Rationale: to assess an adequate cardiovascular function during anaesthesia.

Our challenge is to achieve the most adequate haemodynamic monitoring for a specific setting, optimizing the cardiovascular function and improving patient outcome.

Haemodynamic monitoring provides dynamic measures of cardiovascular system changes, in real time.

During anaesthesia, monitoring target is to guarantee an adequate tissue perfusion and oxygen delivery; predict and correct all causes of instability, avoiding irreversible organ failure and formulate the next step of therapy.

Different conditions could lead to haemodynamic instability: heart failure, fluid shift, hypovolemia and vascular tone modification.

In simple terms, we use dynamic measures to determine if the cardiac output (CO) is adequate or not and if it will increase with specific treatments (fluid administration, vasoactive or inotropic drugs use). Timely an adequate therapeutic strategy is started, the monitoring tool can assess the answer to therapy.

A combination of clinical examination, prior assessment of therapeutic strategy and the treatment response is often called “dynamic or functional monitoring” [1].

Haemodynamic basic measures and CO monitoring can provide information regarding the depth of anaesthesia and the pain control adequacy (ex: sudden increase in blood pressure and heart rate). Perioperative dynamic monitoring (GDT) involves the optimization of tissue oxygen delivery and allows to decrease the incidence of complications, in-hospital length of stay and mortality [2–4].

Despite improvements in technologies, at the moment there is not a haemodynamic monitoring device which can quantify/measure the whole haemodynamic patient assessment.

Many different tools are available. Every system has its own features and also limitations.

Devices which measure systemic blood pressure, heart rate and cardiac output can be extremely basic and non-invasive, but less accurate for some critical setting (ex: poor peripheral perfusion and vasoconstriction).

Minimal invasive (arterial catheter) and more invasive (central venous line and pulmonary artery catheter) devices directly allow measuring cardiac output. They can require time for positioning and lead to some complications.

Between these two categories of devices there are some monitoring systems which indirectly measure the CO and the fluid responsiveness. They can be used in less critical patients providing cardiac output monitoring, during anaesthesia.

Measuring arterial blood pressure (AP) is a cornerstone of haemodynamic assessment.

Blood pressure may be measured non-invasively with a cuff placed around a limb and attached to a sphygmomanometer or an oscillometric device. The oscillometric device measures systolic and diastolic pressure and the mean arterial blood pressure (MAP) through an algorithm. These devices correlate with the old mercury column system. Oscillometric techniques are less accurate in patients with arrhythmia or if the limb cuff is not well positioned; too small or too tight cuff can overestimate pressure values, on the other hand too wide and too big cuff can underestimate it [5].

Despite being easy to perform, the arterial pressure and the heart rate measures are very difficult to examine. Even if a very low value of blood pressure is often matched with a tachycardia, a normal blood pressure value is not always a haemodynamic stability index [6].

Hypotension can be due to an autonomic nervous system inability to balance the decreased cardiac output and the anomalous oxygen delivery.

The degree of the hypotension value can be different according to patient age, depth of anaesthesia, anaesthetic drug effect on haemodynamic status, pain control, and patient comorbidities. Under general condition, if CO decreases, baroreceptor activity tries to increase the sympathetic tone, leading to an increase in heart rate and vascular tone to restore the mean arterial pressure. So patient could have a sudden haemodynamic instability with a low CO, before hypotension appears [7].

Arterial blood pressure alone is a late marker of haemodynamic instability; if we consider simultaneously the heart rate, we can have information of the haemodynamic status.

During low blood flow or fluid loss, HR and non-invasive AP can detect haemodynamic changes.

Recent clinical trials evaluated the accuracy of less invasive devices which can continuously monitor the non-invasive blood pressure [8] and the cardiac output measured by arterial waveform or plethysmography analysis, in the operating room during anaesthesia [9].

Clinical trials suggest that non-invasive devices with more complex algorithm have a better performance and can be used in selected cases [10].

The gold standard for the arterial blood pressure monitoring is the invasive measure, especially in patients with haemodynamic derangement, needing controlled hypotension or increased organ perfusion or multiple arterial blood gas analysis [11].

Radial artery is the most used for cannulation, because it is rapid to detect and it has lower complications. You need a 20 G artery cannula with Seldinger technique or ultrasound-guidance or direct cannulation.

