114 Original article
Treatment algorithm reduces oxygen use in the Emergency Department Robert Sieber and Joseph Osterwalder Objectives Evaluation of an O2 treatment algorithm on the basis of current recommendations to reduce the number of patients unnecessarily treated with O2 in the Emergency Department of a tertiary hospital compared with the traditional application, and analysis of the use of O2. Design This was a single-centre cohort study with 4 weeks of observation before and after the introduction of an O2 treatment algorithm. The main outcome measures were the proportion of patients treated with O2, distribution of indications for O2 therapy and occurrence of hypoxia and hypercapnia as undesired effects. Results The 4-week period before the intervention included 2190 patients and the 4-week period after the intervention included 2122 patients. The indications for O2 therapy were very similar in the two periods. After the intervention, the proportion of patients with supplemental O2 therapy was reduced from 11% (246) to 9% (182) (P < 0.003), a relative decrease of 18%. Undesired effects
Introduction After the harmful effects of hypoxaemia had been identified, high-concentration O2 therapy established itself in the 1970s as a frequent treatment for patients with cardiorespiratory failure and myocardial infarction [1,2]. This development was encouraged by simple methods of measuring oxygen saturation (SpO2) with continuous monitoring. As a consequence, O2 deficiency was increasingly documented. In the Emergency Department (ED), the result of this was the unrestricted use of O2 therapy for all forms of respiratory disturbance and low saturation values of indeterminate cause. In some cases, treatment guidelines have been developed [3]. However, in most hospitals, and particularly in the ED, specific treatment guidelines for the therapeutic use of O2 are not available. Our ED also had no formal recommendations for the use of O2 until recently. On the basis of different publications indicating potentially negative effects of hyperoxaemia, there has been increasing criticism of the unrestricted use of O2 therapy over the past few years, and it is still controversial today [4–14]. In 2008, this led to the first publication proposing comprehensive guidelines for the use of O2 in acute patients [15]. The guidelines treated the use of O2 as a drug in that it had to be specifically prescribed and 0969-9546 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
were similar in nature and frequency in the two observation periods. Conclusion The introduction of an O2 treatment algorithm in the Emergency Department reduced the number of patients treated with O2, and the nature and frequency of undesired effects was similar before and after the intervention. European Journal of Emergency Medicine 23:114–118 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. European Journal of Emergency Medicine 2016, 23:114–118 Keywords: Emergency Department, oxygen, respiratory, treatment algorithm Emergency Department, Cantonal Hospital St Gallen, St Gallen, Switzerland Correspondence to Robert Sieber, MD (RS), Emergency Department, Cantonal Hospital St Gallen, Rorschacherstrasse 95 9007, St Gallen, Switzerland Tel: + 41 714 943 632; fax: + 41 714 942 870; e-mail: [email protected] Received 6 October 2014 Accepted 22 December 2014
continuously monitored [16,17]. To our knowledge, these general guidelines have never been evaluated. We also do not know how frequently O2 therapy is administered to unselected emergency patients and little attention has been paid so far to the considerable expense of O2 therapy. We therefore decided to evaluate the efficacy and safety aspects of an O2 treatment algorithm on the basis of the current recommendation, and at the same time determine the indications for and the proportion of emergency patients administered O2 therapy in our ED.
