Skip to content

Advertisement

  • Review
  • Open Access

Estimating mean circulatory filling pressure in clinical practice: a systematic review comparing three bedside methods in the critically ill

  • Marije Wijnberge1, 2, 3,
  • Daniko P. Sindhunata1,
  • Michael R. Pinsky4Email author,
  • Alexander P. Vlaar2, 3,
  • Else Ouweneel1,
  • Jos R. Jansen5,
  • Denise P. Veelo1 and
  • Bart F. Geerts1
Annals of Intensive Care20188:73

https://doi.org/10.1186/s13613-018-0418-2

Received: 8 February 2018

Accepted: 15 June 2018

Published: 20 June 2018

Abstract

The bedside hemodynamic assessment of the critically ill remains challenging since blood volume, arterial–venous interaction and compliance are not measured directly. Mean circulatory filling pressure (Pmcf) is the blood pressure throughout the vascular system at zero flow. Animal studies have shown Pmcf provides information on vascular compliance, volume responsiveness and enables the calculation of stressed volume. It is now possible to measure Pmcf at the bedside. We performed a systematic review of the current Pmcf measurement techniques and compared their clinical applicability, precision, accuracy and limitations. A comprehensive search strategy was performed in PubMed, Embase and the Cochrane databases. Studies measuring Pmcf in heart-beating patients at the bedside were included. Data were extracted from the articles into predefined forms. Quality assessment was based on the Newcastle–Ottawa Scale for cohort studies. A total of 17 prospective cohort studies were included. Three techniques were described: Pmcf hold, based on inspiratory hold-derived venous return curves, Pmcf arm, based on arterial and venous pressure equilibration in the arm as a model for the entire circulation, and Pmcf analogue, based on a Guytonian mathematical model of the circulation. The included studies show Pmcf to accurately follow intravascular fluid administration and vascular compliance following drug-induced hemodynamic changes. Bedside Pmcf measures allow for more direct assessment of circulating blood volume, venous return and compliance. However, studies are needed to determine normative Pmcf values and their expected changes to therapies if they are to be used to guide clinical practice.

Keywords

Blood pressureVenous pressureBlood volumeIntensive careCritical careHemodynamics

Background

It is difficult to determine the cause for hemodynamic instability in patients and to predict the best treatments. Currently, cardiovascular resuscitation options are triggered by arterial pressure and cardiac output (CO) measures, focusing on the oxygen delivery side of the circulation. However, the primary determinants of CO reside on the venous side. Veins are 30–50 times more compliant than arteries and contain approximately 75% of the total blood volume [15]. Mean circulatory filling pressure (Pmcf) provides vital information on this “forgotten venous side of the circulation” [6].

In 1894, Pmcf was defined as the equilibrium pressure throughout the circulation during circulatory arrest [7]. In the 1950s, Guyton and colleagues described a linear relationship between venous return (VR) and right atrial pressure (Pra), described as: VR = (Pmcf − Pra)/(RVR) [8, 9]. RVR is resistance to VR and defines the slope of the VR curve. This linearity has been confirmed in intact circulations in animal studies and is not affected by hypo- or hypervolemia [1015]. VR curves enable to determine the equilibrium point of the circulation, which is the intersection between the CO and VR curve. Central venous pressure (CVP) is a surrogate of Pra used in clinical practice. CVP at zero flow equals Pmcf (Fig. 1).
Figure 1
Fig. 1

The venous return curve (a) combined with the cardiac output curve (b). The intersection of these two curves (c) is the working point of the circulation. The central venous pressure when venous return equals zero is the Pmcf (d). The slope of the VR is determined by the resistance to venous return

Vascular volume requires a minimal volume before its distending pressure becomes positive. The amount of blood not causing pressure on the vessels is called unstressed volume (Vu) and reflects intravascular volume present with Pmcf of zero. Stressed volume (Vs) is the additional blood causing a distending pressure on the vascular walls and reflects the effective circulating volume. Vu and Vs together define the total blood volume. Vs is approximately 25% of the total blood volume [35]. Vs and vascular compliance (Csys) define Pmcf [16]. An increase in Vs increases Pmcf, and an increase in Csys decreases Pmcf. Fluid loading should increase Pmcf, but VR only increases if the pressure gradient for VR (i.e., Pmcf CVP) increases, RVR decreases, or both. Since in the steady state VR = CO, knowing the determinants of VR is relevant to understanding cardiovascular state.

