Tight control of prehospital ventilation by capnography in major trauma victims

M. Helm*, R. Schuster, J. Hauke and L. Lampl

Department of Anaesthesiology and Intensive Care Medicine, Federal Armed Forces Medical Centre Ulm, D-89070 Ulm, Germany

Corresponding author. E-mail: matthias.helm@extern.uni-ulm.de

Accepted for publication: November 8, 2002


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background. Tracheal intubation combined with controlled ventilation of the lungs is an important part of the prehospital management of major trauma victims, but gauging the adequacy of ventilation remains a major problem.

Methods. Ninety-seven major trauma victims who underwent tracheal intubation in the field and controlled ventilation of the lungs during prehospital treatment by a Helicopter Emergency Medical Service were assigned randomly to one of two groups: (1) monitor group (n=57) and (2) monitor-blind group (n=40), according to whether the anaesthetist could or could not see an attached capnograph screen. In the monitor-blind group ventilation was set by using a tidal-volume of 10 ml kg–1 estimated body weight and an age-appropriate ventilatory frequency. In the monitor group, ventilation was adjusted to achieve target end-tidal carbon dioxide values determined by the ‘physiological state’ of the trauma victim. Arterial blood gases were measured upon hospital admission while maintaining the ventilation initiated in the field and the PaCO2 value obtained was used as the determinant of the adequacy of prehospital ventilation.

Results. The incidence of ‘normoventilation’ was significantly higher (63.2 vs 20%; P<0.0001) and the incidence of ‘hypoventilation’ upon hospital admission was significantly lower (5.3 vs 37.5%; P<0.0001) in the monitor group; patients with severe head and chest trauma and haemodynamically unstable patients and those with a high injury severity score were significantly more likely to be ‘normoventilated’ upon hospital admission in the monitor group than in the monitor-blind group.

Conclusions. The data support the routine use of prehospital capnographic monitoring using target end-tidal carbon dioxide values adapted to the physiological state of the patient in major trauma victims requiring tracheal intubation in the field.

Br J Anaesth 2003; 90: 327–32

Keywords: carbon dioxide; complications, trauma; monitoring, carbon dioxide; ventilation, artificial


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Recent studies1 3 4 have demonstrated a significantly improved outcome in major trauma victims1 when controlled ventilation of the lungs is initiated at the scene of the accident after tracheal intubation. In the prehospital setting, even severely injured trauma victims can usually be adequately oxygenated57 but the ‘quality’ of prehospital artificial ventilation may affect outcome.4 In their study of the quality of prehospital ventilation during primary rescue missions, Kehrberger and colleagues8 found that an appropriate degree of ventilation had been achieved in only 10% of cases. Arterial blood gas analysis provides an optimal assessment of oxygenation and ventilation,9 10 but it is rarely practicable in the prehospital setting.10 End-tidal carbon dioxide measurement has been proposed as an alternative for monitoring the adequacy of prehospital ventilation in major trauma victims.5 6 1113

