Authors: Shuai Yang, Yan He, Shengming Liu, Yuxin Zhang
Categories: Correspondence
Source: Chinese Medical Journal
To the Editor: Acute heart failure (AHF) is a leading cause of hospitalization. Most patients with AHF present with hypoxia and dyspnea. Treatment of these patients often requires oxygenation via facemask and more intensive respiratory support methods, including non-invasive ventilation (NIV) or invasive ventilation through endotracheal intubation.
Conventional oxygen therapy (COT) with a nasal cannula or face mask is the most common oxygen therapy method. However, if the patient's inspiratory flow rate is greater than that provided by the nasal cannula or mask, COT cannot guarantee a sufficient fraction of inspired oxygen (FiO2). NIV, which can create a much higher gas flow rate and positive airway pressure, is recommended as a first-line treatment for acute hypoxia and dyspnea associated with AHF. Nevertheless, NIV may be poorly tolerated and is associated with many complications in certain patients. In recent years, high-flow nasal cannula (HFNC) therapy is increasingly being used for management of hypoxemia and respiratory failure. This technique can provide heated, moist oxygen through the nasal cannula and offer a much higher and predictable gas flow rate (up to 60 L/min) and FiO2 (up to 100%). A few studies have shown that HFNC therapy may be an option in patients with AHF, but others have not drawn the same conclusion. Therefore, we conducted a meta-analysis to investigate whether HFNC therapy has more advantages than COT and NIV in terms of reducing the intubation rate and improving oxygenation efficiency in patients with AHF.
This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. The protocol of this review was registered on PROSPERO (registration No. CRD42021264563). The databases of Cochrane Central Register of Controlled Trials, PubMed, EMBASE, China National Knowledge Infrastructure, China Science and Technology Journal Database, and Wanfang Data were searched from their inception to July 31, 2021. The search term details were presented in Supplementary Table 1. The studies were required to meet the following inclusion being randomized controlled trials (RCTs), comparing HFNC therapy with NIV or COT, including adult patients with AHF-induced hypoxia and dyspnea, and having available outcome data. The primary outcome of our study was to investigate whether HFNC therapy vs. COT or HFNC therapy vs. NIV resulted in a similar intubation rate. The secondary outcomes were the pulse oxygen saturation (SpO2), respiratory rate (RR), heart rate (HR), and arterial blood gas (ABG) parameters (arterial oxygen partial pressure [PaO2], arterial carbon dioxide partial pressure [PaCO2]) after HFNC therapy, NIV, or COT application. The title and abstract of potentially eligible RCTs were independently evaluated by two researchers, and the full text was then assessed according to the eligibility criteria. Using the Cochrane Collaboration Risk of Bias tool, two reviewers independently assessed the methodological quality of the selected studies. Statistical analysis was performed with Review Manager software (RevMan 5.4, Microsoft, Redmond, WA, USA). Mean difference (MD) with 95% confidence interval (95% CI) was calculated for the continuous outcomes and risk ratio with 95% CI for the dichotomous outcomes. Heterogeneity was tested by the I^2^ statistics. A random effect model was used in case of substantial heterogeneity (I^2^ ≥50%); Otherwise, the fixed effect model was used. Two-sided tests were used in all analyses, and P < 0.05 was considered statistically significant.
Fifteen studies involving 1018 subjects (HFNC, n = 508; COT, n = 335; NIV, n = 175) were included in the meta-analysis. The flowchart was shown in Supplementary Figure 1. Nine studies^[1–9]^ compared HFNC therapy with COT, and six studies^[10–15]^ compared HFNC therapy with NIV. The basic characteristics and quality assessment of the included studies were shown in Supplementary Table 2, and Supplementary Figure 2. Funnel plots [Supplementary Figure 3] were visually inspected and did not demonstrate evidence of publication bias. Compared with COT, HFNC therapy was associated with a significant reduction in the intubation rate (relative risk, 0.31; 95% CI, 0.16–0.59; P = 0.0004; heterogeneity, I^2^ = 0%; P = 0.91), RR (MD, −3.35; 95% CI, −3.93 to −2.77; P < 0.0001; heterogeneity, I^2^ = 0%; P = 0.99), and HR (MD, −7.80; 95% CI, −10.42 to −5.17; P < 0.0001; heterogeneity, I^2^ = 54%; P = 0.04); a significant improvement in SpO2 (MD, 2.09; 95% CI, 0.83–3.34; P = 0.001; heterogeneity, I^2^ = 82%; P < 0.0001) and PaO2 (MD, 10.15; 95% CI, 5.68–14.62; P < 0.0001; heterogeneity, I^2^ = 80%; P < 0.0001); and there was no difference in PaCO2 between the two groups [Supplementary Figure 4]. When HFNC therapy was compared with NIV, the post-intervention data showed no significant difference in the intubation rate, RR, HR, SpO2, PaO2, or PaCO2 [Supplementary Figure 5].
