Tag Archives: capnography

Predicting volume responsiveness

IVCiconOne of the current Holy Grails of ED critical care is to find a reliable measure of fluid responsiveness in those patients with impaired organ perfusion, such as those with severe sepsis. This would enable us to identify those patients whose cardiac output would be improved by fluid therapy, and avoid subjecting ‘non-responders’ to the risks associated with fluid overload. Thanks to the uptake of early goal-directed therapy in sepsis, under-resuscitation is now much less common in the ED. However a growing evidence base reveals the dangers of over-resuscitation. We have a responsibility to optimise fluid therapy as best we can with the equipment we have, according to the latest evidence.

Inferior Vena Cava Ultrasound
Some tests of fluid responsiveness rely on the effect of respiration-induced changes in pleural pressure on the circulation. Inferior vena cava (IVC) size and degree of inspiratory collapse correlate with central venous pressure (CVP), but CVP is not a reliable predictor of volume status or responsiveness. Skinny, collapsing IVCs detected on ultrasound suggest volume responsiveness, but the lack of this finding does not exclude fluid responsiveness. IVC size and measurement can be affected by patient position, probe position, and a variety of health states from athleticism to increased abdominal pressure.

Pulse Pressure Variation
Respiratory pulse pressure variation derived from an arterial line trace in mechanically ventilated patients who are adequately sedated and receiving large tidal volumes can predict fluid responsiveness too. Variability in tidal volume, the presence of spontaneous breathing activity in a ventilated patient, and cardiac dysrhythmia can all confound the usefulness of this method.

End expiratory occlusion
Another test in mechanically ventilated patients is the end expiratory occlusion test. A positive pressure inspiratory breath cyclically decreases the left cardiac preload. Occluding the circuit at end-expiration prevents this cyclic impediment in left cardiac preload and acts like a fluid challenge. A 15 second expiratory occlusion is performed and an increase in pulse pressure or (if you can measure it) cardiac index predicts fluid responsiveness with a high degree of accuracy. The patient must be able to tolerate the 15 second interruption to ventilation without initiating a spontaneous breath.

Passive Leg Raise
Passive leg raising (PLR) involves measuring cardiac output (or its surrogate, velocity-time integral, or VTI) before and after tilting the semirecumbent patient supine and raising the legs to 45 degrees. This ‘autotransfuses’ blood from the lower limbs to the core and acts as a reversible fluid challenge. An increase in VTI identifies fluid responders. It would be nice if a PLR-induced increase in blood pressure revealed the answer, but BP does not reliably inform us of changes in cardiac output.

All these tests have limitations. Pulse pressure variation fails in patients with low respiratory system compliance, such as is found in ARDS(1). End-expiratory occlusion and PLR work in low respiratory system compliance, but the former still requires mechanical ventilation, and the latter requires a means of estimating cardiac output or a surrogate – oesophageal Doppler, the velocity-time integral measured by transthoracic echocardiography, and femoral artery flow (measured by arterial Doppler) have all been used. Non-invasive cardiac output monitors that are not operator dependent exist, such as the NICOM(TM) bioreactance device. Bioreactance cardiac output measurement is based on an analysis of relative phase shifts of an oscillating current that occurs when this current traverses the thoracic cavity. Its advantages are that it is noninvasive, it does not require endotracheal intubation or an arterial line, and it provides a good estimate of stroke volume in patients with atrial fibrillation.

A recent study evaluating the combination of PLR with NICOM(TM) bioreactance monitoring revealed that another tool could indicate volume responsiveness: an increase in carotid blood flow after PLR, as measured by carotid Doppler flow imaging(2). A threshold increase in carotid Doppler flow imaging of 20% for predicting volume responsiveness had a sensitivity and specificity of 94% and 86%, respectively. This was studied in a heterogenous group of hemodynamically unstable patients, suggesting applicability to the kind of patients who present to the ED, although numbers were small so more validation is required.

End-tidal carbon dioxide
End-tidal carbon dioxide (ETCO2) levels depend on cardiac output. Increasing cardiac output with a fluid challenge or PLR increases ETCO2,as long as ventilatory and metabolic conditions remain stable. In a recent small study, a PLR-induced increase in ETCO2 ≥ 5 % predicted a fluid-induced increase in cardiac index ≥ 15 % with sensitivity of 71 % (95 % confidence interval: 48-89 %) and specificity of 100 (82-100) %(3). The maximal effects of PLR on CI and ETCO2 were observed within 1 min.