Methods Study design
The indications for and use of O2 before and after the implementation of a staged O2 treatment algorithm into our ED were compared in this prospective, single-centre intervention study. We expected a 20% reduction (from 15 to 12%) in the proportion of patients treated with O2 after the introduction of the algorithm and expected no changes in the frequency or the nature of undesired effects. We also aimed to observe the proportion of admissions to our ED administered O2 therapy. Setting
The study was carried out in the ED of the Cantonal Hospital St Gallen, Switzerland. This is an urban, DOI: 10.1097/MEJ.0000000000000243
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Oxygen reduction due to algorithm Sieber and Osterwalder 115
700-bed, primary and tertiary care centre with an annual incidence of 34 000 ED visits. This is the only ED in the city serving a catchment area of about 120 000 (primary care) and one million (tertiary care). Participants and sample size
We included all adult patients older than 16 years presenting to the ED during two 4-week periods (3 May 2012–30 May 2012; 6 September 2012–3 October 2012). The two periods were seasonally matched to avoid climatic differences and extremes and to minimize differences in holiday behaviour so that a similar incidence of ED visits was expected. Patients referred to another hospital department after triage were excluded (e.g. gynaecology and obstetrics, ophthalmology and paediatrics). On the basis of a P-value of less than 0.05 and a power of 0.8, we calculated that a sample size of 2029 patients in the preintervention and postintervention groups would be required to detect a statistically significant reduction in the proportion of patients receiving O2 therapy from 15 to 12%. Intervention
O2 therapy was administered in the traditional, unregulated manner in the first 4-week observation period before the intervention. During this period, the ED staff were not aware of the purpose and scope of the study. On the basis of the most recent recommendations in the literature, an O2 treatment algorithm was then introduced (Fig. 1) [15–17]. After specific training, the nursing staff were to use the algorithm as a basis for independently deciding when to administer O2 therapy, as an activity delegated by the ED physician, who were also given a general introduction to the algorithm. All staff were given pocket cards of the algorithm, which was also posted in all treatment stations. The algorithm was used for an introductory period of 2 months before study documentation started in the second observational phase. It was possible to deviate from the recommendations of the algorithm if clinical circumstances demanded (e.g. cardiopulmonary instability, intubated patients). The prehospital treatment measures of the emergency services did not differ between the two observation periods. The local ethics committee approved the study (EKSG 12/018). Endpoints
The primary endpoints were the number of patients who had started, finished or were on continued O2 therapy, and the SpO2 level at the beginning and at the end of O2 therapy. The secondary endpoints were the frequency of hypercapnia und hypoxaemia requiring countermeasures and mortality during the first 24 h.
Data collection
The O2 therapy in each patient was documented prospectively on standardized case report forms by registered nurses. The following were documented: start and end time and date of O2 therapy, SpO2 at the beginning and end of O2 therapy, the indication for O2 therapy and any undesired effects (hypercapnia and hypoxia below the target range resulting in a change in O2 supply). The anonymized data were then entered into an Excel sheet by an independent data manager. Data analysis
Systat 13 (Systat Software Inc., Chicago, Illinois, USA) was used for power calculations. Data analyses were carried out by the χ2-test using Epi Info (version 3.4.1; Centre for Disease Control and Prevention, Atlanta, Georgia, USA).
Results In the 4-week period before the intervention, we conducted 2196 consultations and 246 (11%) of these patients received O2 therapy. The corresponding figures for the 4-week period after the intervention were 2122 consultations with 182 (9%) patients with O2 therapy. There was therefore a relative reduction of 18% (P < 0.003) (Table 1). The O2 therapy had been started by the emergency services in 123 (50%) patients before the intervention and 114 (62%) patients afterwards (P = 0.72). The O2 therapy was stopped during the stay in the ED in 56 (23%) patients in the first period and in 48 (26%) patients in the second period (P = 0.52). In the period before the intervention, 190 (77%) patients were discharged from the ED on O2 therapy; the corresponding figure for the second period was 134 (74%) patients (P < 0.003) (Table 2). The SpO2 in patients on admission (P = 0.28) and discharge (P = 0.47) did not differ significantly between the two periods (Table 3). We also found similar baseline characteristics in both groups (age: P = 0.90; sex: P = 0.55; triage categories: P = 0.97). The indications for O2 therapy were the same in both periods (P = 0.26; Table 4). The most frequent indications were lung conditions in 72 (29%) patients in the first period and 61 (34%) patients in the second period, with little difference between the numbers of patients with COPD in the two periods [38 (15%) patients; 32 (18%) patients]. Cardiac problems were the second most frequent indication in 49 (20%) patients in the first period and 34 (19%) in the second period. Trauma was the third most common in 47 (19%) patients before the intervention and 42 (23%) patients afterwards. The only indication with a marked difference between
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
116 European Journal of Emergency Medicine 2016, Vol 23 No 2
Fig. 1
ED patient
Trauma / shock ACS / stroke?