Allen test can be performed to evaluate the collateral arterial circulation, but with low sensitivity. Complication of this procedure can be: thrombosis, arterial spasm, distal embolism, infection, blood loss, and accidental drug injection.

The arterial blood pressure measurement is done by a non-squeezable closed circuit with a pressure transducer which transforms mechanical pulse in electrical one, visible on monitor screen. The zero of the system is done with the transducer at atmosphere pressure, positioned at the right atrium level or at the Willis circle (surgical procedure in sitting position) [12].

Central venous catheter and especially the pulmonary artery catheter (PAC) are the most invasive system of monitoring. They provide characteristic haemodynamic information as no other system can do. PAC is inserted through a central venous access (generally right internal jugular or left subclavian vein) and air inflation into the cuff on the tip of the catheter let it sail through the right heart till the pulmonary artery.

Complications are due to attempt of venous punctures and the passage in the right heart sections. The correct positioning in a pulmonary arterial branch is confirmed by the waveform on the monitor (pulmonary capillary Wedge Pressure Waveform—Pcwp).

Although not perfect, the pulmonary artery catheter has long been considered the optimal form of haemodynamic monitoring. Recent clinical trials have not confirmed the clinical effectiveness of its use [13, 14].

However in these trials no clinical target was selected and data have been often misunderstood. It is important to emphasize that the PAC insertion did not affect patient mortality [15] and it is the unique system that can continuously monitor the cardiac output, the mixed venous oxygen saturation (SvO₂), the intra-thoracic vascular pressure and the oxygen delivery (DO₂).

Shoemaker proposed a pre-operative optimization of haemodynamic values to improve post-operative outcomes, based on DO₂ 600 mL/min/m², haemoglobin 11 g/dL, Pcwp 12 mmHg and targeted inotropic drugs to maximize the oxygen delivery [16–19].

However elevation of DO₂ values to improve haemodynamic status is not always resolving strategy and it can also be harmful [20–22].

Assessing a peri-operative optimization, PAC has to answer the question: could cardiac output provide an adequate oxygen delivery to satisfy metabolic tissue demand? If DO₂ is inadequate, tissue oxygen extraction is increased, with a consequent SvO₂ reduction (<70%). In this setting, low SvO₂ and high intra-thoracic vascular resistances, the cardiac output detects the instability status, leading to a specific therapy. A high CO value with low MAP can show a distributive shock status. On the other hand, a low cardiac output shows: hypovolemic shock (low right atrial pressure CVP and low Pcwp), cardiogenic shock (high CVP and high Pcwp), obstructive shock (CVP > Pcwp and high main pulmonary artery pressure MPAP).

In the presence of haemodynamic instability, the key point is to determine if the cardiac output will increase with fluid administration (preload dependent patient).

The assessment of preload and fluid responsiveness is crucial also in cardiogenic shock [23].

Static pressure (CVP and Pcwp) traditionally have been used to guide fluid management, but they are a poor predictors of fluid responsiveness: low value shows a “poor fluid filling” and high value an “adequate fluid filling”.

Since they were rigorously designed, pulmonary artery occlusion pressure and central venous pressure fail to predict ventricular filling volume or fluid responsiveness, with a 50% reliability [24, 25].

The authors suggest that intra-operative or intensive care unit monitoring, with a single measurements of CVP, is not predictive of the patient’s volemic status and it should not be used [26].

Even if the pulmonary artery catheter can predict the relationship between cardiac output and metabolic demand, it cannot predict the fluid challenge response.

Dynamic measures such as Stroke Volume Variation (SVV) are more accurate than static measurements for assessing fluid responsiveness in mechanically ventilated patients, during anaesthesia.

Stroke volume is the difference between the maximum and the minimum stroke volume over the main stroke volume measured at the same time, over consecutive mechanical breath. During positive pressure inspiration, the increased intra-thoracic pressure is associated with decreased venous return to the right ventricle (RV) and consequent RV cardiac output reduction (RV is preload-dependent). After 2–3 beat time, left ventricle (LV) stroke volume decreases due to reduced RV filling. These changes in LV stroke volume are most marked when a patient is hypovolemic. Given that the pulse pressure variation (PPV) varies beat-to-beat according to the SVV, PVV measure reflects the stroke volume variation [27–29].

Stroke volume variation and pulse pressure variation are specific and sensitive predictor of fluid responsiveness.

