained from a single center that might not be representable for the general population. Finally, the treatment for sepsis evolved during the study and may also affect the result. According to the threshold effect of ROX, we taking more individualized treatment to reduce ICU mortality. Of course, this needs a large-scale prospective study.

In conclusion, our data indicated a curvilinear association between ROX-TWA values and 28-day mortality among septic patient. When ROX--TWA value was less than 8, 28-day mortality decreased with the increase of ROX--TWA value.

DATA AVAILABILITY STATEMENT

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://mimic.physionet.org.

AUTHOR CONTRIBUTIONS

Yukang Dong and Jiangquan Fu conceived the idea, contributed a lot of valuable advice, and drafted the manuscript. Yukang Dong conducted the research and collected data. Rui Huang wrote the draft of the main manuscript and analyzed the data and performed statistical analysis. All authors read and approved the final manuscript.

Declaration Of Conflicting Interests

The authors declare that there were no conflicts of interest.

Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. (2016) Feb 23315(8):801-10. doi:10.1001/jama.2016.0287.
Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. (2001) Jul29(7):1303-10. doi:10.1097/00003246-200107000-00002.
Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021. Crit Care Med. (2021) Nov 149(11):e1063-e1143. doi:10.1097/CCM.0000000000005337.
Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. (1985) Oct13(10):818-29.
Zimmerman JE, Kramer AA, McNair DS, Malila FM. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today's critically ill patients. Crit Care Med. (2006) May34(5):1297-310. doi:10.1097/01.CCM.0000215112.84523.F0.
Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA. (1993) Dec 22-29270(24):2957-63. doi:10.1001/jama.270.24.2957.
Moreno RP, Metnitz PG, Almeida E, Jordan B, Bauer P, Campos RA, et al. SAPS 3--From evaluation of the patient to evaluation of the intensive care unit. Part 2: Development of a prognostic model for hospital mortality at ICU admission. Intensive Care Med. (2005) Oct31(10):1345-55. doi:10.1007/s00134-005-2763-5.
Roca O, Caralt B, Messika J, Samper M, Sztrymf B, Hernandez G, et al. An Index Combining Respiratory Rate and Oxygenation to Predict Outcome of Nasal High-Flow Therapy. Am J Respir Crit Care Med. (2019) Jun 1199(11):1368-1376. doi:10.1164/rccm.201803-0589OC.
Lee CU, Jo YH, Lee JH, Kim J, Park SM, Hwang JE, et al. The index of oxygenation to respiratory rate as a prognostic factor for mortality in Sepsis. Am J Emerg Med. (2021) Jul45:426-432. doi:10.1016/j.ajem.2020.09.052.
Girardis M, Alhazzani W, Rasmussen BS. What's new in oxygen therapy? Intensive Care Med. (2019) Jul45(7):1009-1011. doi:10.1007/s00134-019-05619-9.
Bastarache JA, Ware LB, Bernard GR. The role of the coagulation cascade in the continuum of sepsis and acute lung injury and acute respiratory distress syndrome. Semin Respir Crit Care Med. (2006) Aug27(4):365-76. doi:10.1055/s-2006-948290.
Zhou J, Qian C, Zhao M, Yu X, Kang Y, Ma X, et al. Epidemiology and outcome of severe sepsis and septic shock in intensive care units in mainland China. PLoS One. (2014) 9(9):e107181. doi:10.1371/journal.pone.0107181.
Chen Y, Luo M, Cheng Y, Huang Y, He Q. A nomogram to predict prolonged stay of obesity patients with sepsis in ICU: Relevancy for predictive, personalized, preventive, and participatory healthcare strategies. Front Public Health. (2022) 10:944790. doi:10.3389/fpubh.2022.944790.
Roca O, Messika J, Caralt B, Garcia-de-Acilu M, Sztrymf B, Ricard JD, et al. Predicting success of high-flow nasal cannula in pneumonia patients with hypoxemic respiratory failure: The utility of the ROX index. J Crit Care. (2016) Oct35:200-5. doi:10.1016/j.jcrc.2016.05.022.
Rodriguez M, Thille AW, Boissier F, Veinstein A, Chatellier D, Robert R, et al. Predictors of successful separation from high-flow nasal oxygen therapy in patients with acute respiratory failure: a retrospective monocenter study. Ann Intensive Care. (2019) Sep 119(1):101. doi:10.1186/s13613-019-0578-8.
Chen W, Janz DR, Shaver CM, Bernard GR, Bastarache JA, Ware LB. Clinical Characteristics and Outcomes Are Similar in ARDS Diagnosed by Oxygen Saturation/Fio2 Ratio Compared With Pao2/Fio2 Ratio. Chest. (2015) Dec148(6):1477-1483. doi:10.1378/chest.15-0169.
Gianstefani A, Farina G, Salvatore V, Alvau F, Artesiani ML, Bonfatti S, et al. Role of ROX index in the first assessment of COVID-19 patients in the emergency department. Intern Emerg Med. (2021) Oct16(7):1959-1965. doi:10.1007/s11739-021-02675-2.
Zaboli A, Ausserhofer D, Pfeifer N, Sibilio S, Tezza G, Ciccariello L, et al. The ROX index can be a useful tool for the triage evaluation of COVID-19 patients with dyspnoea. J Adv Nurs. (2021) Aug77(8):3361-3369. doi:10.1111/jan.14848.
Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. (1992) Jun101(6):1644-55. doi:10.1378/chest.101.6.1644.
Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. (2003) Apr31(4):1250-6. doi:10.1097/01.CCM.0000050454.01978.3B.
Johnson AEW, Aboab J, Raffa JD, Pollard TJ, Deliberato RO, Celi LA, et al. A Comparative Analysis of Sepsis Identification Methods in an Electronic Database. Crit Care Med. (2018) Apr46(4):494-499. doi:10.1097/CCM.0000000000002965.
Ahn S, Park J, Song J, Kim J, Cho H, Moon S. Association of ROX Index with Mechanical Ventilator Use in Sepsis Patients in the Emergency Department. J Clin Med. (2022) Jan 1111(2). doi:10.3390/jcm11020342.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Association between ROX index with 28-day mortality in sepsis: A retrospective cohort study

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Data Source and Participants

Data Extraction

Measurement of ROX Over Time

Outcomes

Statistical Analysis

Results

Participants

Association of ROX values with mortality

Plateau Effect of ROX on Mortality

Subgroup Analyses

Discussion

CONCLUSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1