The purpose of this study was to compare the effect of ventilation adjusted to achieve an end-tidal carbon dioxide determined according to the clinical condition of each patient with ventilation determined according to the weight of the patient on admission PaCO2.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
We performed a prospective randomized study in trauma patients treated before hospital admission by the medical team of the Helicopter Emergency Medical Service (HEMS) ‘Christoph 22’ operating from the Federal Armed Forces Medical Centre Ulm, Germany from 1 January 1998 to 31 December 1999. The medical team at this HEMS consists of an experienced trauma anaesthesiologist and a paramedic, both members of the Department of Anaesthesiology and Intensive Care Medicine, Federal Armed Forces Medical Centre Ulm. In all patients included in the study the trachea was intubated electively at the scene of the accident and controlled ventilation of the lungs was instituted and maintained during prehospital treatment. We applied continuous monitoring of cardiovascular and respiratory function (ECG, automated non-invasive arterial pressure, pulse oximetry, and capnography) to every patient upon arrival at the scene using a portable multimodal monitor (Propaq 106EL; Protocol Systems, Inc., Beaverton, OR, USA) and this monitoring was maintained until hospital admission. The patients were assigned randomly to one of two groups: monitor-blind and monitor. In the monitor-blind group, the visual display of the capnograph was covered by tape, so that the trauma anaesthesiologist was not aware of the capnogram or the end-tidal carbon dioxide values during prehospital ventilation. The visual and audio alarm limits were set so that only extremes of hypoventilation or hyperventilation (end-tidal carbon dioxide values 20 mm Hg above or below the target end-tidal carbon dioxide) and circuit obstruction or disconnection were indicated to prevent a catastrophic event but not to bias the trauma anaesthesiologist conducting the ventilation. In the monitor-blind group the tidal-volume and ventilatory frequency were set by using the generally accepted recommendation:14 tidal-volume (TV) = 10 ml kg–1 estimated body weight; in adults the ventilatory frequency was set to 10 bpm, in children the ventilatory frequency was adapted to the corresponding age. In contrast, in the monitor group the display of the capnograph was visible for the trauma anaesthesiologist during prehospital treatment. In this group, ventilation was controlled solely by capnographic monitoring. Based upon the results of studies13 1518 investigating the arterial end-tidal carbon dioxide tension difference P(a-E')CO2 in ventilated patients during prehospital treatment, the target end-tidal carbon dioxide depended on the haemodynamic stability and the presence of severe chest trauma. In those who the trauma anaesthesiologist classified as haemodynamically stable and in whom no signs of severe chest trauma were apparent, ventilation was adjusted to achieve an end-tidal carbon dioxide in the range of 30–35 mm Hg, on the assumption that in these patients P(a-E')CO2 varies from 3 to 5 mm Hg. In haemodynamically unstable trauma victims and/or patients with associated severe chest trauma ventilation was adjusted to achieve an end-tidal carbon dioxide in the range of 25–30 mm Hg on the assumption that in these patients P(a-E')CO2 varies from 10 to 15 mm Hg. In this study we defined haemodynamic instability as a systolic arterial pressure less than 90 mm Hg upon arrival at the scene; severe chest trauma was determined by physical examination and pulse oximetric monitoring as described previously.19 In the total study population, prehospital ventilation was maintained by an automatic emergency respirator (DRÄGER Oxylog 2000®); in all cases the fraction of inspired oxygen (FIO2) was 1.0 during prehospital treatment. In both study groups, the goal was to achieve ‘normoventilation’, that is a PaCO2 between 35 and 45 mm Hg, with a target of 35 mm Hg. Upon hospital admission, and with ventilation maintained from the prehospital regimen, arterial blood gas samples were obtained and analysed (i-STAT, Hewlett Packard, Böblingen, Germany). Recorded data included patient characteristics, mechanism of injury, injured body regions, and injury pattern, injury severity (Injury Severity Score, ISS),20 blood gas analysis on hospital admission (pH, base excess, PaO2, and PaCO2), and end-tidal carbon dioxide values during prehospital treatment. All recorded data were collected concurrently and entered into a relational database management system (Microsoft Access, Microsoft Corporation, Redmond, WA, USA) based upon the national trauma registry of the German Society of Trauma Surgery.21 This study was performed in accordance with the Helsinki Declaration and with agreement of the Ethics Committee of the University of Ulm to its protocol.

Stastical methods
All values in the tables and figures are expressed as means (SD) unless otherwise indicated. Each variable was tested for differences between groups by Student’s t-test or {chi}2 analysis where appropriate. A value of P<0.05 was considered to be statistically significant. Statistical analysis was performed using specialized statistical software (Almo© version 5.0; K. Holm, University of Graz, Austria).


