Ratio of oxygen saturation index for predicting high‑flow nasal cannula outcomes in emergency department for COVID‑19 patients with severe hypoxemia: A retrospective study

Abstract

OBJECTIVES: High flow nasal cannula (HFNC) oxygen therapy has been used as an initial ventilatory support for coronavirus disease 2019 (COVID 19) patients with mixed levels of acute hypoxemic respiratory failure (AHRF). However, the effectiveness of HFNC when used as initial ventilatory support in COVID 19 patients with severe AHRF exclusively is not well documented. Ratio of oxygen saturation (ROX) index (ROX = [SpO2 /fraction of inspired oxygen]/respiratory rate) was shown to predict the outcome of HFNC in intensive care unit patients. Our study aimed to evaluate the utility of the ROX index for predicting HFNC therapy success/failure in COVID 19 patients with severe AHRF when HFNC is used as the first line of ventilatory support.

METHODS: Retrospective study in 67 COVID 19 patients with severe AHRF receiving HFNC in the emergency department at a tertiary care academic medical center. ROX index was determined at 0, 2, 6, 12, and 24 h of HFNC onset. The need to escalate to noninvasive or invasive ventilatory support was documented. The receiver operating characteristic curves were performed and areas under the curves (AUCs) were calculated to evaluate the accuracy of ROX index for differentiating between patients who will succeed or fail HFNC therapy.

RESULTS: HFNC therapy was successful in 19 patients (28.1%) and failed in 48 patients (71.6%). ROX index after 6 h of HFNC initiation had the best predictive capacity for the outcome of HFNC therapy (AUC = 0.78). ROX index >4.4 at 6 h of HFNC onset was significantly associated with HFNC success/failure.

CONCLUSION: ROX index at 6 h after initiating HFNC therapy in COVID 19 patients with severe AHRF has a good predictive capacity for HFNC success/failure.

Introduction

High flow nasal cannula (HFNC) oxygen therapy is an easy to use ventilatory support modality that delivers high fractions of heated and humidified oxygen at flow rates reaching 60–100 L/min.[1] It represents a superior alternative to conventional oxygen therapy for patients with acute hypoxemic respiratory failure (AHRF).[1 3] HFNC can reduce the rate of intubation and invasive mechanical ventilation (IMV) in patients with AHRF.[4] Although HFNC may avoid the need for mechanical ventilation (MV) in some patients with AHRF, it may unduly delay the initiation of MV in others and worsen their outcome.[5] As such, it is essential to identify as early as possible patients who will fail HFNC trials so that they are escalated to more aggressive respiratory support modalities such as noninvasive ventilation (NIV) and invasive MV.

Roca et al. first described the ratio of oxygen saturation (ROX) index and showed that it can predict HFNC success/failure in intensive care unit (ICU) patients with AHRF.[6,7] ROX index, expressed as (SpO₂ /FiO₂ )/RR, where SpO₂ is the oxygen saturation by pulse oximetry, FiO₂ is the fraction of inspired oxygen, and RR is the respiratory rate, can be easily obtained and used at the bedside. Roca et al. showed that a ROX value of >4.88 determined at 12 h of HFNC therapy is a valuable predictor of patients at low risk for HFNC failure.[7]

In recent years, the coronavirus disease 2019 (COVID 19) pandemic has resulted in an unprecedented number of patients with AHRF flooding health care facilities all over the world. ICUs were filled up quickly and COVID 19patients were boarded in emergency departments (EDs) for substantial periods. HFNC was extensively utilized in EDs for COVID 19 patients, particularly in those with mild to moderate hypoxemia.[8,9] However, no data have been published about tools for predicting the outcome of HFNC when utilized as the initial form of respiratory support in COVID 19 patients with severe hypoxemia managed in ED. The aim of the current study is to assess whether the ROX index is a valuable predictor of HFNC therapy success/failure in COVID 19 patients with severe AHRF treated initially with HFNC in ED.

Material and Methods

This study was approved by the Institutional Review Board at the (BIO 2021 0318; November 25, 2021). Given the retrospective nature of the study, no informed consent was deemed necessary.