So what can I use?
In summary, differentiating fluid responders from non-responders in the ED remains a challenge. The method used depends on available equipment and expertise, and whether the patient is spontaneously breathing or mechanically ventilated. The NICOM(TM) shows great promise but until your department can afford one, ultrasound is the way to go; small collapsing IVCs suggest fluid responders. Learning to measure a VTI on transthoracic echo or carotid Doppler flow will help you assess the response to a PLR in spontaneously ventilating patients. If they’re mechanically ventilated, then looking for an ETCO2 rise after PLR could be a simpler alternative.

Fluid responsiveness assessment – options in the Emergency Department

Inferior Vena Cava Ultrasound
Helpful if skinny / large degree of respirophasic collapse – suggests fluid responsive – ventilated or spontaneous breathing

Passive Leg Raise
Good in ventilated or spontaneous breathing patients; need to measure cardiac output or a surrogate, such as VTI (echo), NICOM(TM), carotid Doppler flow, or ETCO2 (if ventilation and metabolic status constant)

Pulse Pressure Variation
Requires full mechanical ventilation; no good if low respiratory compliance / disturbed heart-lung interaction

End expiratory occlusion
Requires mechanical ventilation and patient tolerance of 15 seconds of apnoea. Acts like a passive leg raise so need a measure of cardiac output or surrogate

I look forward to more studies on these modalities, and to trying some of them in the resus room at every available opportunity.


1. Passive leg-raising and end-expiratory occlusion tests perform better than pulse pressure variation in patients with low respiratory system compliance
Crit Care Med. 2012 Jan;40(1):152-7

OBJECTIVES: We tested whether the poor ability of pulse pressure variation to predict fluid responsiveness in cases of acute respiratory distress syndrome was related to low lung compliance. We also tested whether the changes in cardiac index induced by passive leg-raising and by an end-expiratory occlusion test were better than pulse pressure variation at predicting fluid responsiveness in acute respiratory distress syndrome patients.

DESIGN: Prospective study.

SETTING: Medical intensive care unit.

PATIENTS: We included 54 patients with circulatory shock (63 ± 13 yrs; Simplified Acute Physiology Score II, 63 ± 24). Twenty-seven patients had acute respiratory distress syndrome (compliance of the respiratory system, 22 ± 3 mL/cm H2O). In nonacute respiratory distress syndrome patients, the compliance of the respiratory system was 45 ± 9 mL/cm H2O.

MEASUREMENTS AND MAIN RESULTS: We measured the response of cardiac index (transpulmonary thermodilution) to fluid administration (500 mL saline). Before fluid administration, we recorded pulse pressure variation and the changes in pulse contour analysis-derived cardiac index induced by passive leg-raising and end-expiratory occlusion. Fluid increased cardiac index ≥ 15% (44% ± 39%) in 30 “responders.” Pulse pressure variation was significantly correlated with compliance of the respiratory system (r = .58), but not with tidal volume. The higher the compliance of the respiratory system, the better the prediction of fluid responsiveness by pulse pressure variation. A compliance of the respiratory system of 30 mL/cm H2O was the best cut-off for discriminating patients regarding the ability of pulse pressure variation to predict fluid responsiveness. If compliance of the respiratory system was >30 mL/cm H2O, then the area under the receiver-operating characteristics curve for predicting fluid responsiveness was not different for pulse pressure variation and the passive leg-raising and end-expiratory occlusion tests (0.98 ± 0.03, 0.91 ± 0.06, and 0.97 ± 0.03, respectively). By contrast, if compliance of the respiratory system was ≤ 30 mL/cm H2O, then the area under the receiver-operating characteristics curve was significantly lower for pulse pressure variation than for the passive leg-raising and end-expiratory occlusion tests (0.69 ± 0.10, 0.94 ± 0.05, and 0.93 ± 0.05, respectively).