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Out of 183 trauma victims, 86 patients were excluded from the study because of failure to obtain an arterial blood gas sample upon hospital admission before the prehospital ventilation regimen was changed (n=55), incomplete clinical follow-up data (n=25) and death during the prehospital treatment phase (n=6). Therefore, 97 (ntot=97) trauma patients (70 male; 27 female/age 36.5 (20.0) yr, range: 6–85 yr/ISS 32.9 (17.8)) were enrolled into the study. The mechanism of injury in all 97 patients was blunt trauma. The detailed analysis of mechanism of injury and injured body regions is presented in Tables 1 and 2, respectively. In 23 patients a single body region was injured, in 33 patients there was a combination of head and chest trauma and in a further 15 patients there was a combination of head, chest, and limb trauma.


View this table:
[in this window]
[in a new window]
 
Table 1 Mechansim of injury within the study population (n=97)
 

View this table:
[in this window]
[in a new window]
 
Table 2 Patient characteristics and nature of injuries (monitor group vs monitor-blind group). *Expressed as mean (SD) [range]; n.s, not significant
 
The arterial blood gas analysis upon hospital admission showed no statistically significant difference in pH, PaO2, PaCO2, and BE between the monitor-blind and monitor group (Fig. 1). The number of patients in whom ‘normoventilation’ (defined as: 35<=PaCO2<=45 mm Hg), ‘hyperventilation’ (defined as PaCO2<35 mm Hg) and ‘hypoventilation’ (defined as PaCO2>45 mm Hg) was achieved in the monitor and monitor-blind group is presented in Figure 2. There was a significantly higher proportion of patients ‘normoventilated’ upon hospital admission in the monitor group than in the monitor-blind group (63.2 vs 20.0%; P<0.0001), whereas the occurrence of ‘hypoventilation’ was significantly reduced in the monitor-group (5.3 vs 37.5%; P=0.0001). The incidence of ‘hyperventilation’ was not significantly different in the monitor and the monitor-blind group (P=0.31). As shown in Figure 3, patients with severe head trauma (AIS>3) and severe chest trauma (AIS>3), those who were haemodynamically unstable and those with a high injury severity score (ISS>19) were significantly more likely to be ‘normoventilated’ upon hospital admission in the monitor group than in the monitor-blind group (Figure 3).



View larger version (20K):
[in this window]
[in a new window]
 
Fig 1 Blood gas analysis (pH, PaO2, PaCO2, and BE) upon hospital admission within the study population (n=97) displayed as combined box-and-whisker and dot plots.

 


View larger version (24K):
[in this window]
[in a new window]
 
Fig 2 Incidence of ‘normoventilation’ (35<=PaCO2<=45 mm Hg), ‘hypoventilation’ (PaCO2>45 mm Hg) and ‘hyperventilation’ (PaCO2<35 mm Hg) upon hospital admission within the study population (n=97). *P<0.0001, n.s., not significant.

 


View larger version (38K):
[in this window]
[in a new window]
 
Fig 3 Incidence of ‘normoventilation’ (35<=PaCO2<=45 mm Hg) within the study population upon hospital admission (n=97). SBP, systolic arterial pressure; ISS, injury severity score; AIS, abbreviated injury severity score.

 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Various studies1 3 22 23 have demonstrated that tracheal intubation and controlled ventilation of the lungs instituted in the field significantly improves outcome in major trauma victims, but some of these and other studies3 6 2426 also indicate that the ‘quality’ of the controlled ventilation has to be improved. The ventilation devices used in the prehospital setting are a major factor in the ‘quality’ of the controlled. In the case of manual ventilation by self-inflating manual resuscitator (i.e. AMBU®-bag) there is usually no measurement of minute-volume; the ‘quality’ of ventilation depends solely upon the experience and skill of the person squeezing the bag.27 28 Most of the automatic emergency respirators permit the operator to preset the ventilatory frequency and minute-volume but only some of them (i.e. Oxylog 2000®) actually measure these preset variables. Furthermore, a number of studies27 29 30 have shown that in nearly all commonly used automatic emergency respirators, the delivered minute-volume differs up to ±20% from the preset value. On the other hand there are only very limited capabilities for monitoring controlled ventilation in the prehospital setting. It is therefore not surprising that in major trauma victims despite tracheal intubation and controlled ventilation in the field, PaCO2 values between 16 and 86 mm Hg upon hospital admission were noted.5 Kuhnighk and colleagues7 found in their study using calculated prehospital ventilation in multi-system-trauma victims that in only 18% of the cases was the target PaCO2 range actually achieved upon hospital admission and conclude that ‘to date a calculated prehospital ventilation in multi-system-trauma victims is not practicable’.

