Uncategorized

Prehospital oxygen for stroke victims

Journal logoUnlabelled imageAmerican Journal of Emergency Medicine 40 (2021) 205-207

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage:

Prehospital oxygen for stroke victims

To the Editors,

It was of great interest to read Dr. Dylla and colleagues’ well-done retrospective cohort study investigating prehospital Oxygen treatment in acute stroke patients [1]. It was also very informative to read Dr. Zhang’s review [2] and the authors’ response [3].

Dr. Dylla and colleagues’ investigation is quite impressive. I would like to especially highlight and support the authors’ assertion of the high frequency of prehospital stroke encounters that result in supple- mental oxygen by sharing some unpublished data from my institution in central New Jersey [4]. A few years ago, concerned about the potential harm of hyperoxia in acute stroke and other Disease states [5-7], my co- investigators and I examined Emergency Medical Services’ (EMS) Prehospital use of oxygen with acute stroke patients to study how many of these patients might be at risk of hyperoxia.

We performed a retrospective review of a single Basic Life Support (BLS)-Advanced life support EMS system database of acute stroke patients transported to a single comprehensive stroke center in central New Jersey from 2010 to 2012 (full description of methods avail- able). We included patients N21 years old who were determined to be a “brain attack” by EMS providers with stroke symptoms b12 h from onset. We excluded patients b21 years old, pregnant women and cases with data missing from the EMS record that could not be ab- stracted from the subjects’ emergency department (ED) record. We col- lected the following data from the EMS Electronic Medical Record database: name, age, sex, date of service, time of symptom onset, level of provider training (BLS/ALS), oxygen therapy delivered (oxygen flow), oxygen saturation, heart rate, respiratory rate, blood pressure, and medications given by EMS. Our vital sign data from EMS database reflected patient vital signs upon arrival in the Emergency Department (ED).

Our final analysis included 395 cases (see Table 1) (full description of results available). We found oxygen flow levels were non-normally distributed and clustered around 0, 4, 10 and 15 L/min. So, for analysis purposes we divided subjects into four groups – zero, low, mid, and high according to oxygen flow delivered by EMS providers. The groups correlated to 0 L/min, 2-6 L/min, 10-15 L/min, 25-60 L/min respec- tively (see Table 2). Multinomial logistic regression was used to assess the effects of covariates on oxygen flow.

We found 116 (29%) of our subjects arrived in the ED with oxygen

saturations of 100% (see Table 3). Seventy-eight (67%) of these subjects

O2 flow (L/min)

O2 flow (L/min)

O2 sat (%)

N

ALS

BLS

received supplemental oxygen. There were 29 (25%) subjects treated

0

0

100

38

31

7

with low flow (2-6 L/min) oxygen, while another 48 (41%) subjects

2-6

2-6

100

29

28

1

We presented the results of our multinomial logistic regression for oxygen flow levels as odds ratios (Table 4). Gender, EMS level of training (BLS vs ALS), oxygen saturation, heart rate and respiratory rate were all significantly associated with oxygen flow rate. Females had lower odds of receiving the mid-level (10-12 L/min) of oxygen flow than no oxygen (0 L/min) compared to males (OR = 0.39, p = 0.01) but no different odds of receiving low (2-6 L/min) or high (15-60 L/min) flow levels of oxygen. For subjects transported by ALS, the odds of low oxygen

Table 1

Oxygen flow, age, gender and ems training distribution.

O2

Frequency

%

Age

Gender

EMS

training level

(L/min)

(N)

(Range yrs)

M F

ALS BLS

0

151

38

23-96

66

85

125 26

2

22

5.6

37-94

10

12

19 3

3

4

1

49-77

4

0

4 0

4

101

25.6

24-93

49

52

99 2

5

1

0.3

78

0

1

1 0

6

1

0.3

52

1

0

1 0

10

46

11.7

43-92

30

16

46 0

12

2

0.5

27-91

2

0

2 0

15

60

15.2

36-94

25

35

58 2

25

3

0.8

65-83

1

2

3 0

60

4

1

36-91

2

2

2 2

Total

395

100

190

205

360 35

Table 2

Oxygen flow, age, gender, ems training distribution grouping for statistical analysis.

