The benefit of continuous positive airway pressure (CPAP) applied by face mask as an adjuvant treatment for acute cardiogenic pulmonary oedema (APCE) is well established.1 Several studies indicate that early administration of CPAP in ACPE reduces the number of endotracheal intubations and shortens intensive care unit (ICU), coronary care unit (CCU) and hospital length of stay.1^–4 The effectiveness of CPAP has also been demonstrated in prehospital settings when applied by attending emergency physicians.5 6 In the Netherlands the ambulance service network is manned by well-trained nurses, not by physicians. These are all registered nurses, often specialised in intensive care, emergency care or anaesthesiology, with additional ambulance training for 1 year. The ambulance nurses are extensively trained in diagnosing and initiating treatment for different conditions that are set out in protocols.

The application of CPAP by nurses in ambulances has not previously been described. The administration of CPAP has been proved to be effective with the Boussignac CPAP (BCPAP) system in the treatment of ACPE in the emergency department.7^–10 Given the experiences with non-physicians on ambulances in identifying ST elevation myocardial infarction (STEMI), we hypothesised that diagnosis of ACPE could also be made by ambulance nurses.11 We also hypothesised that BCPAP can be used by nurses in the ambulance and would result in a rapid improvement in oxygen saturation.

Our aims were to demonstrate that ACPE can be identified by trained nurses and to show that BCPAP results in rapid improvement in oxygenation during emergency transportation by ambulance.

METHODS

This study on a prospective case series was performed by the Department of Cardiology of the University Medical Center Groningen (UMCG) and AmbulanceZorg Groningen in the setting of a regional cooperation between the emergency departments and departments of cardiology in six hospitals in our region.

Before BCPAP was implemented as the standard treatment for patients with ACPE on the ambulance, all ambulance nurses received training for 1 h on applying BCPAP on the ambulance by means of a lecture and an e-learning program (which can be viewed at www.bcpap.umcg.nl). All 33 ambulances in the region (AmbulanceZorg Groningen) were equipped with the disposable BCPAP system, with four different sizes of face mask and a BCPAP instruction manual. The BCPAP system (Vygon, 95440 Ecouen, France) is a light-weight (10 g) FDA- and CE-approved disposable cylindrical plastic device. Oxygen is supplied by means of an integrated oxygen tube 2 m in length connected to an oxygen cylinder. A jet flow of oxygen that accelerates to nearly the speed of sound through four microchannels generates a flow-dependent pressure in the BCPAP system.9^–12 The BCPAP system was directly connected to a small mobile 2-litre oxygen cylinder (Linde Gas Therapeutics Benelux, Eindhoven, The Netherlands) pressurised at 200 bar. Once in the ambulance, the BCPAP system was connected to larger oxygen cylinders (2×5 l) pressurised at 200 bar. Depending of the oxygen flow administered (5–20 l/min), a CPAP of 2–8 cm H₂O was created.

Patients with presumed ACPE were immediately treated with BCPAP. Presumed ACPE was defined as acute dyspnoea associated with suspected heart disease. When the ambulance nurse triaged a patient for treatment with BCPAP, the severity of presumed ACPE and the respiratory condition including peripheral oxygen saturation (Spo₂) were carefully considered. Criteria for initiation of BCPAP were defined as presumed ACPE with a respiratory rate of >25 breaths/min and an Spo₂<95% while receiving oxygen.4 5 8^–10 13 14 If deemed necessary, treatment with the disposable BCPAP system could be continued without interruption in the emergency department, the CCU or the ICU. In addition to BCPAP, patients were also treated with 80 mg frusemide intravenously and 0.4 mg nitroglycerin sublingually, provided the systolic blood pressure was >90 mm Hg. When the clinical response of the patient was deemed insufficient, nitroglycerin was repeated together with 2 mg morphine intravenously.

Patient demographic data, medical history and physiological information from the ambulance monitoring system as well as data from hospital admission and subsequent outcome and long-term survival were retrieved. For this purpose the investigators visited all hospitals.