CONCLUSIONS: The ability of pulse pressure variation to predict fluid responsiveness was inversely related to compliance of the respiratory system. If compliance of the respiratory system was ≤ 30 mL/cm H2O, then pulse pressure variation became less accurate for predicting fluid responsiveness. However, the passive leg-raising and end-expiratory occlusion tests remained valuable in such cases.

2. The use of bioreactance and carotid doppler to determine volume responsiveness and blood flow redistribution following passive leg raising in hemodynamically unstable patients
Chest. 2013 Feb 1;143(2):364-70

BACKGROUND: The clinical assessment of intravascular volume status and volume responsiveness is one of the most difficult tasks in critical care medicine. Furthermore, accumulating evidence suggests that both inadequate and overzealous fluid resuscitation are associated with poor outcomes. The objective of this study was to determine the predictive value of passive leg raising (PLR)- induced changes in stroke volume index (SVI) as assessed by bioreactance in predicting volume responsiveness in a heterogenous group of patients in the ICU. A secondary end point was to evaluate the change in carotid Doppler fl ow following the PLR maneuver.

METHODS: During an 8-month period, we collected clinical, hemodynamic, and carotid Doppler data on hemodynamically unstable patients in the ICU who underwent a PLR maneuver as part of our resuscitation protocol. A patient whose SVI increased by . 10% following a fluid challenge was considered a fluid responder.

RESULTS: A complete data set was available for 34 patients. Twenty-two patients (65%) had severe sepsis/septic shock, whereas 21 (62%) required vasopressor support and 19 (56%) required mechanical ventilation. Eighteen patients (53%) were volume responders. The PLR maneuver had a sensitivity of 94% and a specificity of 100% for predicting volume responsiveness (one false negative result). In the 19 patients undergoing mechanical ventilation, the stroke volume variation was 18.0% 5.1% in the responders and 14.8% 3.4% in the nonresponders ( P 5 .15). Carotid blood fl ow increased by 79% 32% after the PLR in the responders compared with 0.1% 14% in the nonresponders ( P , .0001). There was a strong correlation between the percent change in SVI by PLR and the concomitant percent change in carotid blood fl ow ( r 5 0.59, P 5 .0003). Using a threshold increase in carotid Doppler fl ow imaging of 20% for predicting volume responsiveness, there were two false positive results and one false negative result, giving a sensitivity and specificity of 94% and 86%, respectively. We noted a significant increase in the diameter of the common carotid artery in the fluid responders.

CONCLUSIONS: Monitoring the hemodynamic response to a PLR maneuver using bioreactance provides an accurate method of assessing volume responsiveness in critically ill patients. In addition, the study suggests that changes in carotid blood fl ow following a PLR maneuver may be a useful adjunctive method for determining fluid responsiveness in hemodynamically unstable patients.

3. End-tidal carbon dioxide is better than arterial pressure for predicting volume responsiveness by the passive leg raising test
Intensive Care Med. 2013 Jan;39(1):93-100

PURPOSE: In stable ventilatory and metabolic conditions, changes in end-tidal carbon dioxide (EtCO(2)) might reflect changes in cardiac index (CI). We tested whether EtCO(2) detects changes in CI induced by volume expansion and whether changes in EtCO(2) during passive leg raising (PLR) predict fluid responsiveness. We compared EtCO(2) and arterial pulse pressure for this purpose.

METHODS: We included 65 patients [Simplified Acute Physiology Score (SAPS) II = 57 ± 19, 37 males, under mechanical ventilation without spontaneous breathing, 15 % with chronic obstructive pulmonary disease, baseline CI = 2.9 ± 1.1 L/min/m(2)] in whom a fluid challenge was decided due to circulatory failure and who were monitored by an expiratory-CO(2) sensor and a PiCCO2 device. In all patients, we measured arterial pressure, EtCO(2), and CI before and after a fluid challenge. In 40 patients, PLR was performed before fluid administration. The PLR-induced changes in arterial pressure, EtCO(2), and CI were recorded.