O2 (L/min)

Frequency

%

Age

Gender

EMS

training level

(N)

(Range yrs)

M F

ALS BLS

0

151

38.23

23-96

66

85

125

26

2-6

129

32.66

24-94

64

65

124

5

10-15

108

27.34

27-94

57

51

106

2

25-60

7

1.77

36-91

3

4

5

2

Total

395

100

190

205

360

35

Table 3

Summary of oxygen therapy for patients with O2 saturation 100%.

were treated with mid oxygen flow (10-15 L/min) and one subject was treated with high oxygen flow (25-60 L/min). Only 38 (33%) of these cases received no oxygen treatment.

10-15

10-15

100

48

47

1

25-60

25-60

100

1

1

0

Total

116

107

9

https://doi.org/10.1016/j.ajem.2020.05.062

0735-6757/(C) 2020

206 Prehospital oxygen for stroke victims

Table 4

Multinomial logistic regression results as odds ratios.

Covariate Comparing odds of oxygen flow levels at specified level relative to 0 Overall p-value

2-6 L/min

10-12 L/min

15-60 L/min

Gender Female v. male

1.08 (0.60, 1.94)

p = 0.79

0.39 (0.20, 0.77)

p = 0.01

0.79 (0.49, 1.26)

p = 0.32

0.03

Transport ALS v. BLS 3.13 (1.04, 9.37)

p = 0.04

NEa

4.92 (1.83, 13.27) 0.01

p = 0.00

O2 Saturation b100 vs. 100 0.62 (0.34, 1.16)

0.32 (0.16, 0.63)

1.15 (0.67, 1.99) 0.00

p = 0.13

p = 0.00

p = 0.62

Age IQR = 20 1.10 (0.73, 1.66)

0.92 (0.59, 1.43)

0.92 (0.67, 1.28) 0.84

p = 0.64

p = 0.69

p = 0.63

Systolic BP IQR = 44 0.96 (0.67, 1.38)

0.91 (0.60, 1.39)

0.92 (0.68, 1.24) 0.94

p = 0.82

p = 0.67

p = 0.57

Diastolic BP IQR = 24 1.11 (0.79, 1.56)

0.94 (0.63, 1.39)

1.03 (0.78, 1.37) 0.89

p = 0.55

p = 0.74

p = 0.82

Heart rate IQR = 26 1.45 (1.00, 2.09)

1.70 (1.15, 2.50)

1.35 (0.98, 1.85) 0.05

p = 0.05

p = 0.01

p = 0.06

Respiratory rate IQR = 2 0.69 (0.54, 0.87)

0.92 (0.71, 1.20)

1.02 (0.85, 1.23) 0.01

p = 0.00

p = 0.55

p = 0.84

a NE = Not Estimable due to small cell counts.

flow and high oxygen flow were 3.13 and 4.92, respectively, the odds for those transported with BLS (p-values = 0.04, 0.00). The odds of receiv- ing mid-level (10-12 L/min) oxygen flow (relative to no oxygen) were lower among those with oxygen saturation b100% than among those with 100% saturation (OR = 0.32, p = 0.00). Increases in heart rate were associated with increases in the odds for low (2-6 L/min) and mid (10-12 L/min) oxygen flow (OR 1.45 and 1.70, p-values 0.05 and 0.01 respectively) and increases in respiratory rate were associated with decreased low oxygen flow (OR 0.69, p-value 0.00).

From our results it was difficult to discern a clear pattern of EMS ox- ygen treatment for our acute stroke subjects. It would have been en- couraging to see vital sign markers of respiratory distress (i.e. oxygen saturation and respiratory rate) to have a consistent association with changes in oxygen flow. For example, to see increasing respiratory rate be associated with increasing oxygen flow could indicate oxygen therapy being applied in response to patient physiologic needs. Al- though our results did show increasing heart rate to be associated with increased odds of low and mid oxygen flow, increased respiratory rate was associated with decreased odds of low flow oxygen and oxygen saturation of 100% was associated with decreased odds of receiving only mid flow oxygen. Thus, we did not see a consistent pattern of oxygen flow associated with vital sign indicators of respiratory distress.

Our study was limited by a lack of physical exam data as our data set did not provide data on subjects’ initial respiratory exam, stroke scale, or GCS and the only complete vital sign data we could access were vital signs upon arrival at the hospital. In addition, we were also limited by the retrospective nature of the study and that 183 cases (46%) in the data set were missing data that required abstracting from the subjects’ ED record which could be a source of bias in our results.

