Article Text
PDF
Abstract
Background Non-invasive ventilation (NIV) is increasingly being implemented by many ambulance jurisdictions as a standard of care in the out-of-hospital management of acute cardiogenic pulmonary oedema (ACPO). This implementation appears to be based on the body of evidence from the emergency department (ED) setting, with the assumption that earlier administration by paramedics would give benefits with regard to inhospital mortality and the rate of endotracheal intubation beyond those seen when initiated in the ED. This paper sought to identify and review the current level of evidence supporting NIV in the prehospital setting.
Methods Electronic searches of Medline, EMBASE, CINAHL, Cochrane Database of Systematic Reviews and Cochrane Database of Controlled Trials were conducted and reference lists of relevant articles were hand searched.
Results The search identified 12 primary studies documenting the use of NIV, either continuous positive airway pressure or bi-level non-invasive ventilation, for ACPO in the out-of-hospital setting. Only three studies were randomised controlled trials, with none addressing inhospital mortality as a primary outcome measure. The majority of articles were non-comparative descriptive studies.
Conclusion Early prehospital NIV appears to be a safe and feasible therapy that results in faster improvement in physiological status and may decrease the need for intubation when compared with delayed administration in the ED. There is weak evidence that is may decrease mortality. The cost versus benefit equation of system-wide prehospital implementation of NIV is unclear and, based on the current evidence, should be considered with caution.
- Cardiac care
- continuous positive airway pressure
- emergency ambulance systems
- emergency medical services
- heart failure
- non-invasive ventilation
- nursing
- out-of-hospital
- prehospital
- pulmonary oedema
- ventilation
Statistics from Altmetric.com
- Cardiac care
- continuous positive airway pressure
- emergency ambulance systems
- emergency medical services
- heart failure
- non-invasive ventilation
- nursing
- out-of-hospital
- prehospital
- pulmonary oedema
- ventilation
Acute cardiogenic pulmonary oedema (ACPO) is a common pathology among patients presenting to emergency medical services (EMS) with acute breathing difficulty. Despite significant advances in hospital-based care, prehospital emergency management of ACPO has until quite recently remained largely unchanged, with standard care continuing to be structured around loop diuretic therapy, sublingual nitrates and either high-concentration oxygen therapy or ventilatory assistance using intermittent positive pressure ventilation. In the emergency department (ED) and other acute care hospital settings, non-invasive ventilation (NIV) has for some time constituted a core component of the acute treatment regime for patients with ACPO who have adequate respiratory effort, and is well documented to improve patient outcomes.1 Based on this inhospital body of evidence an increasing number of EMS agencies have integrated or are considering introducing NIV into their standard care algorithms. This emergence has been aided partly by the emergence of smaller, more portable continuous positive airway pressure (CPAP) delivery systems that have increased the feasibility and practicality of providing this therapy in the prehospital setting. The intuitive assumption based on the strength of the ED-based evidence is that initiating early NIV before arrival at the ED would further improve upon the beneficial outcomes seen when NIV is implemented in the hospital. It is unclear though whether the presumed benefits of early prehospital NIV in relation to patient outcomes are considerable enough to justify the significant organisational cost associated with system-wide implementation in EMS jurisdictions. The impact of out-of-hospital NIV on significant patient outcomes such as reduction in mortality and mechanical ventilation with intubation, when implemented less than 30 min earlier in the patient's journey, is not well described.
Objectives
The aims of this review are to identify primary studies and systematic reviews describing the use of prehospital NIV for patients with ACPO, and to determine the effect of prehospital NIV on mortality, the rate of intubation, physiological variables and length of stay (LOS).
Methods
An extensive search of major electronic databases was conducted by a researcher with formal training in electronic literature searching. The search included the following databases: Cochrane Database of Systematic Reviews (to 3 Quarter 2009); Cochrane Database of Controlled Trials (to third quarter 2009); Medline (1950 to September week 3 2009); CINHAL; and EMBASE (1980 to 2009 week 39). Keywords ‘out-of-hospital’, ‘prehospital’, ‘emergency medical services’, ‘paramedic’, ‘emergency care’, ‘CPAP’, ‘BIPAP ‘continuous positive airway pressure’, ‘NIV’, ‘non-invasive ventilation’, ‘positive pressure ventilation’, ‘pulmonary oedema’, ‘heart failure’, ‘lung edema’ and ‘respiratory distress’ were used.
As ‘NIV’ is an umbrella term for a variety of non-invasive modalities with varying terminologies, we included all studies that reported the use of CPAP, bi-level non-invasive ventilation (bi-level NIV), bi-level positive airway pressure, non-invasive pressure support ventilation and non-invasive positive pressure ventilation.
We limited our search to published studies and those with English language abstracts. The reference lists of suitable articles were hand-searched to identify any articles that were not found in the electronic search. We excluded non-systematic reviews or commentaries.