Recently, methods have emerged to enable clinicians to estimate Pmcf at the bedside. Our objectives for this review were to describe the techniques and to highlight their clinical applicability, precision, accuracy and limitations in critically ill patients.

Materials and methods

Publication selection

This review was performed according to PRISMA guidelines [17] (Additional file 1) and methodology outlined in the Cochrane Handbook for systematic reviews [18]. No study protocol was published. A PubMed, Embase and Cochrane Library database search was performed with help of a clinical librarian with no restriction on publication date. The search was performed up to May 18, 2017. The search strategy combined the following concepts: (1) “mean systemic filling pressure” or “mean circulatory filling pressure” or “static filling pressure” and (2) “intensive care” or “critical care” or “perioperative” or “intraoperative” (Additional file 1). Titles, abstracts and full-texts were independently screened by two reviewers for relevance (MW and DPS), and discrepancies were resolved by a third reviewer (BFG). The references of the selected articles were examined for additional eligible articles. Studies were included when available in English and full-text, described prospective studies in which Pmcf estimation methods were examined in heart-beating ICU patients and contained a description of their clinical applicability, precision and accuracy or limitations.

Data extraction and analysis

Data were extracted into predefined forms. No additional analyses were performed. Critical appraisal was based on the Newcastle–Ottawa Scale for cohort studies [19] to assess the quality of non-randomized studies at study level. A modified version of the scale was used since only five out of nine questions were applicable, resulting in a possible highest score of five stars (Additional file 1).

Results

Study selection and characteristics

The initial search identified 369 articles, of which 300 were excluded after screening title and abstract. A total of 53 articles were excluded based on full-text. Two relevant articles were found by citation tracking. Consequently, 17 prospective cohort studies estimating Pmcf in heart-beating ICU patients were included (Additional file 1). Three different bedside measurement techniques were found. Eight studies estimated Pmcf applying inspiratory hold maneuvers (Pmcf hold), three studies during a circulatory stop-flow in the arm (Pmcf arm) and four studies using a mathematical algorithm (Pmcf analogue). Two studies compared multiple techniques.

Eleven studies were performed in postoperative cardiac surgery patients (Table 1). All patients were hemodynamically stable without alteration in vasopressor use or fluid therapy during the study protocol. All patients were sedated and mechanically ventilated. In one study, spontaneous breathing efforts were observed [20]. The number of included patients ranged from nine to 80. In all studies, CVP was measured via a catheter in the right internal jugular vein. CO measurement techniques differed between studies (Additional file 1).
Table 1

Baseline characteristics for included studies

References

Method

N

Patient population (all adult ICU patients)

Age

Male

Timeframe Pmcf measurement

Maas et al. [21]

Pmcf hold

12

Postoperative cardiac surgery

64 (10)

10 (83%)

Not described

   

10 CABG

   
   

2 AVR

   

Keller et al. [23]

Pmcf hold

9

Postoperative cardiac surgery

Median 61

4 (44%)

Not described

   

3 CABG

IQR 55–75

  
   

6 AVR

   

Maas et al. [22]

Pmcf hold

10

Postoperative cardiac surgery

64 (11)

9 (90%)

Within 1 h after ICU admission

   

2 AVR

   
   

1 MVP + TVP

   
   

7 CABG

   

Persichini et al. [27]

Pmcf hold

16

Septic shock

67 (16)

8 (50%)

Not described

Maas et al. [25]

Pmcf hold

16

Postoperative cardiac surgery

64 (11)

Not described

Within 1 h after ICU admission

   

1 MVP

   
   

15 CABG

   

Guerin et al. [28]

Pmcf hold

30

Shock

65 (12)

21 (70%)

Not described

De Wit et al. [24]

Pmcf hold

17

Postsurgical gastrointestinal

62 (9)

14 (82%)

Not mentioned

   

16 esophageal resection

   
   

1 pancreaticoduodenectomy

   

Helmerhorst et al. [26]

Pmcf hold

22

Postoperative cardiac surgery

63 (59–66)

17 (85%)

1 h after ICU admission

   

22 CABG

   

Geerts et al. [43]

Pmcf arm

24

Postoperative cardiac surgery

64 (10)

19 (79%)