Monitoring of end-tidal carbon dioxide has become standard in the prehospital management and air transfer of severely traumatized patients. Capnography applied at the scene of an accident assists confirmation of tracheal intubation, diagnosis of circuit disconnection and assessment of cardiopulmonary resuscitation.31 Furthermore, on-scene capnography has been proposed as an alternative to intermittent PaCO2 measurements for monitoring the adequacy of prehospital ventilation in major trauma victims.5 6 11 12 31 The problem of capnography in this context is that it is not solely a measurement of the respiratory function and capnograms must be interpreted in conjunction with other clinical findings.32 Some authors11 13 33 34 doubt that capnography may be of benefit in calculating prehospital ventilation. Böbel and colleagues13 concluded in their study that calculation of prehospital ventilation by end-tidal carbon dioxide monitoring ‘seems to be problematic because of unknown arterial end-tidal carbon dioxide tension difference in the prehospital setting’. In major trauma victims, severe chest trauma, hypotension and heavy blood loss were found to reduce end-tidal carbon dioxide and increase P(a-E')CO2 differences to more than 10 mm Hg.9 13 15 16 18 36 37 These increased P(a-E')CO2 differences are thought to be largely because of an increased alveolar dead space, which is probably caused by decreased alveolar perfusion resulting from a reduced cardiac output and/or maldistribution or occlusion of portions of the pulmonary blood flow.9

Our study demonstrated that prehospital ventilation calculated according to the ‘physiological state’ of the trauma victim and controlled by capnography, facilitates tighter control of PaCO2. The incidence of normoventilation was significantly increased (63.2 vs 20.0%; P<0.0001) while the incidence of hypoventilation upon hospital admission was significantly reduced (5.3 vs 37.5%; P<0.0001) in the monitor group compared with the monitor-blind group, whereas the incidence of hyperventilation upon hospital admission was not reduced in the monitor group. This result may be explained by using a PaCO2 value of 35 mm Hg both as the target in the monitor group and the cut-off value for classifying the outcome as normo- or hyperventilation. Palmon and colleagues12 found similar results in their study on the in-hospital transport of intubated patients from the operating room to the intensive care unit (ICU) and from the ICU for examinations in the radiology department; they conclude that using an end-tidal carbon dioxide monitor for transport helps guide ventilation of patients who require tight control of PaCO2.

The prehospital phase seems to be the most critical interval in determining the ultimate outcome after traumatic brain injury. Therefore, especially patients with associated severe head trauma may profit from a close control of PaCO2. Schüttler and colleagues4 have shown in their study on the quality of prehospital management of severe head injury that mortality is significantly reduced within a group of prehospitally intubated and ventilated patients in those with ‘good’ compared with those with ‘poor’ respiratory therapy (25 vs 61%). The significant increase of intended normoventilation in the population of patients with associated severe head injury (AIS>3) in our study in the monitor group compared with the monitor-blind group (57.5 vs 12.9%; P<0.001) is therefore of major importance.

Based upon the results of a number of studies,13 1518 patients in the monitor group of our study with associated severe chest trauma and/or haemodynamic instability were ventilated to achieve an end-tidal carbon dioxide in the range of 25–30 mm Hg; both subpopulations were significantly more patients meeting these criteria were normoventilated upon hospital admission in the monitor group than in the monitor-blind group (Fig. 3). Therefore, we have shown that it is possible without knowledge of P(a-E')CO2 to ventilate trauma victims, including subpopulations with probable increased P(a-E')CO2, more tightly in the prehospital setting than has been possible before.

