Assessing Fluid Responsiveness: SIGH35 vs. EEOT in ICU

SIGH₃₅ and end-expiratory occlusion test for assessing fluid responsiveness in critically ill patients undergoing pressure support ventilation

What was it?

A prospective study conducted in a general ICU in Milan, Italy, to assess the reliability of two haemodynamic tests, SIGH35 and end-expiratory occlusion test (EEOT), to predict fluid responsiveness in critically ill patients on PSV.

PSV creates a challenge for haemodynamic tests, as it alters heart-lung physiological interplay where the presence of some spontaneous breathing renders the tests more difficult to interpret compared to those receiving controlled mechanical ventilation.

What are SIGH35 and EEOT?

They were first described in ICU patients receiving mechanical ventilation and demonstrated a high positive predictive value for assessing fluid responsiveness. SIGH35 is a ventilator generated sigh at 35cmH20 for 4 seconds with changes in pulse pressure (PP) noted. It utilises the Hering-Breuer reflex whereby inhibition of inspiration occurs when pulmonary stretch receptors detect overinflation therefore prolonging expiration and reducing interference from spontaneous inspiratory efforts.

EEOT involves stopping the ventilator at PEEP for a short time period, during which the venous return and thus RV preload improves, which can contribute to an overall rise in stroke volume (SV). It can thus indicate biventricular preload responsiveness.

The Devil in the details!

60 ICU patients over a 19 month period (February 2022 to September 2022 and January 2024 to November 2024).
Patient selection was based on whether they were receiving PSV with varying degrees of inspiratory effort (quantified by airway occlusion pressures (p0.1) as a marker for respiratory drive and requiring volume expansion (VE).
MOSTCARE cardiac output (CO) monitoring was used to acquire haemodynamic variables and arterial waveform analysis was performed to assess beat to beat measurements, to gather information on MAP, PP variation and SV.
The protocol was initiated during a stable ventilatory period and the two tests were applied to each patient, prior to fluid bolus but randomised in order to avoid bias. A 3 minute stabilisation period was included in between both tests.
Baseline measurements (including patient characteristics and clinical status) were taken followed by the two tests and then the final set at the end of VE (defined as 4ml/kg crystalloid over 10 minutes). Patients were classified as fluid responsive if CO ≥ 10% after VE administration.
Statistical analysis via Area Under the Receiving Operating Characteristic (ROC) curve (AUC) was used to assess diagnostic performance of the study and comparison of the two tests.

The results!

Out of 60, 56 patients were analysed. 3 patients were excluded due to the presence of extrasystoles in the beat to beat analysis of SIGH35 and 1 patient due to ventilation triggered during EEOT.

SIGH35 showed significantly better reliability in predicting fluid responsiveness vs EEOT overall.

For SIGH35:

AUC for PP changes after SIGH35=0.93 indicating high accuracy (95% CI, p<0.001, 93.1% sensitivity, 91.6% specificity).
It predicted correct fluid responsiveness in 51 out of 56 patients (ie 91% accuracy) by using the threshold of -25% PP reduction from baseline.
A mean reduction of PP of -44.6% was observed in responders.

For EEOT:

AUC for CO changes after EEOT= 0.67 indicating moderate accuracy (95% CI, p= 0.003, 72.4% sensitivity, 70.3% specificity)
Correctly identified 38/56 patients (67.9%) with 18 patients (32.1%) being wrongly classified.
Best threshold for CO = 4% from baseline
The difference in CO increase between responders and non-responders was not statistically significant. This lack of a significant difference in CO response between the groups contributed to the relatively low AUC for the EEOT in the overall population.