Within 2 h after ICU admission

   

17 CABG

   
   

7 CABG plus valve repair

   

Aya et al. [41]

Pmcf arm

20

Postoperative cardiac surgery

63 (11)

17 (85%)

Initial period at ICU (not further defined)

   

13 CABG

   
   

4 AVR

   
   

4 MVR

   

Aya et al. [42]

Pmcf arm

80

Postoperative cardiac surgery

70

62 (78%)

Initial period at ICU (not further defined)

   

36 CABG

Range 52–80

  
   

27 AVR + CABG

   
   

12 MVR + CABG

   
   

5 Other

   

Parkin et al. [49]

Pmcf analogue

10

Multi-organ failing patients receiving CVVH for acute renal failure

65

7 (70%)

Not described

    

Range 24–77

  

Cecconi et al. [48]

Pmcf analogue

39

22 Cardiac surgery

68 (12)

26 (67%)

Not described

   

8 Shock

   
   

6 Non-cardiac surgery

   
   

3 Other

   

Gupta et al. [20]

Pmcf analogue

61

Postoperative cardiac surgery

63 (11)

46 (75%)

Within 6 h after ICU admission

   

40 CABG

   
   

8 CABG + valve replacement

   
   

8 Valve replacement

   
   

5 Bentall’s procedure

   
   

7 DDD pacing

   

Aya et al. [51]

Pmcf analogue

26

Postoperative fluid challenge

68

16 (62%)

Initial period at ICU (not further defined)

   

7 Cardiac surgery

Range 53–80

  
   

19 Non-cardiac surgery

   

Maas et al. [16]

Pmcf hold

11

Postoperative cardiac surgery

64

9 (82%)

Within 2 h after ICU admission

 

Pmcf arm

11

9 CABG

Range 50–80

  
 

Pmcf analogue

11

2 AVR

   

Maas et al. [30]

Pmcf arm

15

Postoperative cardiac surgery

64 (11)

Not described

Within 1 h after ICU admission

 

Pmcf hold

12

9 CABG

   
   

5 Valve

   
   

1 CABG + valve

   

Age is presented as mean with standard deviation (SD) or median with range or interquartile range (IQR). Number of males per study is presented as counts with percentage

CABG coronary artery bypass, MVR mitral valve replacement, MPV mitral valve prolapse, AVR aortic valve replacement, TVP tricuspid valve prolapse, CVVH continuous veno-venous hemodiafiltration

P mcf hold

Technique description

Pmcf hold is based on the linear relation between CVP and VR (Pmcf = (VR − CVP)/RVR). CVP is raised by performing a series of end-inspiratory hold maneuvers. In 2009, the method was first studied in humans [21]. Inspiratory hold maneuvers at 5, 15, 25 and 35 cmH2O incremental ventilatory plateau pressures (Pvent) were performed, and CO was measured in the last 3 s of the 12 s inspiratory hold. They validated that after 7–10 s a steady state consists when VR = CO. By plotting the CVP and CO values, a VR curve is constructed and the zero-flow pressure (Pmcf) extrapolated. Seven studies [16, 2126] estimated Pmcf hold using these four plateau pressures. Two studies [27, 28] used two points (Pvent 5 and 30 cmH2O) at 15-s inspiratory and expiratory hold plateau phase. Between the Pmcf hold measurements, either 1-min pauses were used to re-establish the initial hemodynamic steady state [16, 21, 22, 24, 28], or the consecutive inspiratory hold was performed when CO had returned to baseline [23, 26, 27].

Clinical applicability

The average baseline Pmcf hold values found in the eight included studies range from 19 to 33 mmHg with a wide standard deviation (Tables 2, 3). Five studies [2123, 26, 28] demonstrated fluid administration caused an increase in Pmcf hold, confirming that in humans, as in animals before [14, 15], Pmcf hold follows hemodynamic changes (Table 2). One of these studies found passive leg raising (PLR) to significantly increase Pmcf hold values [28]. RVR was not significantly affected by different volumetric conditions nor by PLR. Vs was calculated from Pmcf as a measure for effective circulating volume [22]. In one study, Pmcf was used to assess the hemodynamic effects of arterial hyperoxia (FiO2 = 90% for 15 min) in ICU patients [26]. During this hyperoxia, left ventricular afterload increased and contractility remained similar; however, CO did not decrease. Both Pmcf and RVR increased significantly (Table 3), explaining why VR (thus CO) remained unaltered.
Table 2