In conclusion, the data of our study support the routine use of an end-tidal carbon dioxide controlled prehospital ventilation in trauma victims, which considers the ‘physiological state’ of the patient. The protocol used in this study seems to be simple, practicable and effective for usage in the prehospital setting.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1 Winchell RJ, Hoyt DB. Endotracheal intubation in the field improves survival in patients with severe head injury. Arch Surg 1997; 132: 592–7[Abstract]

2 Nolan JP, Paar MJA. Aspects of resuscitation in trauma. Br J Anaesth 1997; 79: 226–40[Free Full Text]

3 Trupka A, Waydhas C, Nast-Kolb D, et al. Early intubation in severely injured patients. Eur J Emerg Med 1994; 1: 1–8[Medline]

4 Schüttler J, Schmitz B, Bartsch AC, et al. Efficacy of advanced trauma life support in patients with brain injury and multiple trauma. A contribution to quality assurance in emergency medicine (German). Anaesthesist 1995; 44: 850–8[CrossRef][ISI][Medline]

5 Helm M, Hauke J, Sauermüller G, et al. About the quality of prehospital emergency ventilation—a prospective study in trauma patients (German). Unfallchirurg 1999; 102: 347–53[CrossRef][ISI][Medline]

6 Helm M, Hauke J, Lampl L. About the quality of prehospital emergency ventilation—a prospective study in patients with severe head injury. Br J Anaesth 2002; 88: 345–9[Abstract/Free Full Text]

7 Kuhnigk H, Lingg V, Sefrin P, et al. Kalkulierte Beatmung bei Polytraumatisierten—Zielsetzung des Notarztes und Ergebnisse bei Klinikaufnahme (German). Anaesthesiol Intensivmed Notfallmed Schmerzther 1998; 33/3: 190

8 Kehrberger E, Hörtling H. Blutgasanalysen nach präklinischer kontrollierter Beatmung durch den Notarzt. Der Notarzt 1989; 5: 140–4

9 Russel GB, Graybeal JM. Reliability of the arterial to end-tidal carbon dioxide gradient in mechanically ventilated patients with multisystem trauma. J Trauma 1994; 36: 317–22[ISI][Medline]

10 Santos LJ, Varon J, Pic-Aluas L, et al. Practical use of end-tidal carbon dioxide monitoring in the emergency department. J Emerg Med 1994; 12: 633–44[CrossRef][Medline]

11 Hoffmann RA, Krieger BP, Kramer MR, et al. End-tidal dioxide in critically ill patients during changes in mechanical ventilation. Am Rev Respir Dis 1989; 140: 1265–8[ISI][Medline]

12 Palmon SC, Liu M, Moore LE, et al. Capnography facilitates tight control of ventilation during transport. Crit Care Med 1996; 24: 608–11[CrossRef][ISI][Medline]

13 Böbel M, Graf S, Geitner K, et al. Bestimmung arterio-endexspiratorischer CO2-Differenzen in der präklinischen Versorgungsphase. Der Notarzt 1997; 13: 126–31

14 Schou J. Endotracheal intubation and ventilation. In: Schou J, ed. Prehospital Emergency Medicine. Amsterdam: Harwood Academic Publishers, 1992; 79–91

15 Ensle G, Altemeyer KH. The arterial to end expiratory CO2 difference with ventilated patients in emergency medicine (German). Notfall Rettungsmedizin 1998; 6: 347–54

16 Whitesell R, Asiddao C, Gollmann D, et al. Relationship between arterial and peak expired carbon dioxide pressure during anethesia and factors influencing the difference. Anesth Analg 1981; 60: 508–12[Abstract]