Results
The search strategy located a total of 87 articles, of which 23 were relevant to the use of NIV in the prehospital setting. After excluding non-systematic reviews and commentaries there were only 12 primary research articles relating to the use of NIV by paramedics in the prehospital setting, the characteristics of which are detailed in table 1. Of these, only three studies used a randomised controlled trial (RCT) methodology.2–4 Two papers reported non-randomised cohort studies5 6 and five reported non-comparative case series.7–12 One article reported on cost-effectiveness,13 while another described the viability of NIV as a prehospital treatment option.14 The majority of studies used surrogate outcomes such as oxygenation, patient-reported dyspnoea score or other clinical measures as the primary outcome; only one RCT3 was powered on the rate of intubation, while none were powered to detect differences in mortality as the primary outcome.
Summary of literature investigating the prehospital use of NIV for ACPO
Synthesis of evidence
Mortality
Reducing mortality in patients with ACPO is perhaps the most important outcome to consider when investigating the merits of the prehospital administration of NIV. There is some evidence that prehospital NIV reduces mortality compared with standard care regimes for ACPO; however, none of the current evidence arises from studies adequately powered to detect differences in mortality as a primary outcome. Plaisance et al4 conducted a prehospital RCT of 124 patients with ACPO comparing the immediate administration of CPAP with delayed (15 min) in a two-tiered physician-led advanced life support prehospital system in Europe. A subgroup analysis demonstrated weaker evidence of a lower inhospital mortality rate in the group assigned to receive immediate CPAP (2% vs 13%; OR 0.22; 95% CI 0.04 to −1.00; p=0.05), suggesting that a delay of 15 min in commencement of CPAP impacts on short-term mortality. Thompson et al3 reported a secondary outcome analysis of 17 patients, in which inhospital mortality was lower in patients who received prehospital CPAP compared with standard care (14% vs 35%; OR 0.3; 95% CI 0.09 to −0.99; p<0.05) in a North American EMS jurisdiction. It should be noted that the population for this study included not only those with ACPO but also all acute respiratory distress in adults, including pneumonia, chronic obstructive pulmonary disease and asthma. The majority of patients (66%) in the final analysis were, however, ultimately confirmed as having ACPO. The RCT by Weitz et al2 investigated the impact of non-invasive pressure support ventilation (also known as non-invasive positive pressure ventilation or bi-level positive airway pressure) in addition to standard care compared with standard care alone on ACPO in a small study of 23 patients in a German EMS jurisdiction. Subgroup analysis of inhospital mortality in that study identified no difference, with only one patient in each group dying (1/13 (8%) vs 1/10 (10%)). A suggestion of mortality benefit has also been reported in non-randomised comparative studies. A prospective cohort study by Hubble et al6 reported a significant difference in mortality favouring CPAP with standard care compared with standard care alone (5% vs 23%; p<0.001) for the 215 patients with a field diagnosis of ACPO. Approximately 24% of enrolled patients ultimately received a diagnosis other than ACPO; the investigators therefore diligently conducted an analysis of the 163 patients with confirmed ACPO and found similar results favouring the addition of CPAP to standard care (6% vs 25%; p<0.001). The study design compared outcomes from two adjacent ambulance services, one of which administered standard care plus CPAP, while the other administered standard care alone. Although the authors state that the participating services are homogenous in nature, the result could be confounded by organisational, clinical or cultural heterogeneity that makes direct comparison difficult.
Rate of intubation
Reducing the need for endotracheal intubation is an important goal in the management of acute cariogenic pulmonary oedema. There is good evidence that early initiation of NIV in the prehospital setting reduces the rate of intubation. Thompson et al3 powered their RCT on the rate of intubation as the primary outcome, and found that the odds of being intubated were 75% less for patients who received CPAP compared with those who received standard care alone (OR 0.25; 95% CI 0.09 to −0.73). In the control group, 50% were intubated compared with 20% in the CPAP group. The high intubation rates in both groups may reflect the severity of the patients enrolled in the study, as patients were required to have severe respiratory distress with failing respiratory efforts. CPAP is indicated in less severe patients with ACPO, so the study population may not be generalisable to less severe cases; anecdote suggests that the frequency of severe cases of ACPO is decreasing as a result of improved primary care management. The 50% intubation rate in the control group may also reflect the elective intubation system in which the study was conducted, therefore the transferability of these results to a non-elective intubation emergency medical system is arguable. The French RCT by Plaisance et al4 also reported reduced intubation in the early CPAP group compared with the control group who received it later, with six out of 63 (9.5%) intubated in the early group compared with 16 out of 61(26%) in the delayed group (OR 0.30; 95% CI 0.09 to −0.89; p=0.01). Results from a cohort study by Hubble et al6 were consistent with the randomised trials described above, with standard care (no CPAP) patients significantly more likely to be intubated (OR 4.21; 95% CI 1.09 to −16.28).