Mean circulatory filling pressure during different volumetric state

Study

Method

N

Patient population

Baseline position

Baseline Pmcf

Hypervolemia (induced by fluid administration)

p value*

Amount of fluid administered to induce hypervolemia

Hypovolemia (induced by HUT)

p value

Maas et al. [21]

Pmcf hold

12

Cardiac surgery

Supine

18.7 (4.5)

29.1 (5.2)

0.001

500 mL colloid in 15–20 min

14.5 (3.0)

0.005

Keller et al. [23]

Pmcf hold

9

Cardiac surgery

Semirecumbent

19.7

26.9

< 0.05

500 mL colloid

     

IQR 17.0–22.6

IQR 18.4–31.0

    

Maas et al. [22]

Pmcf hold

10

Cardiac surgery

Not described

18.7 (4.0)

26.4 (3.2)

< 0.001

500 mL colloid

Guerin et al. [28]

Pmcf hold

30

Shock

Semirecumbent

Responder: 25 (13)

32 (17)

< 0.01

500 mL saline in 10 min

  
     

Non-responders: 24 (10)

28 (12)

< 0.01

   

Geerts et al. [43]

Pmcf arm

24

Cardiac surgery

Supine

Responders: 16.2 (6.3)

22.0 (7.6)

< 0.001

500 mL colloid

     

Non-responders: 24.3 (8.2)

29.9 (9.1)

< 0.001

   

Aya et al. [41]

Pmcf arm

20

Cardiac surgery

Supine

22.4 (7.7)

Aya et al. [42]

Pmcf arm

80

Cardiac surgery

Supine

23.0

     

Range: 17.3–29.8

     

Parkin et al. [49]

Pmcf analogue

10

CVVH

Not described

Target state = 15.9

CVVHD

Cecconi et al. [48]

Pmcf analogue

39

Heterogenous

Not described

Responders: 17.8 (5.1)

20.9 (5.1)

< 0.001

Mean 252 (8.9) mL

     

Non-responders: 17.9 (5.1)

21.0 (4.9)

< 0.001

52.5% crystalloid

  
        

37.6% colloid

  
        

8.8% FFP & RBC

  

Gupta et al. [20]

Pmcf analogue

61

Cardiac surgery

Supine

Responders: 17 (3.7)

19 (4.3)

0.02

Mean 264 (16) mL

     

Non-responders: 17 (3.6)

19 (4.1)

0.03

50% saline. Other 50%: mix of FFP, platelets, albumin, packed RBC, return of pump blood

  

Aya et al. [51]

Pmcf analogue

26

Heterogenous

Not described

Responders: 13.7 IQR: 10.9–16.9

  

250 mL crystalloid

  
     

Non-responders: 16.7 IQR: 10.5–18.9

     

Maas et al. [16]

Pmcf hold

11

Cardiac surgery

Supine

19.7 (3.9)

28.3 (3.6)

< 0.001

500 mL colloid

16.2 (3.0)

0.001

 

Pmcf arm

   

18.4 (3.7)

27.1 (4.0)

< 0.001

 

15.4 (3.1)

0.001

 

Pmcf analogue

   

14.7 (2.7)

19.2 (1.1)

< 0.001

 

10.9 (2.0)

< 0.001

Maas et al. [30]

Pmcf arm

15

Cardiac surgery

Supine

21.0 (6.8)

27.7 (7.4)

< 0.001

500 mL colloid (10 steps of 50 mL)

 

Pmcf hold#

         

Data presented as mean with SD or median with interquartile range (IQR). Pmcf in mmHg. Hypovolemic state induced by head up tilt (HUT) to 30°. Responders = fluid responsiveness was defined by a 10% increase in CO

p value, difference between baseline and hypervolemia induced by fluid administration

p value, difference between baseline and hypovolemic state

Table 3

Pmcf and pharmacodynamics

References

Method

n

Situation A

Situation B

p value*

Situation C

p value#

Persichini et al. [27]

 

16

NE 0.30

NE 0.19

   
   

Range 0.10–1.40

Range 0.08–1.15

   
 

Pmcf hold (in mmHg)

 

33 (12)

26 (10)

0.003

  

Maas et al. [25]