Subgroups with low Resp effort:

24 patients (42.8%) showed p0.1 ≤1.5cmH20 at baseline.
AUC of PP changes after SIGH35 =0.98 (95% CI, p<0.001, 92.8% sensitivity, 100% specificity).
AUC of CO changes after EEOT = 0.89 (95% CI, p=0.001, 100% sensitivity, 80% specificity.
The AUCs of the two tests were therefore comparable (p=0.26).
High accuracy: SIGH35 correctly identified 22 of 24 patients (91.5%) and EEOT correctly identified 21 of 24 patients (87.5%)

They concluded…

SIGH35 reliably predicted fluid responsiveness by assessing percentage PP change with a best threshold of -25% reduction from baseline in ICU patients undergoing PSV. While SIGH35 outperformed EEOT, the latter is still reliable where there is a small extent of spontaneous inspiratory effort.

Gripe point summary!

Unique study which addresses the challenges of assessing fluid responsiveness in those receiving PSV with spontaneous respiratory effort which has not extensively been studied to date.
It is known that patient effort can cause EEOT failure but the extent of this and how it affects the test’s accuracy has not been thoroughly investigated.
The study highlights the use of p0.1 to measure the extent of inspiratory efforts and therefore improve the reliability of the test. While EEOT has a reported failure rate of 22.5% due to patient’s effort against an occluded airway, the study shows that it remains reliable if p0.1≤1.5cmH20.
With varying degrees of inspiratory effort, EEOT is less reliable than SIGH35 due to smaller AUC.

Limitations

Generalisability is limited due to single centre study and strict patient selection criteria
Potential for carry-over of haemodynamic effects when performing each test however bias is reduced by randomising the sequence and the inclusion of the 3 minute stabilisation period between tests to mitigate against this
MOSTCARE system is uncalibrated therefore its use may be viewed as debatable however they examined percentage changes which is more robust than absolute numbers.
Underpowered for certain subgroups eg obese patients
SIGH35 can cause coughing and transient distress in awake patients but the authors have suggested that this can be managed by optimising ventilator settings

Our summary

Overall, SIGH35 is a more reliable indicator of fluid responsiveness but the patient’s own breathing effort needs to be examined. If there is a small degree of spontaneous breathing activity then EEOT is just as good as SIGH35, but if there is a lot of spontaneous effort then EEOT is not as robust. Therefore, EEOT can be reliably used if p0.1 is low otherwise SIGH35 is a reliable form of measurement. We still require a deeper insight into how individual factors can affect our decision making when optimising fluid status, but the study does provide us with a good starting point.

Who’s worked on this before?

Sigh maneuver to enhance assessment of fluid responsiveness during pressure support ventilation Messina A, et al. Critical Care. 2019. (Introduces the sigh maneuver under PSV, showing high predictive value for PP variation; foundational for SIGH35.)

The end-expiratory occlusion test for detecting preload responsiveness: a systematic review and meta-analysisPréau S, et al. Annals of Intensive Care. 2020. (Meta-analysis validating EEOT’s reliability in mechanically ventilated patients, with AUC ~0.90.)

End-Expiratory Occlusion Test Predicts Fluid Responsiveness in Patients Receiving Positive End-Expiratory Pressure Gavelli F, et al. Anesthesia & Analgesia. 2018. (Demonstrates EEOT’s accuracy in protective ventilation settings, with >5% SV increase as threshold.)

Prediction of Fluid Responsiveness Using Combined End-Expiratory and End-Inspiratory Occlusion Tests in Elective Cardiac Surgical Patients in the Intensive Care Unit Dong TS, et al. Journal of Clinical Medicine. 2023. (Combines EEOT with other tests in post-op ICU patients, highlighting improved preload assessment.)

End-expiratory occlusion manoeuvre does not accurately predict fluid responsiveness in the operating theatre Biais M, et al. British Journal of Anaesthesia. 2017. (Explores limitations of EEOT in surgical contexts, noting challenges with spontaneous breathing.)

Using the ventilator to predict fluid responsiveness Monnet X, Teboul JL. Intensive Care Medicine. 2024. (Reviews ventilator-based tests including sigh and EEOT under PSV, discussing interference from spontaneous efforts.)

Written by Dr Anusha Roopram

Peer reviewed by JW