We performed a single center retrospective cohort analysis of a clinical database of patients treated for severe AHRF secondary to COVID 19 and immediately received HFNC (Airvo 2; Fisher and Paykel, Auckland, New Zealand) upon presentation to the ED at an academic medical center staffed with emergency medicine specialists from March 2020 to March 2021. Patients were included if ≥18 years of age, with a laboratory confirmed diagnosis of COVID 19 by polymerase chain reaction testing, with either partial pressure of arterial oxygen (PaO₂ )/FiO₂≤100 from first arterial blood gas results or SpO₂ /FiO₂≤140 and treated with HFNC for at least 2 h in the ED. Patients were excluded if endotracheal intubation was performed, or noninvasive bilevel positive airway pressure was applied before initiation of HFNC. Moreover, patients with a “do not intubate order” were excluded.

HFNC was initiated with flow of 50–60 L/min, and FiO₂ was adjusted to maintain SpO₂ ≥92%. Patients were monitored by noninvasive measurements of heart rate, blood pressure, oxygen saturation, and RR. HFNC failure was defined as escalation to noninvasive ventilatory (NIV) support or the need for intubation and initiation of MV. Escalation of therapy was generally based on the presence of hypoxemia with the inability to maintain SpO₂ ≥92% despite receiving maximal FiO₂ and/or breathing frequency >35 breaths/min with associated signs of respiratory distress/failure.

Adjunct therapies targeting COVID 19 were administered at the discretion of the ED team and commonly included systemic glucocorticoids, remdesivir, and anticoagulation.

Data were collected from the patients’ electronic medical records during December 2021–January 2022. Patients’ demographics, relevant clinical data, and past medical history were obtained and followed up until patients were discharged from the ED or expired. Clinical data included heart rate, oxygen saturation, blood pressure, and RR. Laboratory data included D dimer, procalcitonin, C reactive protein, lactate levels, and arterial blood gases. ROX index was determined for all patients at 2 h and subsequently at 6, 12, and 24 h after initiation of HFNC therapy or until HFNC failure is observed. Acute Physiology and Chronic Health Evaluation II scores in the first 24 h of ED stay, pneumonia severity index, and sequential organ failure assessment scores upon ED admission were determined. The incidence of intubation and MV and the use of noninvasive ventilation (NIV) were also recorded.

The main outcome was either the escalation to NIV or the need of intubation and MV both reflecting the failure of HFNC therapy. Quantitative variables were expressed as mean and standard deviation or median and interquartile range (IQR) if normality criteria, as tested with the Kolmogorov–Smirnov test, were not met. Categorical variables were expressed as frequencies and percentages. Continuous variables were compared using the Student’s t test or Mann–Whitney U test, as appropriate. For categorical variables, the comparison was made using the Chi square test or Fisher’s exact test, as appropriate. Multivariate logistic regression analysis was performed to assess the factors associated with HFNC therapy outcome. Adjusted odds ratio and 95% confidence interval (CI) were reported. Receiver operating characteristic (ROC) curves were performed and areas under the curves (AUCs) were calculated to evaluate variables for differentiating patients who will succeed or fail HFNC therapy. AUC values were analyzed using the DeLong statistical test. Cutoff values that best discriminate between HFNC success and failure was chosen to maximize the sum of sensitivity and specificity. Kaplan–Meier curves were used to determine the probability of HFNC failure/success at follow up time intervals. These curves were compared using the log rank test. Statistical analyses were performed using the SPSS statistical package (SPSS Inc., Chicago, IL, USA). P < 0.05 was considered statistically significant.

Results

Sixty seven patients with COVID 19 and with severe hypoxemia were initially treated with HFNC in the ED [Figure 1]. Baseline characteristics of the patients’ population are presented in Table 1. Nineteen patients (28.4%) were categorized as HFNC therapy success, while the remaining 48 patients (71.6%) required escalation of respiratory support to either NIV or intubation and MV in the ED and were categorized as HFNC failure. There were no statistically significant differences in baseline characteristics among patients who succeeded or failed HFNC therapy [Table 1].

Figure 1. Study flowchart. PaO2 : Partial pressure of arterial oxygen, FiO2 : Fraction of inspired oxygen, SpO2 : Oxygen saturation by pulse oximetry

Table 1. Demographics and baseline characteristics

Patients with successful HFNC therapy had significantly higher SpO₂ /FiO₂ at 12 h, lower RR at 6 h, and higher ROX values at 6 and 24 h of HFNC onset [Table 2]. The areas under the ROC curve (AUROCs) predicting the accuracy of SpO₂ /FiO₂ , RR, and ROX at different time intervals after the onset of HFNC are presented in Table 3. Only the ROX index determined at 6 and 24 h after the onset of HFNC had clinically significant predictive capacity (AUROC ≥0.7) for HFNC success/failure [Table 3]. However, the ROX index after 6 h of HFNC treatment demonstrated the best prediction accuracy with AUROC of 0.78 (95% CI: 0.66–0.90) compared to AUROC of 0.71 (95% CI: 0.54–0.89) at 24 h of HFNC treatment.