 

16

Baseline 1

NE increase of 0.04 (0.02)

 

Baseline 2

 
   

NE 0.04 (0.03)

  

NE 0.04 (0.03)

 
 

Pmcf hold (in mmHg)

 

21.4 (6.1)

27.6 (7.4)

< 0.001

22.0 (5.3)

 

de Wit et al. [24]

 

17

Propofol low

Propofol medium

 

Propofol high

 
   

Cb 3.0 (0.90) μg/mL

Cb 4.5 (1.0) μg/mL

 

Cb 6.5 (1.2) μg/mL

 
 

Pmcf hold (in mmHg)

 

27.9 (5.4)

24.6 (4.9)

0.01

21.4 (4.2)

< 0.001

Helmerhorst et al. [26]

 

22

FiO2 21–30%

FiO2 90%

   
 

Pmcf hold (in mmHg)

 

20.8 (3.5)

23.1 (4.0)

< 0.001

  

NE norepinephrine dose in μg/kg/min presented as mean with range or mean with standard deviation. Pmcf values are presented as mean with standard deviation. Cb target blood concentration of propofol in μg/mL. Pmcf hold values presented in mmHg. FiO 2 fractional oxygen concentration

p value, p value for situation A compared to B

#p value, p value for situation A compared to C

Studies have used Pmcf hold to describe hemodynamic changes caused by propofol [24] and norepinephrine [25, 27] (Table 3). In septic shock patients, decreasing the dose of norepinephrine decreased both Pmcf and RVR [27]. Further, after increasing norepinephrine CO decreased in ten patients and CO increased in six patients [25]. In all patients, Pmcf and RVR increased, though the “balance” between the two values determined whether CO increased. One study showed an increase in propofol caused a decrease in Vs without a change in CO [24]. These studies show Pmcf behaves within the framework of hemodynamic reasoning and lends itself to being used as a less invasive method to assess drug-induced physiology. Since Pmcf exists at the intersection of arterial and venous flow, it enables to calculate the true arterial and venous resistance by calculating the critical closing pressure (Pcc). Pcc is the mean arterial pressure (MAP) to zero CO-intercept. Arterial resistance is calculated as (MAP − Pcc)/CO [22].

Precision and accuracy

The technique precision has not yet been assessed in humans. However, in an animal study the averaged coefficient of variation for repeated measurements of Pmcf hold was 6% [29]. Comparing the techniques’ accuracy, no significant differences between Pmcf hold and Pmcf arm existed, whereas Pmcf analogue values were significantly lower [16, 30].

Limitations

The use of Pmcf hold is restricted to mechanically ventilated and sedated patients with a central venous catheter. The procedure of the inspiratory hold maneuvers is not yet automated and requires a direct link between monitor and ventilator, or advanced monitor analytics to detect the inspiratory holds and to perform the instantaneous CO calculations. Furthermore, it is not suitable during cardiac arrhythmia. This method is not suitable to measure rapid changes in hemodynamic status since it takes a couple of minutes to perform the multiple end-inspiratory (and end-expiratory) holds. Potentially, this technique is operator-dependent because a proper inspiratory plateau pressure is needed. CVP can be altered due to incorrect catheter placement. An absolute CO value is not necessary for Pmcf hold as the technique extrapolates to zero CO. If the trend measurements are accurate, the RVR slope might change, but the intersection Pmcf point remains constant. The latter holds only true for the Pmcf itself, the RVR is dependent of the slope of the curve. In clinical practice, a physician would use Pmcf together with RVR; therefore, for clinical use of the Pmcf an accurate CO value is needed.

Potentially, the inspiratory hold maneuver overestimates Pmcf by the blood translocation from the pulmonary into the systemic circulation [3133]. However, the potential volume shifts relative to Csys suggest that this effect is minimal [10, 34]. During inspiratory hold maneuvers, arterial pressure decreases. If sustained, baroreflex-induced increased sympathetic tone may cause Pmcf to increase [35, 36]. Indeed one study performed in pigs found the Pmcf hold overestimating compared to a method using right atrial balloon occlusion in euvolemic conditions, in bleeding and hypervolemia; however, the values found between the two methods were similar [34]. Two clinical studies [16, 30] have shown Pmcf hold and Pmcf arm values not being significantly different, debating the former result found in pigs. Future studies in humans are needed. Moreover, all patients undergoing inspiratory holds are on neuro-humoral suppressive agents, probably dampening the baroreflex and other autonomic influences [3739].