Despite these limitations and the age of our data, our findings were striking in that of the 395 acute stroke patients we studied, 244 (64%) received oxygen. One-hundred-sixteen patients (29%) had oxygen saturations of 100% and 78 of these (20%) were treated with oxygen indicating that a considerable portion of our patients could be at risk for hyperoxia. These results support Dylla and col- leagues’ assertion that prehospital hyperoxia seems widespread in acute stroke patients.

From Dylla and colleagues’ conclusions, it is encouraging to know the practice of prehospital hyperoxia in acute stroke patients does not appear harmful. Perhaps this will encourage additional research to elucidate EMS’ rationale for applying oxygen to acute stroke pa- tients, clarify if this practice offers benefit and ultimately lead to bet- ter guidelines for Prehospital providers to improve care for acute stroke patients.

Sources of funding

We received no funding support for the research and writing of this study.

Declaration of competing interest

Dr. McCoy was site PI for RE VERSE AD study, a multi-center study on dabigatran reversal with support from Boerhinger-Ingelheim. The study is now closed and the aims were unrelated to our current study.

Sincerely, John Collins, MD

Associate Professor Department of Emergency Medicine Rutgers – Robert Wood Johnson Medical School One Robert Wood Johnson Place

MEB 104

New Brunswick, NJ 08901 Tel: 732-235-8717

Fax: 732-235-7379

[email protected]

Acknowledgements

We are grateful to William Camarda, MS, NR-P and Anthony Cascio, MS, MICP for their assistance in extracting our initial data from EMS- Charts for our project.

References

  1. Dylla L, Adler DH, Abar B, et al. Prehospital supplemental oxygen for acute stroke – a retrospective analysis. Am J Emerg Med. 2019. https://doi.org/10.1016/j.ajem.2019. 11.002.
  2. Zhang K. Is prehospital supplemental oxygen effective for acute stroke? Am J Emerg Med. 2020. https://doi.org/10.1016/j.ajem.2020.02.048.
  3. The authors respond: studying prehospital supplemental oxygen in acute stroke. Am J Emerg Med. 2020. https://doi.org/10.1016/j.ajem.2020.02.051.
  4. Collins J, Sumner M, Barlas M, Ohman-Strickland P, McCoy J. Emergency medical ser- vice oxygen therapy in acute stroke patients: analysis of practice patterns. [Unpub- lished results].
  5. Rincon F, Kang J, Maltenfort M, et al. Association between hyperoxia and mortality after stroke: a Multicenter cohort study. Crit Care Med. 2014;42:387-96. https:// doi.org/10.1097/CCM.0b013e3182a27732.
  6. Ronning OM, Guldvog B. Should stroke victims routinely receive supplemental oxy- gen?: A quasi-randomized controlled trial. Stroke. 1999;30:2033-7. https://doi.org/ 10.1161/01.str.30.10.2033.
  7. Kilgannon JH, Jones AE, Shapiro NI, et al. Association between arterial hyperoxia fol- lowing resuscitation from cardiac arrest and in-hospital mortality. Jama. 2010;303: 2165-71. https://doi.org/10.1001/jama.2010.707.

Prehospital oxygen for stroke victims 207

John Collins M.D Rutgers-Robert Wood Johnson Medical School, Department of Emergency Medicine, One Robert Wood Johnson Place, MEB 104, New Brunswick, NJ

08901, USA

Corresponding author.

E-mail address: [email protected].

Michael Sumner D.O1 School of Graduate Studies-Biomedical Sciences, Rutgers-Robert Wood Johnson Medical School, 675 Hoes Lane West, Research Tower Room 102,

Piscataway, NJ 08854-5635, USA

Mehwish Barlas BBE, M.B.S. School of Graduate Studies-Biomedical Sciences, Rutgers-Robert Wood Johnson Medical School, 675 Hoes Lane West, Research Tower Room 102,

Piscataway, NJ 08854-5635, USA

Pamela Ohman Strickland Ph.D., M.S

Rutgers School of Public Health, Room 218, 683 Hoes Lane West,

Piscataway, NJ 08854, USA E-mail address: [email protected].

Jonathan McCoy M.D Rutgers-Robert Wood Johnson Medical School, Department of Emergency Medicine, One Robert Wood Johnson Place, MEB 104, New Brunswick, NJ

08901, USA

E-mail address: [email protected].

21 May 2020

1NAS Oceana Branch medical clinic, Carrier Air Wing 8, 1640 Tomcat Blvd, Virginia Beach, VA 23454.