For all patients treated with BCPAP the diagnosis of presumed ACPE was verified by the hospital physicians and/or study investigators (EEMW, ICCvdH and MWNN) including a cardiologist. ACPE was diagnosed by a combination of physical examination (pulmonary rales, third heart sound, jugular venous distension), chest radiography (signs of congestion) and other available data including clinical history, echocardiography and natriuretic peptides.

Some of the surviving patients were queried about their experiences by one of the investigators (WD) with three questions:

• Can you remember receiving treatment with a special oxygen mask during ambulance transport? (yes/no)

• Directly after the start of this treatment with an oxygen mask, how did you feel? (better/the same/worse/don’t know)

• How did you experience the “comfort” of the mask? (good/reasonable/poor/don’t know)

Eight months after introduction we anonymously queried the ambulance nurses about their experiences with respect to BCPAP treatment in patients with ACPE with the following questions:

• Did you use BCPAP? (yes/no)

• Do you think BCPAP is useful in the ambulance? (yes/no)

• Treatment with BCPAP for ACPE is time consuming (agree fully/partially agree/no opinion/disagree/fully disagree)

• Any comments with regard to BCPAP in the ambulance (open question)

Data analysis

Data are presented as medians and associated interquartile ranges (IQR) or means with standard deviation (SD) for continuous variables, or as group percentages for categorical variables. All statistical analyses were performed using commercially available software (SPSS Version 14.0; SPSS, Chicago, Illinois, USA).

RESULTS

From March to December 2006, 32 patients (15 women) of median age 82 years (IQR 73–86) were identified as having ACPE and received treatment with BCPAP according to the protocol (table 1). The patients were transported to one of six hospitals in the region. Baseline clinical variables at admission are shown in table 1.

The prevalence of pre-existing cardiovascular disease and comorbidity was considerable in these patients (table 2).

A diagnosis of ACPE was confirmed in 26 patients (81%). In the other six patients (19%), three (9%) were diagnosed with an exacerbation of chronic obstructive pulmonary disease, two (6%) had pneumonia and one (3%) intracranial haemorrhage. Three patients (all with ACPE) were admitted to the ICU; intubation was needed in two patients and one was treated with non-invasive bilevel positive pressure ventilation.

After initiation of BCPAP, median (IQR) Spo₂ increased from 79% (69–94%) to 96% (89–98%) within 20 min (fig 1, table 1). Mean (SD) arterial pressure decreased from 124 (30) mm Hg to 108 (21) mm Hg after 20 min (table 1). The mean heart rate in the same interval did not change, with an initial heart rate of 98 beats/min.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Changes in median peripheral oxygen saturation (Spo₂) before treatment and after 10 and 20 min of treatment with Boussignac continuous positive airway pressure (BCPAP) in 32 patients with acute cardiogenic pulmonary oedema.

One patient (3%) died in the ambulance, 10 patients (31%) died during their hospital stay and three patients (9%) died within a median observation period of 4 months (table 3). The cause of death was related to severe heart failure in all but one case.

Four patients with proven ACPE (13%) were interviewed about their experiences with BCPAP directly after treatment. Fourteen patients (44%) with proven ACPE were also questioned after a mean interval of 4 months about their experiences with BCPAP. None of the patients had a negative recollection of the treatment. Eleven patients (61%) could not remember specific details of the treatment. The other seven patients (39%) reported that the BCPAP treatment immediately reduced dyspnoea. No discomfort was reported by any of the patients.

Twenty-two ambulance nurses were questioned 0–8 months after the use of BCPAP; 20 (91%) were positive about the practical feasibility and medical effectiveness of BCPAP treatment for ACPE. No technical or logistical problems were reported.

DISCUSSION

To our knowledge, this is the first study in which BCPAP has been used in ambulances by nurses for the treatment of patients with presumed ACPE. The study shows that BCPAP could be used on the ambulance by trained ambulance nurses. They can recognise ACPE equally as well as physicians who correctly diagnosed ACPE in 77% of cases.7 The treatment of ACPE with BCPAP on the ambulance has been described before but, in the previous study, the BCPAP was applied by physicians.7

The accuracy of the ambulance personnel in diagnosing patients with ACPE was good. Although 19% of the patients had a final diagnosis different from ACPE, treatment with BCPAP in these patients did not lead to deterioration in their condition. The skills in diagnosing ACPE may improve over time with more frequent use of BCPAP and by feedback from the physicians at the hospitals.

