Background:

Fluid administration is the first line of treatment in patients with acute circulatory failure. While hypovolemia affects tissue oxygenation leading to organ dysfunction and death¹, excessive fluid loading is associated with increased complications, mortality and length of intensive care unit (ICU) stay ^{2, 3}. Only half the patients with circulatory failure respond positively to fluid administration⁴. Hence it is necessary to detect fluid responders. Traditionally used static hemodynamic parameters like central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP) are of limited value in predicting fluid responsiveness in critically ill patients^5-8. Various dynamic indices which depend on cardiopulmonary interactions like stroke volume variation (SVV) and pulse pressure variation (PPV) are superior to static indices and shown to accurately predict fluid responsiveness ^9-11. The major limitation for using PPV and SVV are low tidal volume (Vt) ventilation i.e. Vt ≤6 mL/kg predicted body weight (PBW) ^{12, 13} and poor lung compliance (Crs is <30 mL/cmH2O) ¹⁴. Low tidal volume ventilation is commonly used in ICU today, it is recommended not only in patients with adult respiratory distress syndrome (ARDS) ¹⁵, but also in patients at risk of ARDS, with septic shock, potential organ donors, and patients undergoing high risk intra- abdominal and thoracic surgeries^{16, 17}.

During low Vt ventilation the magnitude of change in airway driving pressure may not be enough to produce a adequate change in intra-cardiac pressure. Thus PPV and SVV may be low even in fluid responders. To overcome this limitation with the use of PPV and SVV during low tidal volume we developed a new test called the “tidal volume challenge” ¹⁸. This test involves transiently increasing the tidal volume from 6mL/kg predicted body weight (PBW) to 8mL /kg PBW for one minute and thereafter reducing the tidal volume back to 6mL/kg PBW. The change in PPV and SVV following the “tidal volume challenge” (i.e. ΔPPV_6-8 and ΔSVV_6-8) reliably predicts fluid responsiveness in patients ventilated at low Vt with cut off values of 3.5% and 2.5 % respectively.

The PPV and SVV depends not only on the magnitude of driving pressure but also on how much of this driving pressure is transmitted to pleural and pericardial cavity, which in turn is proportional to  lung compliance. Thus PPV and SVV do not reliably predict fluid responsiveness when the lung compliance is low. Monnet et al have shown that PPV was the unreliable when the respiratory system compliance was <30 mL/cmH2O and predictability varies with the lung compliance¹⁴. In our study on 19 out of the 30 occasion, patients had low respiratory compliance (<30 mL/cmH2O) ¹⁸. In all 19 situations PPV and SVV reliably predicted fluid responsiveness after the “tidal volume challenge”. Thus our study suggested that the “tidal volume challenge” may also help improve the reliability of PPV and SVV in patients with low compliance who are ventilated using low Vt. Thus, we would like to test whether “tidal volume challenge” can improve the reliability of PPV and SVV in patients with low compliance of the respiratory system.

Another test commonly used to predict fluid responsiveness is the variability in inferior vena cava diameter using 2D echocardiography. Inferior vena cava diameter variability (IVC-V) is routinely measured in all patients with shock, during screening using echocardiography which is calculated as IVC distensibility index (IVC-DI) during controlled ventilation and has been shown to reliably predict fluid responsiveness in mechanically ventilated patients ^19-21.

The IVC-V depends on the variations in intrathoracic pressure, abdominal pressure and compliance of the vessel. Studies that tested IVC-DI used tidal volume > 8 mL/kg PBW²¹. When Vt less than 8 mL/kg is used, there may be smaller variations in intrathoracic blood volume and pressure leading to smaller variations in IVC dimensions, thus leading to false negative results irrespective of the volume status²². Whether IVC distensibility index reliably predict fluid responsiveness in patients ventilated with low tidal volume (≤6mL/kg PBW) has not been tested. A recent study that compared various dynamic parameters used to predict fluid responsiveness, in large population of patients receiving controlled ventilation, showed poor performance of IVC-DI in predicting fluid responsiveness²³. In this study 66% patients had lower tidal volume i.e. < 8mL/kg. Thus the IVC-DI may also have the same limitation as PPV and SVV to predict fluid responsiveness during low Vt ventilation. We would like to determine the reliability of IVC-DI at low Vt and also whether the “tidal volume challenge” can improve the reliability of IVC-DI in predicting fluid responsiveness in patients receiving low Vt ventilation.

OBJECTIVE:

1.      To determine whether the “tidal volume challenge” can improve the reliability of PPV and SVV in predicting fluid responsiveness in patients having low lung compliance ventilated using low Vt ventilation.

2.      To determine whether IVC-DI can reliably predict fluid responsiveness in patients receiving low Vt ventilation

3.      To determine whether the “tidal volume challenge” can improve the reliability of IVC-DI in predicting fluid responsiveness in patients receiving low Vt ventilation.