RESULTS: Considering the whole population, the fluid-induced changes in EtCO(2) and CI were correlated (r (2) = 0.45, p = 0.0001). Considering the 40 patients in whom PLR was performed, volume expansion increased CI ≥ 15 % in 21 “volume responders.” A PLR-induced increase in EtCO(2) ≥ 5 % predicted a fluid-induced increase in CI ≥ 15 % with sensitivity of 71 % (95 % confidence interval: 48-89 %) and specificity of 100 (82-100) %. The prediction ability of the PLR-induced changes in CI was not different. The area under the receiver-operating characteristic (ROC) curve for the PLR-induced changes in pulse pressure was not significantly different from 0.5.

CONCLUSION: The changes in EtCO(2) induced by a PLR test predicted fluid responsiveness with reliability, while the changes in arterial pulse pressure did not.

Colorimetric CO2 detectors and newborns

Colorimetric CO2 detectors may fail to indicate successful tracheal tube placement in adults in certain circumstances, such as low cardiac output states, and waveform capnography is considered the gold standard. We now have data that demonstrate their inadequacy for neonatal intubation. Ideally, waveform devices should be used by all professionals who intubate patients – from paramedics to neonatologists.

AIM: Clinical assessment and end-tidal CO(2) (ETCO(2)) detectors are routinely used to verify endotracheal tube (ETT) placement. However, ETCO(2) detectors may mislead clinicians by failing to identify correct placement under a variety of conditions. A flow sensor measures gas flow in and out of an ETT. We reviewed video recordings of neonatal resuscitations to compare a colorimetric CO(2) detector (Pedi-Cap®) with flow sensor recordings for assessing ETT placement.
METHODS: We reviewed recordings of infants <32 weeks gestation born between February 2007 and January 2010. Airway pressures and gas flow were recorded with a respiratory function monitor. Video recording were used (i) to identify infants who were intubated in the delivery room and (ii) to observe colour change of the ETCO(2) detector. Flow sensor recordings were used to confirm whether the tube was in the trachea or not.
RESULTS: Of the 210 infants recorded, 44 infants were intubated in the delivery room. Data from 77 intubation attempts were analysed. In 35 intubations of 20 infants both a PediCap® and flow sensor were available for analysis. In 21 (60%) intubations, both methods correctly identified successful ETT placement and in 3 (9%) both indicated the ETT was not in the trachea. In the remaining 11 (31%) intubations the PediCap® failed to change colour despite the flow wave indicating correct ETT placement.
CONCLUSION: Colorimetric CO(2) detectors may mislead clinicians intubating very preterm infants in the delivery room. They may fail to change colour in spite of correct tube placement in up to one third of the cases.

Assessment of flow waves and colorimetric CO2 detector for endotracheal tube placement during neonatal resuscitation
Resuscitation. 2011 Mar;82(3):307-12

End tidal CO2 in cardiac arrest

Measuring end-tidal CO2 in cardiac arrest patients is helpful for

  1. confirming tracheal tube placement
  2. assessing the effectiveness of chest compressions
  3. predicting likelihood of return of spontaneous circulation (ROSC), in that a persistently low ETCO2 tends to predict death, whereas a high or rising ETCO2 is associated with a higher chance of ROSC.

It may be however that its predictive ability depends on the type of cardiac arrest, and how far into the resuscitation you’ve got when you measure the ETCO2. Consider this new study from Slovenian pre-hospital emergency physicians:

Methods: The study included two cohorts of patients: cardiac arrest due to asphyxia with initial rhythm asystole or pulseless electrical activity (PEA), and cardiac arrest due to arrhythmia with initial rhythm VF or pulseless VT. The causes of asphyxia were: asthma, severe acute respiratory failure, tumor of the airway, suicide by hanging, acute intoxication, pneumonia and a foreign body in the airway.PetCO2 was measured for both groups immediately after intubation and repeatedly every minute, both for patients with or without return of spontaneous circulation (ROSC). We compared the dynamic pattern of PetCO2 between groups. Resuscitation procedures were performed by an emergency medical team (emergency medical physician and two emergency medical technicians or registered nurses) in accordance with 2005 ERC Guideline

Results: Between June 2006 and June 2009 resuscitation was attempted in 325 patients and in this study we included 51 patients with asphyxial cardiac arrest and 63 patients with VF/VT cardiac arrest. The initial values of PetCO2 were significantly higher in the group with asphyxial cardiac arrest (6.74 ± 4.22 kilopascals (kPa) versus 4.51 ± 2.47 kPa; P = 0.004). In the group with asphyxial cardiac arrest, the initial values of PetCO2 did not show a significant difference when we compared patients with and without ROSC (6.96 ± 3.63 kPa versus 5.77 ± 4.64 kPa; P = 0.313). We confirmed significantly higher initial PetCO2 values for those with ROSC in the group with primary cardiac arrest (4.62 ± 2.46 kPa versus 3.29 ± 1.76 kPa; P = 0.041).