The treatment of ACPE with CPAP was first described by Barach in 1938.15 CPAP can increase lung volume, improve oxygenation and reduce the work of breathing.16 It can also reduce venous return, decreasing ventricular filling pressures and improve cardiac performance.17 18 Several studies indicate that CPAP in patients with acute hypoxaemia due to heart failure may decrease the need for intubation.1 19 20 Compared with oxygen therapy, treatment with CPAP is associated with significant improvement in the Pao₂/Fio₂ ratio, subjective dyspnoea score and respiratory and heart rates. CPAP produces a rapid physiological and symptomatic improvement in these patients, especially within the first hour. Interestingly, non-invasive ventilation is equivalent to CPAP alone.9 21 The use of CPAP as well as many other interventions performed by non-physicians on the ambulance, with the exception of thrombolysis in patients with STEMI,22 has been poorly investigated. We do not consider that acute myocardial infarction or shock is a reason to exclude patients with ACPE from treatment with BCPAP, provided that BCPAP treatment does delay definitive treatment for myocardial infarction or shock. Although ambulance personnel were instructed only to use BCPAP for patients with ACPE, we do not believe that the patients without ACPE were harmed by BCPAP. Previously, these patients who were also hypoxaemic would have received a high inspired fraction of oxygen through a non-rebreathing mask.

The rise in life expectancy and the improved survival after myocardial infarction have led to an increase in predominantly elderly patients with heart failure and ACPE. This is underscored by a mean age of 80 years in our patients. Our patients had a high in-hospital mortality rate as well as after discharge from hospital. Recent studies on similar patients with ACPE also found high hospital mortality rates varying from 27% to 46%.14 22 23

In our study we have also shown that Spo₂ in patients with hypoxaemia due to ACPE rapidly improved when treated with BCPAP. Similar benefits were reported with in-hospital treatment of patients with ACPE treated with CPAP from more sophisticated CPAP systems or mechanical ventilators.1 3 4 The utility of early aggressive therapy for ACPE is underscored by the results seen after introduction of a mobile coronary care unit in Sweden. With this mobile unit, CPAP was used in 91% of 158 patients. This resulted in marked clinical improvement upon hospital arrival in comparison with a control group of 158 patients who did not receive CPAP.24

Over the last 4 years we have also designated a pivotal role for ambulance nurses in diagnosing STEMI for immediate transport to our catheterisation laboratory for primary percuntaneous intervention (PCI). The feasibility, safety and speed of paramedic-referred primary PCI has recently been demonstrated in a Canadian study and was accurate in 95% of patients in a Dutch study.11 25 As with STEMI and primary PCI, the ambulance nurses were well motivated to diagnose ACPE and initiate treatment with BCPAP. Undoubtedly, an important motivation for nurses to apply BCPAP will be the rapid clinical improvement induced by this treatment in patients who are in severe distress.

Our study has a number of limitations. It is an observational study with a limited number of patients. This design does not allow us to draw conclusions on the favourable outcome with BCPAP over standard oxygen therapy. Recall bias may have affected the positive reactions of the ambulance nurses, since some nurses were queried 8 months after they used BCPAP. Nevertheless, we have shown that the BCPAP system is well suited for use in the ambulance. After admission to one of six hospitals, different treatment protocols may have been used and this might have affected the outcome. However, the primary outcomes were determined in the ambulance and not after admission. We hypothesise that it may even be possible to use the BCPAP system in air ambulances for patients with ACPE. The low weight and small size of the system is very relevant in this respect. Future studies with the BCPAP system are needed to confirm this hypothesis. We are currently in the process of implementing BCPAP in other ambulance regions, with the goal of using it across the whole of the Netherlands. With the introduction and more frequent use of BCPAP, the rate of correctly diagnosed ACPE might increase further.

CONCLUSION

This study shows that immediate treatment of ACPE with BCPAP indicated by and administrated by ambulance nurses was feasible and effective in improving oxygenation. We believe that BCPAP can be implemented in ambulances in countries with similar emergency systems for the treatment of patients with ACPE.