Table 2. Respiratory variables during high‑flow nasal cannula treatment

Table 3. Diagnostic accuracy of respiratory variables at different time points of need for NIV or mechanical ventilation in patients treated with high‑flow nasal cannula

The best cutoff threshold for the ROX index at 6 h of HFNC onset was estimated to be 4.40. A ROX index ≤4.40 at 6 h after HFNC onset had a sensitivity of 71%, a specificity of 76%, a positive predictive value of 88.6%, a negative predictive value of 50%, a positive likelihood ratio of 3.0, and a negative likelihood ratio of 0.4 in predicting HFNC therapy failure. The unadjusted and confounder adjusted odds ratios for HFNC success (i.e., no need for NIV/IMV) when ROX >4.4 were 7.8 (95% CI [2.1–28.3]) and 5.2 (95% CI [1.0–28.2]), respectively.

The follow up time on patients was 45.78 ± 67.14 h. The median (IQR) duration of HFNC therapy was 1 (1–2) days in patients who failed HFNC compared to 1 (1–5) days in patients who succeeded HFNC therapy (P = 0.135). The median time for HFNC treatment without the need for NIV or intubation and MV was significantly higher in patients whose ROX score was >4.4 at 6 h post HFNC initiation compared to patients with ROX score ≤4.4 (48 h, 95% CI [7.8–88.1] vs. 24 h, 95% CI [15.8–32.2] respectively; P = 0.016).

Kaplan–Meier plots showing the probability of HFNC success according to the ROX group are shown in Figure 2. Patients with ROX index >4.4 after 6 h of HFNC were less likely to need NIV or intubation and MV (P = 0.016).

Figure 2. Kaplan–Meier plot showing the cumulative probability of remaining free of either noninvasive ventilation or intubation and mechanical ventilation in patients with COVID‑19 and severe hypoxemia treated with HFNC therapy in the emergency department. HFNC: High‑flow nasal cannula, MV: Mechanical ventilation, ROX: Ratio of oxygen saturation, NIV: Noninvasive ventilation

Discussion

We showed that the ROX index determined at 6 h from the onset of HFNC therapy that is >4.4 is a good predictor of HFNC success/failure in COVID 19 patients with severe hypoxemia treated in the ED.

During the COVID 19 pandemic, HFNC was feasible in treating patients with AHRF due to COVID 19 in ICU and non ICU settings.[10,11] To date, there is a lack of robust data from randomized controlled trials (RCTs) on the timely use of HFNC in COVID 19 associated AHRF as it is hard to conduct such RCTs during a pandemic. Nevertheless, several studies have reported important clinical benefits of HFNC in COVID 19 patients and the potential role of the ROX index in predicting the outcome of HFNC.[12,13] However, none of these studies focused on a cohort group of patients with severe hypoxemia (i.e., PaO₂ /FiO₂ ≤100) who were managed exclusively in ED. Hu et al. showed in COVID 19 patients receiving HFNC in specialized respiratory units that ROX >5.5 after 6 h of HFNC therapy is a good predictor of HFNC success.[12] Although our findings are similar, there remain several important differences between the two studies. First, in Hu et al. study, patients were managed in specialized respiratory units while our patients were managed exclusively in the ED. Second, the median (interquartile) PaO₂ /FiO₂ ratio for patients in Hu et al. study was 116 (102.1–132.0) compared to 64.5 (57.3– 75) in the current study. As such and as per the Berlin definition of acute respiratory distress syndrome,[14] the patients in Hu et al. study can be classified as moderate AHRF while ours are severe AHRF.[12] Third, in Hu et al. study, the ROX index >5.55 at 6 h of HFNC therapy was associated with HFNC success, while in the current study, the ROX index >4.4 was associated with HFNC success. Finally, the rate of HFNC failure was higher in the current study compared to Hu et al. study (71.6% vs. 38.1%, respectively). The higher failure rate could be attributed to the higher severity of hypoxemia in our patients. Recently, Costa et al. reported a similar high HFNC failure rate of 69.6% in COVID 19 patients with severe hypoxemia.[14]