17 Rückoldt H, Marx G, Leuwer M, et al. Pulsoxymetrie und Kapnometrie bei Intensivtransporten: kombinierter Einsatz verringert das Transportrisiko. Anaesthesiol Intensivmed Notfallmed Schmerzther 1998; 32: 32–36

18 Helm M, Hauke J, Lampl L, et al. Arterial to end-tidal carbon dioxide gradient and Horovitz-quotient – of value in diagnostic blunt chest trauma? Br J Anaesth 1995; 74: 127

19 Helm M, Hauke J, Esser M, et al. Prehospital diagnosis in blunt chest trauma: efficacy of continuous pulse oximetric monitoring (German). Chirurg 1997; 68: 606–12[CrossRef][ISI][Medline]

20 Baker SP, O’Neill B, Haddon W. The Injury Severity Score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974; 14: 187–96[ISI][Medline]

21 Arbeitsgemeinschaft ‘Scoring’ der Deutschen Gesellschaft für Unfallchirurgie (DGU) Das Traumaregister der Deutschen Gesellschaft für Unfallchirurgie. Unfallchirurg 1994; 97: 230–7[ISI][Medline]

22 Miller JD, Sweet RC, Narayan R, et al. Early insults of the injured brain. JAMA 1978; 240: 439–42[Abstract]

23 Gildenberg PL, Makela M. The effect of early intubation and ventilation on outcome following head trauma. In: Winn WR, Rimel R, Jane JA, eds. Recent Advances in Neurotrauma. New York: Raven Press, 1985; 79–90

24 Kehrberger E, Hörtling H. Blutgasanalysen nach präklinischer kontrollierter Beatmung durch den Notarzt. Der Notarzt 1990; 5: 2–4

25 Kuhnigk H, Zischler K. Narkose im Rettungsdienst. In: Sefrin P, ed. Beatmung im Rettungsdienst. Munich: Zuckschwerdt, 1995; 89–95

26 Gervais HW, Eberele B, Konietzke D, et al. Comparison of blood gases of ventilated patients during transport. Crit Care Med 1987; 15/8: 761–3

27 Carden E, Friedman D. Further studies of manually operated self-inflating resuscitation bags. Anesth Analg 1977; 56: 202–6[Abstract]

28 Rossi R, Lotz P, Keller A. Charakteristika von Geräten zur Beatmung von Patienten am Notfallort und auf dem transport. Anästhesist 1989; 39 (Suppl 1): 142

29 Rossi R, Rockemann M, Keller A. Einsatz von Notfallrespiratoren beim innerklinischen transport. Notfallmedizin 1993; 19: 26–9

30 Morley AP. Prehospital monitoring of trauma patients: experience of an emergency helicopter medical service. Br J Anaesth 1996; 76: 726–30[Free Full Text]

31 Napolitano LM: Capnography in critical care: accurate assessment of ARDS therapy? Crit Care Med 1999; 27/5: 862–3[CrossRef]

32 Prause G, Hetz H, Lauda P, et al. A comparison of the end-tidal CO2 documented by capnometry and the arterial pCO2 in emergency patients. Resuscitation 1997; 35: 145–8[CrossRef][ISI][Medline]

33 Wahba RWM, Tessler MJ. Misleading end-tidal CO2 tensions. Can J Anaesth 1996; 43: 862–6[Abstract]

34 Wilson R, Tyburski JG, Kubinec SM, et al. Intraoperative end-tidal carbon dioxide levels and derived calculations correlated with outcome in trauma patients. J Trauma 1996; 41: 606–11[ISI][Medline]

35 Lenz G, Klöss T, Schorer R. Grundlagen und Anwendung der Kapnometrie. Anästhesiol Intensivmed 1985; 26: 133–41

36 Craig GR, Randalls PB. Analysis of arterial to end-tidal carbon dioxide gradients in ventilated trauma patients. J Trauma 1993; 35: 331

37 Prough DS, Lang J. Therapy of patients with head injuries: key parameters for management. J Trauma 1997; 42/5: 10–17