Physiological variables
There is strong evidence that prehospital administration of NIV improves the clinical presentation of patients with ACPO on arrival in the ED. Patients appear to benefit in the short to medium term of their journey through achieving significant relief of acute symptoms such as self-reported dyspnoea,4 6 and significant improvements in physiological parameters including heart rate, respiratory rate, blood pressure, oxygen saturation2 5 6 11 12 and blood chemistry.9
Length of stay (hospital/intensive care unit)
The effect of prehospital CPAP on LOS in hospital at any level of care is not well described, with only one study reporting outcomes of this kind.3 No significant differences were found in LOS in either critical care or hospital overall, although the number of patients in this secondary analysis are too low to be considered meaningful (n=16). LOS in hospital is an important consideration in the assessment of CPAP from the prehospital perspective; however, it is an outcome that is difficult to measure due to the many non-clinical factors that contribute to how long the patient stays. Staffing in the ED, bed availability in wards and the intensive care unit and ED congestion all have the potential to confound this outcome.
Discussion
The current body of prehospital evidence appears to be of insufficient strength to warrant a definitive recommendation supporting the inclusion of NIV into out-of-hospital standard of care for patients with ACPO. Based on the level and strength of evidence presented in this review, such a recommendation could be considered premature.
It is well established that NIV, especially CPAP, is a therapy with confirmed efficacy when administered to patients with ACPO in the ED, producing significant reductions in the need for intubation and less convincingly for reducing mortality.1 15–17 The research question requiring investigation is probably not therefore about the effectiveness of NIV itself, but rather about whether receiving NIV in the field typically less than 30 min earlier than in the ED results in any significantly greater benefits in relation to non-surrogate outcomes such as mortality and intubation, along with the health economics linked to the length of hospital stay and hospital resource utilisation. A limitation of the current body of prehospital research is a poor account of how many patients in the various control groups received NIV as part of ‘standard care’ after arriving at the ED. If we assume that NIV is a component of standard care for ACPO, as is the case in most Australian ED, the question is about early versus late NIV, rather than NIV versus no NIV. Currently, the study by Plaisance et al4 provides the best evidence of a benefit from ‘early’ administration, with a delay of 15 min influencing outcomes significantly. In that study all patients from both groups received CPAP during the study period, but during different periods.
The implementation of CPAP as a prehospital standard of care could come at significant financial burden, particularly for large EMS agencies, which would require hundreds of NIV units, ongoing costs of consumables, and an enormous commitment to training before the implementation of this intervention. This outlay could of course be balanced by the potential cost benefits arising from decreased LOS and reduced cost and burden of broader ongoing care. Hubble et al13 created a health economic model evaluating the implementation of CPAP in their EMS system based on their earlier study from 2006, reporting a cost-effective result. It is important to note that this health economic evaluation was based on an absolute risk reduction of 18% for mortality and 16% for intubation, and that the EMS agency was a small system consisting of only 14 ambulances. While organisational heterogeneity means the economics of the study are not easily applied to other EMS systems, it demonstrates that the cost of implementation in that jurisdiction was not cost prohibitive and can in fact be cost-effective.
A well-designed prospective RCT with adequate statistical power to detect a clinically meaningful reduction in mortality is required. Such a trial should also collect meaningful secondary outcome data on not only clinical outcomes and physiological variables but also non-clinical measures such as LOS, resource utilisation and cost-effectiveness.
Limitations
This narrative review has several limitations. This review of the evidence is not entirely systematic and there may be gaps in the included studies. For pragmatic reasons we excluded articles that were not in English and we did not seek to identify ‘grey’ literature (eg, unpublished studies). Our objective was to identify and discuss the evidence for prehospital NIV and assess the impact of this intervention on various aspects of prehospital practice. A systematic review of only controlled trials of prehospital NIV would have excluded most studies and precluded discussion of the domains of interest.
Conclusion
There is some evidence that receiving NIV for ACPO in the prehospital setting results in significant improvements in short-term mortality. There is good evidence that NIV appears to decrease the need for endotracheal intubation. Early NIV before arrival at the ED does improve acute symptoms in patients with ACPO predominantly in regard to decreases in patient-reported dyspnoea and improvements in vital signs. A large RCT powered to detect a meaningful difference in mortality is required to realise fully the potential benefits of the early administration of NIV in the prehospital setting and to inform clinical policy.
Acknowledgments
The authors would like to thank Associate Professor Paul Middleton, Mark Goodger and Monika Sitkowski from the Ambulance Research Institute for reviewing the paper and providing valuable advice.
References
Footnotes
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Primary survey