A significant difference in PetCO2 values for those with and without ROSC was achieved after five minutes of CPR in both groups (asphyxial arrest: 6.09 ± 2.63 kPa versus 4.47 ± 3.35 kPa; P = 0.006; primary arrest: 5.63 ± 2.01 kPa versus 4.26 ± 1.86; P = 0.015)

In mmHg, the PetCO2 values for those with and without ROSC after five minutes of CPR was: asphyxial arrest: 42.3 ± 20 mmHg versus 34 ± 25.5 mmHg; P = 0.006; primary arrest: 42.8 ± 15.3 mmHg versus 32.3 ± 14.1 mmHg; P = 0.015

Graphically, this difference in ROSC vs non-ROSC PetCO2 for both groups appeared to be even greater at ten minutes, with higher statistically significance (p<0.001), although the values of PetCO2 are not given in the paper.

In all patients with ROSC the initial PetCO2 was again higher than 1.33 kPa (10.1 mmHg).

Conclusions: The dynamic pattern of PetCO2 values during out-of-hospital CPR showed higher values of PetCO2 in the first two minutes of CPR in asphyxia, and a prognostic value of initial PetCO2 only in primary VF/VT cardiac arrest. A prognostic value of PetCO2 for ROSC was achieved after the fifth minute of CPR in both groups and remained present until final values. This difference seems to be a useful criterion in pre-hospital diagnostic procedures and attendance of cardiac arrest.

The authors summarise with the following key messages:

  • Initial values of PetCO2 are higher in asphyxial cardiac arrest than in primary cardiac arrest.
  • Initial values of PetCO2 in asphyxial cardiac arrest do not have a prognostic value for resuscitation outcome.
  • The prognostic value of PetCO2 for ROSC was achieved after the fifth minute of CPR in both groups and remained present until the final values.
  • The values of PetCO2 seem to be useful in differentiating the causes of cardiac arrest in a pre-hospital setting.

I think that last one’s a bit of a stretch. For me, this paper confirms that the longer you are into a cardiac arrest resuscitation, the worse news a low PetCO2 is. The lack of predictive value of initial PetCO2, particularly in the asphyxia group, is interesting but not surprising.

The dynamic pattern of end-tidal carbon dioxide during cardiopulmonary resuscitation: difference between asphyxial cardiac arrest and ventricular fibrillation/pulseless ventricular tachycardia cardiac arrest
Critical Care 2011, 15:R13


One for the ‘hardly surprising’ category….

A study of end-tidal CO2 during out-of-hospital adult and child cardiac arrest resuscitation showed a sudden rise in CO2 was associated with return of spontaneous circulation (ROSC), suggesting that witnessing this would be a good time for a pulse check. Data from the 59 patients who achieved ROSC are shown below, time zero being time of ROSC. There was no such observed rise in the 49 patients who did not achieve ROSC.

A Sudden Increase in Partial Pressure End-Tidal Carbon Dioxide (PETCO2) at the Moment of Return of Spontaneous Circulation
The Journal of Emergency Medicine, Vol. 38, No. 5, pp. 614–621, 2010

ETCO2 all over the place in trauma

In 180 intubated trauma patients in the ED, there was little correlation between arterial carbon dioxide tension (PaCO2) and end-tidal carbon dioxide levels (ETCO2) (R2 = 0.277). In fact, in those patients ventilated to the ‘normal range’ of 35-40 mmHg (4.6-5.2 kPa), PaCO2 was over 50 mmHg 30% of the time. Slightly reassuring that in isolated brain injury the correlation was better (r2 = 0.52)

The Utility of Early End-Tidal Capnography in Monitoring Ventilation Status After Severe Injury
J Trauma. 2009 Jan;66(1):26-31