An SVV>15% in patients during mechanically ventilation with tidal volume >8 mL/kg or >10% with tidal volume 6 mL/kg predicts a fluid responsiveness [30–32].

Pulmonary artery catheter cannot quantify the SVV, and in the operating room, clinicians can use two alternative devices to get these information: pulse contour analysis (arterial waveform analysis) and oesophageal-Doppler.

The more experienced calibrated device uses the transpulmonary thermodilution. A bolus of 20 mL cold saline (<8 °C) is injected into the right atrium via a venous central line and the thermal profile is registered in a central artery (femoral). This calibration method does not recommend the use of the radial artery, so the catheter must be placed in the femoral, axillary or brachial artery.

This system provides the measurement of the CO, the global end-diastolic volume (GEVD), dynamic indices (PPV and SVV). Although some studies have shown that GEDV can be superior to other pressure static measurements in predicting response to preload [34, 35] other studies have however shown a lower correlation.

However, the effectiveness of SVV and PVV in the intraoperative fluid-management has been confirmed [36]. To ensure the accuracy of the continuous measurement of CO, it is important to calibrate the system every 8 hours or if changes in clinical status occur.

An alternative method to measure the cardiac output is using the lithium dilution to calibrate this device. Unlike the transpulmonary thermodilution techniques, the dilution with lithium does not require a central line. Strictly speaking, this method uses the analysis of the pulse oscillatory power to provide continuous cardiac output data.

The analysis of the pressure waves power converts the arterial wave form into a “volume-time wave” using an autocorrelation to determine the stroke volume [37]. The analysis of the pulse power is less influenced by the reflected waves and by the variations in the transducer set-up because it is less dependent on the pulse wave form.

The system needs a catheter in the radial artery, allowing for monitoring from more conventional arterial access site. As for transpulmonary system, you can measure the CO, the PPV and SVV and the system should/must be recalibrated after 8 h.

The last calibrated device based on the pulse contour analysis is a new hemodynamic platform that uses the transpulmonary thermodilution tool for calibration. It provides the same parameters than other devices, despite its greater precision in the measurement of some pulmonary hemodynamic parameters (extravascular lung water) [38].

Among non-calibrated systems, that achieved success for their easy use, a special mention is for the system provided by a transducer, easily and quickly connectable to any arterial line already placed, allowing the measurement of CO. It does not need calibration for calculating the CO, but it is necessary to insert patient’s age, height, sex and weight into the system, to determine the cardiac output from the pulse contour analysis.

Not surprisingly, in critical situations/setting, a non-calibrated system is not so reliable [39], although the third generation software shows a marked improvement in performance [40]. The measurement of SVV and PPV does not depend on accurate calibration, but when it is used to guide the perioperative fluid administration, non-calibrated system reduces complications in major surgery [41, 42].

The second not-calibrated device records the pressure value with analytical method. It is a technique designed for the continuous CO monitoring, derived from the blood pressure without initial calibration or central venous catheterization. Therefore, the system requires only an arterial line without a dedicated pressure transducer.

The technology of this system is based on the principle that in a vessel the volume variations mainly occur due to the radial expansion of pressure variations; so alterations of the systolic portion of the area under the pressure curve reflect the stroke volume variations [43].

This technique calculates the CO by physical parameters, such as left ventricle ejection strain, arterial impedance opposing the blood flow pulsatility, arterial compliance and peripheral vascular resistances [44]. The sampling frequency of this system is 1000 Hz, compared to other methods that sample at 100 Hz. The system captures 1000 times/s, compared to other methods that capture 100 times/s. A higher sampling frequency should allow for a greater precision of the measured data (CO, SVV, PPV and SVR).

Overall, these devices are very effective in predicting the preload responsiveness and protocols guided by SVV or PVV for the haemodynamic optimization, all lead to improvements in surgical outcome [45, 46].

However, it is important to remember that the SVV and PPV require the chest to be closed, to predict the preload responsiveness, although the one-lung ventilation in thoracic surgery does not compromise the predictive ability of these techniques [47].

The presence of arrhythmias or atrial fibrillation is a bias in the measurement of SVV and PPV; in fact, they seem to be a result of cyclical changes in the ventricular filling, rather than a cyclical changes due to mechanical ventilation. In these situations, the dynamic indices/parameters are unable to predict the preload/fluid responsiveness. Unlike non-calibrated systems, calibrated monitoring tool provides CO measurements that correlate with PAC measures [48, 49].