The main finding of our study is that HFNC therapy significantly reduced the rate of intubation compared with COT in adult patients with AHF. We also clarified the beneficial effect of HFNC therapy in patients with AHF using objective parameters of RR, HR, SpO2, and ABG parameters. This meta-analysis demonstrated that compared with COT, the application of HFNC therapy can significantly increase PaO2 and SpO2 and decrease HR and RR in patients with AHF-induced hypoxia and dyspnea. This is encouraging and indicates the beneficial physiological effects of HFNC therapy. The inspiratory flow changes with the patient's inspiratory effort. If the inspiratory flow of a patient with heart failure is greater than the oxygen flow provided by COT, the air will dilute the oxygen concentration, leading to a lower FiO2. By providing high and constant oxygen flow, HFNC therapy overcomes this air dilution and maintains FiO2 as high as 100%.^[16]^ Moreover, the high flow through the HFNC can provide a small amount of positive pressure in the respiratory system. It was postulated that with the mouth closed and a maximum flow of 60 L/min, positive pressure generated by an HFNC may reach 5.6 cmH2O.^[17]^ The positive pressure within the nasopharyngeal space and thoracic cavity is helpful for recruiting the collapsed alveoli or increasing the lung volume. Thus, in patients with AHF, the application of HFNC therapy provides the same positive end-expiratory pressure effect as NIV and can increase cardiac output by reducing both preload and afterload, reduce intrapulmonary shunting, and decrease pulmonary edema.
Our study showed no difference in PaCO2 between HFNC therapy and COT because most of the patients did not have hypercapnic respiratory failure. Therefore, it was difficult to explore the effects of HFNC therapy on hypercapnic respiratory failure in these patients with AHF. In a recent prospective observational study conducted by Marjanovic et al,^[18]^ 27 patients with a discharge diagnosis of hypercapnic cardiogenic pulmonary edema were analyzed. The median decrease in PaCO2 from baseline to after 1 h of HFNC treatment was 7 mmHg (interquartile range, 4–11 mmHg; P = 0.002). However, because that was a small prospective observational study, further research is needed to assess HFNC therapy as a possible management strategy for AHF-induced acute hypercapnic respiratory failure.
Cardiogenic pulmonary edema caused by AHF is a well-established indication for NIV. NIV has been proven to improve gas exchange and reduce the need for intubation and mortality in patients with AHF. In contrast, there is limited evidence supporting the effectiveness of HFNC therapy in patients with AHF, and some conclusions are even controversial. Therefore, we collected evidence from RCTs to compare the effects of HFNC therapy vs. NIV in patients with AHF. Our study showed no significant difference in the intubation rate, RR, HR, SpO2, or ABG parameters between HFNC therapy and NIV, indicating that HFNC therapy is not inferior to NIV. This finding might be due to the following features of HFNC therapy. First, the high flow rate from the cannula against the expiratory airflow from the patient generates the positive end-expiratory pressure effect as discussed above. Second, community-acquired pneumonia is a frequent triggering factor for decompensation of chronic HF. Clearance of sputum and alveolar fluid is important for these patients with AHF. Compared with NIV, the heated, humidified flow provided by the HFNC system facilitates easier clearance of secretions. Third, HFNC therapy has higher patient compliance. Unlike the intermittent application of NIV, most patients with AHF use HFNC therapy continuously, and this feature ensures a sufficient treatment time. We cannot exclude the fact that NIV is still an effective oxygen therapy method for respiratory failure in patients with AHF. However, from the perspective of preventing intubation, ease of application, and oxygenation efficiency, we consider that HFNC therapy may be another alternative strategy for patients with AHF.
Some limitations of our study need to be noted. First, blinding of the participants and personnel was impossible in all of the included RCTs, which might have led to performance bias. Second, the parameters of HFNC therapy ranged so extensively that we could not perform a subgroup analysis to investigate the influence of the flow rate and FiO2 on the effects of HFNC therapy. Third, the variable duration of oxygen support methods and significant heterogeneity of some outcomes might make it difficult to reach definitive conclusions regarding the effects of HFNC therapy in patients with AHF.
In conclusion, our meta-analysis showed that compared with COT, HFNC therapy in patients with AHF can not only reduce the intubation rate but also improve oxygenation, RR, and HR. HFNC therapy might serve as an alternative to NIV in terms of respiratory support in patients with AHF. More RCTs are required to fully assess the application of HFNC therapy in patients with hypercapnic respiratory failure specifically due to AHF.
None.