Identifying a reliable and easy to use predictor of the success/failure of HFNC is of great importance. It provides not only an objective index on which to base critical interventions and decisions such as termination of unduly HFNC therapy and escalation to intubation and MV but also specifies a cutoff value to use throughout the process. Unnecessary delays in intubation and initiation of MV might increase mortality in patients receiving HFNC therapy.[15] Subsequently, the identification of patients who can be maintained on HFNC therapy without being exposed to unnecessary risks and with the intention of improving their outcomes is paramount. However, close attention should be made on how, when, and what threshold to use when applying the ROX index during HFNC therapy. Several studies on using HFNC as first line treatment at different locations in the hospital and for different patient’s populations reported different ROX cutoff values ranging from 4.94 to 5.99.[12 14] However, ROX thresholds as high as 11.17 and as low as 3.0 were reported in COVID 19 patients who received HFNC therapy after liberation from MV.[15,16] These findings suggest that despite the feasibility of ROX in predicting HFNC outcome, no single ROX value is appropriate and thus different ROX values should be used at specific time intervals for different patients’ categories at different locations in the hospital. In the current study, we suggest using an ROX threshold of 4.4 at 6 h from HFNC onset in COVID 19 patients with severe hypoxemia receiving HFNC in the ED.

HFNC could be a valuable and feasible treatment option for patients with COVID 19 since its easy setup allows for rapid training even for nonexpert clinicians with heterogeneous backgrounds.[1,2,12 14] Thus, its implementation in a non ICU setting such as in ED is crucial for countries and health care systems with shrinking critical care capacities and resources for invasive MV. Our patients were COVID 19 patients who received HFNC as an initial form of respiratory support in the ED. HFNC failure rate was 71.6% and is probably the highest HFNC failure rate reported in the literature.[2,7,12 14] However, our patients’ cohort could be considered the sickest cohort with the most severe form of AHRF with a median (IQR) PaO₂ /FiO₂ of 64.5 (57.3–75). Nevertheless, the current study provides valuable information for the management of COVID 19 patients in ED. First, 28.4% of COVID 19 patients with severe AHRF were successfully managed with HFNC in ED. Second, a simple ROX measured as early as 6 h after HFNC onset may be used to identify COVID 19 patients with severe AHRF who will succeed/fail HFNC therapy in ED so that HFNC is either maintained or patients are escalated to NIV or invasive MV.

In our patients, the median duration of HFNC was 1 day for both HFNC success and failure groups with no statistical difference. However, when ROX was >4.4 at 6 h from HFNC onset, the median time of HFNC without the need for either NIV or intubation and MV was significantly increased to 48 h. This suggests that in COVID 19 patients with severe AHRF, ROX >4.4 at 6 h from HFNC onset is valuable in identifying patients who will tolerate an additional day without failing HFNC in ED.

Limitations

The current study has some limitations. First, this was a single center study and as such the current findings might not be generalized to other clinical settings; nevertheless, our findings should help in providing guidance for possible implementation at other health care facilities. Second, the number of patients was not large enough (only 67 patients) and based on a convenient sample approach of selected COVID 19 patients who presented to our ED between March 2020 and March 2021 with a PaO₂ /FiO₂ ≤100. However, despite the relatively small number, this group of patients represents a unique cohort of patients with the most severe forms of AHRF (PaO₂ /FiO₂≤100) who are usually immediately intubated and started on invasive MV upon presentation to the ED or at least given a very short trial of NIV to save them from intubation and MV. To our knowledge, very few, if any, studies have reported findings on such a cohort group of COVID 19 patients with severe AHRF who received HFNC therapy as the first line of intervention for severe hypoxemia. Third, the transition from HFNC to NIV or IMV was decided by the medical team. Different ED physicians have different opinions on the criteria for terminating HFNC and switching to NIV or IMV. However, this study can still reflect on how HFNC was actually used in ED for COVID 19 patients. Fourth, since we included COVID 19 patients with only severe AHRF, our identified ROX of 4.4 cannot be generalized and used in COVID 19 patients with mild or moderate AHRF. Fifth, despite no statistically significant difference in the frequency of comorbidities between the success and failure groups, the low sample size might have masked a possible effect of lower comorbidities in the success group. This can be further evaluated with a study with larger group of patients. Finally, the ROX index was determined at specified and discrete time intervals after HFNC onset (i.e., 2, 6, 12, and 24 h), while HFNC failure may occur at any point in time. Nevertheless, previous studies have shown that the median duration of HFNC treatment in patients with AHRF was at least 24 h and hence most patients may be assessed with ROX index at the currently specified time intervals.[17]

Conclusion

Our study assessed the use of HFNC therapy in a homogenous cohort of COVID 19 patients with severe AHRF treated in ED. ROX index >4.4 at 6 h after HFNC initiation had a good predictive ability for HFNC therapy success in the ED.