If SpO2
Treatment algorithm reduces oxygen use in the Emergency Department Robert Sieber and Joseph Osterwalder Objectives Evaluation of an O2 treatment algorithm on the basis of current recommendations to reduce the number of patients unnecessarily treated with O2 in the Emergency Department of a tertiary hospital compared with the traditional application, and analysis of the use of O2. Design This was a single-centre cohort study with 4 weeks of observation before and after the introduction of an O2 treatment algorithm. The main outcome measures were the proportion of patients treated with O2, distribution of indications for O2 therapy and occurrence of hypoxia and hypercapnia as undesired effects. Results The 4-week period before the intervention included 2190 patients and the 4-week period after the intervention included 2122 patients. The indications for O2 therapy were very similar in the two periods. After the intervention, the proportion of patients with supplemental O2 therapy was reduced from 11% (246) to 9% (182) (P < 0.003), a relative decrease of 18%. Undesired effects
Introduction After the harmful effects of hypoxaemia had been identified, high-concentration O2 therapy established itself in the 1970s as a frequent treatment for patients with cardiorespiratory failure and myocardial infarction [1,2]. This development was encouraged by simple methods of measuring oxygen saturation (SpO2) with continuous monitoring. As a consequence, O2 deficiency was increasingly documented. In the Emergency Department (ED), the result of this was the unrestricted use of O2 therapy for all forms of respiratory disturbance and low saturation values of indeterminate cause. In some cases, treatment guidelines have been developed [3]. However, in most hospitals, and particularly in the ED, specific treatment guidelines for the therapeutic use of O2 are not available. Our ED also had no formal recommendations for the use of O2 until recently. On the basis of different publications indicating potentially negative effects of hyperoxaemia, there has been increasing criticism of the unrestricted use of O2 therapy over the past few years, and it is still controversial today [4–14]. In 2008, this led to the first publication proposing comprehensive guidelines for the use of O2 in acute patients [15]. The guidelines treated the use of O2 as a drug in that it had to be specifically prescribed and 0969-9546 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
were similar in nature and frequency in the two observation periods. Conclusion The introduction of an O2 treatment algorithm in the Emergency Department reduced the number of patients treated with O2, and the nature and frequency of undesired effects was similar before and after the intervention. European Journal of Emergency Medicine 23:114–118 Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. European Journal of Emergency Medicine 2016, 23:114–118 Keywords: Emergency Department, oxygen, respiratory, treatment algorithm Emergency Department, Cantonal Hospital St Gallen, St Gallen, Switzerland Correspondence to Robert Sieber, MD (RS), Emergency Department, Cantonal Hospital St Gallen, Rorschacherstrasse 95 9007, St Gallen, Switzerland Tel: + 41 714 943 632; fax: + 41 714 942 870; e-mail: [email protected] Received 6 October 2014 Accepted 22 December 2014
continuously monitored [16,17]. To our knowledge, these general guidelines have never been evaluated. We also do not know how frequently O2 therapy is administered to unselected emergency patients and little attention has been paid so far to the considerable expense of O2 therapy. We therefore decided to evaluate the efficacy and safety aspects of an O2 treatment algorithm on the basis of the current recommendation, and at the same time determine the indications for and the proportion of emergency patients administered O2 therapy in our ED.