PATIENTS AND METHOD:

This prospective study will be conducted at mixed medical and surgical Intensive Care Unit, Tata Memorial Hospital, Mumbai, INDIA 

Inclusion Criteria

• Adult patients   (Age > 18 yrs )
• Acute circulatory failure
• Receiving  protective lung ventilation  ≤ 6ml/kg IBW using Volume Assist Control mode, without any spontaneous activity
• With  cardiac output monitoring using Flotrac/ Volume View with Phillips Intel View monitor (MP700) 
• Treating physician has decided to give fluid bolus

The Exclusion Criteria:

• Cardiac arrhythmias, Acute myocardial infarction
• Previously known significant valvular disease or intracardiac shunt
• Air leakage through chest drains
• Right heart failure
• An urgently required fluid challenge
• Abdominal compartment syndrome, and pregnancy

MATERIAL AND METHODS:

All patients enrolled will have either an internal jugular vein or a subclavian vein catheter and a continuous cardiac output monitoring device (The Volume View set from Edwards Life Sciences in combination with EV1000 clinical platform or Flotrac from Edwards Life Sciences). The PPV will be recorded using the Philips Intel View Monitor (MP 70) and SVV will be recorded from cardiac output monitor. The arterial pressure and central venous pressure transducers will be attached to the patient’s upper arm at the level of the cardiac cavities. Central venous pressure will be measured at end expiration. IVC maximum and minimum diameters will be measured using ultrasound machine (SonoSite, M-Turbo, USA) with phased array probe (frequency 1-5MHz). The IVC will be examined subcostally in longitudinal section. Its maximum and minimum diameter will be measured in M-mode coupled to two-dimensional mode, just upstream of the origin of the suprahepatic vein on expiration. Measurements will be considered only when the M-mode tracing is exactly perpendicular to the IVC (Fig 1). All measurement will be done by intensivists trained in echocardiography. The IVC distensibility index (IVC-DI) will be calculated using the following formulae.

1.      maximum diameter – minimum diameter /minimum diameter of IVC × 100

2.      maximum diameter – minimum diameter /mean diameter of IVC × 100

This study will not be performed more than once in a same patient on same day. Not more than two readings per patient will be taken. Demographic data of patients including age, sex, height, weight, predicted body weight (PBW) APACHE II on admission, primary diagnosis and reason for ICU admission of patients will be recorded. Indication for low tidal volume ventilation will be recorded. Dose of vasopressors used if any, will be recorded. The vasoactive medications would be kept constant during the study period. The Positive end expiratory pressure (PEEP) will be kept constant during the study period.Indication for the fluid challenge will be noted. Various Respiratory and hemodynamic parameters will be recorded at various intervals as per study design below. ICU and hospital outcome of the patient will be recorded.

Study Design

Arterial lactate and mixed central venous oxygen saturation values (ScVO₂) will be recorded. All patients will be monitored for 15 minutes at 6 ml/kg PBW for any variability in vital parameters. The following set of hemodynamic variables, i.e. heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), cardiac index (CI), PPV, SVV and CVP and respiratory variables i.e. plateau pressure (Pplat) and static compliance of the respiratory system (calculated using the formula Vt/ Pplat-PEEP)  and inferior vena cava diameters  will be recorded. This set of readings will be recorded at baseline and repeated at various intervals (refer to flowchart). The “tidal volume challenge” will be performed and after one minute a set of readings will be recorded. The tidal volume will be decreased back to 6 ml/kg IBW. After one minute a set of readings will be recorded.

A fluid challenge of 7 ml/kg actual body weight of saline will be given over 10 minutes. Following this a set of readings will be recorded. Arterial lactate and mixed central venous oxygen saturation values will be recorded after the fluid challenge.

The ratio of the heart rate and respiratory rate (HR/RR) and difference between Pplat and PEEP will be calculated for each interval. Patients will be divided into two groups Responders and Non-Responders based on increase in Cardiac index >15% after giving the fluid bolus.

STATISTICAL ANALYSIS 

Demographic, clinical and disease related variable will be presented as frequency (Percentage) and mean (S.D), Median as appropriate. Change in continuous variable from base line 6ml/ kg to TV 8 ml/kg will be compared using paired t test or Wilcoxon signed rank sum test as per distribution of data. Group comparisons will be made using independent t-test or Man Whitney U test as per the distribution of the data for continuous variables. Categorical variables will be analyzed using Chi-square test or Fisher’s exact test (for binary variable).Correlations between the different scales will be compared with Pearson correlation and Crammers V correlation for categorical. ROC Curve will be used to determine the ability of various preload responsiveness indicators to discriminate between responders and non responders. p value < 0.05 will be considered statistical significant.

For PPV our study18 showed an AUC of 0.65 in patients with low lung compliance and IVC distensibility index has an AUC of 0.65 23 to  predict fluid responsiveness. Using the Statstodo computer program to calculate the sample size requirement for comparing two receiver-operating characteristic (ROC) curves with expected areas under the curves of 0.65 (PPV6) and 0.90 (ΔPPV6–8) for PPV in patients with low compliance and AUC 0.65 (IVC6) to AUC 0.9 (IVC8) for IVC distensibility index, assuming an Î± error of 0.05 and power of 80% a sample size of 26 patients/reading respectively are required. A total of 40 consecutive patients/readings will be included in the study as it may not be feasible to use both PPV and IVC distensibility index in all patients and also to account for a few dropouts.