How to cite this article: Karam C, Oseili A, Shebbo FM, Fakih M, El-Khatib MF. Ratio of oxygen saturation index for predicting high‑flow nasal cannula outcomes in emergency department for COVID-19 patients with severe hypoxemia: Aretrospective study. Turk J Emerg Med 2024;24(1):41-7.

This study was approved by the Institutional Review Board at the American University of Beirut, Beirut, Lebanon (IRB# BIO‑2021‑0318). IRB approval date: November 25, 2021.

All authors had full access to the data, contributed to the study, reviewed, edited, and approved the final version for publication, and were responsible for its accuracy and integrity.

None Declared.

None.

References

1. El Khatib MF. High flow nasal cannula oxygen therapy during hypoxemic respiratory failure. Respir Care 2012;57:1696 8.
2. Frat JP, Thille AW, Mercat A, Girault C, Ragot S, Perbet S, et al. High flow oxygen through nasal cannula in acute hypoxemic respiratory failure. N Engl J Med 2015;372:2185 96.
3. Demoule A, Vieillard Baron A, Darmon M, Beurton A, Géri G, Voiriot G, et al. High flow nasal cannula in critically III patients with severe COVID 19. Am J Respir Crit Care Med 2020;202:1039 42.
4. Ni YN, Luo J, Yu H, Liu D, Ni Z, Cheng J, et al. Can high flow nasal cannula reduce the rate of endotracheal intubation in adult patients with acute respiratory failure compared with conventional oxygen therapy and noninvasive positive pressure ventilation? A systematic review and meta analysis. Chest 2017;151:764 75.
5. Kang BJ, Koh Y, Lim CM, Huh JW, Baek S, Han M, et al. Failure of high flow nasal cannula therapy may delay intubation and increase mortality. Intensive Care Med 2015;41:623 32.
6. Roca O, Caralt B, Messika J, Samper M, Sztrymf B, Hernández 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;199:1368 76.
7. Roca O, Messika J, Caralt B, García 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;35:200 5.
8. Crimi C, Pierucci P, Renda T, Pisani L, Carlucci A. High flow nasal cannula and COVID 19: A clinical review. Respir Care 2022;67:227 40.
9. Long B, Liang SY, Lentz S. High flow nasal cannula for adult acute hypoxemic respiratory failure in the ED setting. Am J Emerg Med 2021;49:352 9.
10. Franco C, Facciolongo N, TonelliR, DongilliR, VianelloA, PisaniL, et al. Feasibility and clinical impact of out of ICU noninvasive respiratory support in patients with COVID 19 related pneumonia. Eur Respir J 2020;56:2002130.
11. Tonetti T, Grasselli G, Zanella A, Pizzilli G, Fumagalli R, Piva S, et al. Use of critical care resources during the first 2 weeks (February 24 March 8, 2020) of the COVID 19 outbreak in Italy. Ann Intensive Care 2020;10:133.
12. Hu M, Zhou Q, Zheng R, Li X, Ling J, Chen Y, et al. Application of high flow nasal cannula in hypoxemic patients with COVID 19: A retrospective cohort study. BMC Pulm Med 2020;20:324.
13. Costa WN, Miguel JP, Prado FD, Lula LH, Amarante GA, Righetti RF, et al. Noninvasive ventilation and high flow nasal cannula in patients with acute hypoxemic respiratory failure by COVID 19: A retrospective study of the feasibility, safety and outcomes. Respir Physiol Neurobiol 2022;298:103842.
14. ARDS Definition Task Force, Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, et al. Acute respiratory distress syndrome: The Berlin definition. JAMA 2012;307:2526 33.
15. Simioli F, Annunziata A, Langella G, Polistina GE, Martino M, Fiorentino G. Clinical outcomes of high flow nasal cannula in COVID 19 associated postextubation respiratory failure. A single Centre case series. Anaesthesiol Intensive Ther 2020;52:373 6.
16. Chandel A, Patolia S, Brown AW, Collins AC, Sahjwani D, Khangoora V, et al. High flow nasal cannula therapy in COVID 19: Using the ROX index to predict success. Respir Care 2021;66:909 19.
17. Goligher EC, Slutsky AS. Not just oxygen? Mechanisms of benefit from high flow nasal cannula in hypoxemic respiratory failure. Am J Respir Crit Care Med 2017;195:1128 31.