Methods Study design
The indications for and use of O2 before and after the implementation of a staged O2 treatment algorithm into our ED were compared in this prospective, single-centre intervention study. We expected a 20% reduction (from 15 to 12%) in the proportion of patients treated with O2 after the introduction of the algorithm and expected no changes in the frequency or the nature of undesired effects. We also aimed to observe the proportion of admissions to our ED administered O2 therapy. Setting
The study was carried out in the ED of the Cantonal Hospital St Gallen, Switzerland. This is an urban, DOI: 10.1097/MEJ.0000000000000243
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Oxygen reduction due to algorithm Sieber and Osterwalder 115
700-bed, primary and tertiary care centre with an annual incidence of 34 000 ED visits. This is the only ED in the city serving a catchment area of about 120 000 (primary care) and one million (tertiary care). Participants and sample size
We included all adult patients older than 16 years presenting to the ED during two 4-week periods (3 May 2012–30 May 2012; 6 September 2012–3 October 2012). The two periods were seasonally matched to avoid climatic differences and extremes and to minimize differences in holiday behaviour so that a similar incidence of ED visits was expected. Patients referred to another hospital department after triage were excluded (e.g. gynaecology and obstetrics, ophthalmology and paediatrics). On the basis of a P-value of less than 0.05 and a power of 0.8, we calculated that a sample size of 2029 patients in the preintervention and postintervention groups would be required to detect a statistically significant reduction in the proportion of patients receiving O2 therapy from 15 to 12%. Intervention
O2 therapy was administered in the traditional, unregulated manner in the first 4-week observation period before the intervention. During this period, the ED staff were not aware of the purpose and scope of the study. On the basis of the most recent recommendations in the literature, an O2 treatment algorithm was then introduced (Fig. 1) [15–17]. After specific training, the nursing staff were to use the algorithm as a basis for independently deciding when to administer O2 therapy, as an activity delegated by the ED physician, who were also given a general introduction to the algorithm. All staff were given pocket cards of the algorithm, which was also posted in all treatment stations. The algorithm was used for an introductory period of 2 months before study documentation started in the second observational phase. It was possible to deviate from the recommendations of the algorithm if clinical circumstances demanded (e.g. cardiopulmonary instability, intubated patients). The prehospital treatment measures of the emergency services did not differ between the two observation periods. The local ethics committee approved the study (EKSG 12/018). Endpoints
The primary endpoints were the number of patients who had started, finished or were on continued O2 therapy, and the SpO2 level at the beginning and at the end of O2 therapy. The secondary endpoints were the frequency of hypercapnia und hypoxaemia requiring countermeasures and mortality during the first 24 h.
Data collection
The O2 therapy in each patient was documented prospectively on standardized case report forms by registered nurses. The following were documented: start and end time and date of O2 therapy, SpO2 at the beginning and end of O2 therapy, the indication for O2 therapy and any undesired effects (hypercapnia and hypoxia below the target range resulting in a change in O2 supply). The anonymized data were then entered into an Excel sheet by an independent data manager. Data analysis
Systat 13 (Systat Software Inc., Chicago, Illinois, USA) was used for power calculations. Data analyses were carried out by the χ2-test using Epi Info (version 3.4.1; Centre for Disease Control and Prevention, Atlanta, Georgia, USA).
Results In the 4-week period before the intervention, we conducted 2196 consultations and 246 (11%) of these patients received O2 therapy. The corresponding figures for the 4-week period after the intervention were 2122 consultations with 182 (9%) patients with O2 therapy. There was therefore a relative reduction of 18% (P < 0.003) (Table 1). The O2 therapy had been started by the emergency services in 123 (50%) patients before the intervention and 114 (62%) patients afterwards (P = 0.72). The O2 therapy was stopped during the stay in the ED in 56 (23%) patients in the first period and in 48 (26%) patients in the second period (P = 0.52). In the period before the intervention, 190 (77%) patients were discharged from the ED on O2 therapy; the corresponding figure for the second period was 134 (74%) patients (P < 0.003) (Table 2). The SpO2 in patients on admission (P = 0.28) and discharge (P = 0.47) did not differ significantly between the two periods (Table 3). We also found similar baseline characteristics in both groups (age: P = 0.90; sex: P = 0.55; triage categories: P = 0.97). The indications for O2 therapy were the same in both periods (P = 0.26; Table 4). The most frequent indications were lung conditions in 72 (29%) patients in the first period and 61 (34%) patients in the second period, with little difference between the numbers of patients with COPD in the two periods [38 (15%) patients; 32 (18%) patients]. Cardiac problems were the second most frequent indication in 49 (20%) patients in the first period and 34 (19%) in the second period. Trauma was the third most common in 47 (19%) patients before the intervention and 42 (23%) patients afterwards. The only indication with a marked difference between
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
116 European Journal of Emergency Medicine 2016, Vol 23 No 2
Fig. 1
ED patient
Trauma / shock ACS / stroke?
If SpO2