P mcf arm

Technique description

As Pmcf is defined as the steady-state blood pressure during no-flow conditions, instantaneously Pmcf should mainly be similar for different vascular compartments even though each compartment may have different Vu and Vs [2, 40]. Four studies [16, 4143] used the arm to estimate Pmcf. For arm occlusion, a rapid cuff inflator (inflates in 0.3 s) [16, 43] or a pneumatic tourniquet (inflates in 1.4 s) [41, 42] was inflated around the upper arm to 50 mmHg above systolic blood pressure. Arterial and venous pressures were measured via a radial artery catheter and a peripheral venous cannula in the forearm. When these two pressures equalize, Pmcf arm values are achieved. An initial study determined that a 25–30 s stop-flow time was adequate to achieve this equilibration [16]. Following this, in two studies Pmcf arm was measured as the average radial arterial pressure at 30 s after stop-flow [16, 43]. One study found the smallest difference between venous and arterial pressure after 60 s of stop-flow [41]. This discrepancy could be explained by different inflation time, i.e., induction of stop-flow.

Clinical applicability

The average baseline Pmcf arm values found in the included studies range from 16 to 24 mmHg (Table 2). Pmcf arm can be performed in spontaneously breathing subjects and requires only one measure. In two studies, Pmcf arm was assessed as a predictor of fluid loading responsiveness (FLR) [16, 43]. One study showed that a low Pmcf arm (< 22 mmHg) predicts FLR with 71% sensitivity and 88% specificity, where responders were defined when CO increased > 10% after 500 mL colloid administration [43]. Another study showed changes in circulating volume (500 mL colloid) are tracked well by changes in Pmcf arm [16]. Finally, one study indicated a minimum of 4 mL/kg fluid challenge was needed to define FLR [42].

Precision and accuracy

Repeated measurements of Pmcf arm showed no significant differences [41]. The coefficient of variation for a single measurement was 5%, which reduced to 3% after four measurements. Bland–Altman analysis showed a bias of − 0.1 ± 1.68 mmHg for the first two measurements. The least significant change [44] for a single measurement was 14% (i.e., ± 3 mmHg for a Pmcf arm of 22 mmHg). One study observed a negligible bias of two Pmcf arm determinations at baseline position and after fluid expansion [16]. Two studies [16, 30] found no significant differences in Pmcf arm to Pmcf hold measures.

Limitations

Theoretically, a limitation of the technique is the influence of an auto regulatory hypoxia-induced response causing arterial vasodilation. The time of measuring Pmcf after arm occlusion should be enough for arterial and venous pressures to equilibrate, but before hypoxia-induced vasodilation causes an underestimation of Pmcf [45]. One study observed plateau pressures after 20–30 s and saw a further decrement after 35–40 s which indicates hypoxia-induced vasodilation [16]. Potentially, arm occlusion causes a small accumulation of blood volume because the venous outflow stops before the arterial inflow stops [16]. Though, this potential overestimation is negligible since the inflow is small compared to the total distal arm volume as long as cuff inflation is rapid. To note, Pmcf arm is only reliable when a stable plateau pressure is achieved [2].

In contrast to Pmcf hold, Pmcf arm measures can be made in non-sedated patients with cardiac arrhythmias. However, the possible influence of the rapid cuff inflator on reflex mechanisms needs to be studied. In septic patients, central and peripheral vasomotor tone might be altered differently [46]. Shortly after cardiac surgery differences between aortic and radial pressure can occur [47], still, the original validation studies were on postoperative cardiac surgery patients.

P mcf analogue

Technique description

Based on a Guytonian model of the systemic circulation (CO = VR = (Pmcf − CVP)/RVR), an analogue of Pmcf can be derived using a mathematical model: Pmcf analogue = axCVP + bxMAP + cxCO [5, 20, 48, 49]. In this formula, a and b are dimensionless constants (a + b = 1). Assuming a veno-arterial compliance ratio of 24:1, a = 0.96 and b = 0.04; c resembles arteriovenous resistance and is based on a formula including age, height and weight [5, 4850].

Clinical applicability

The average baseline Pmcf analogue values found in the included studies range from 14 to 18 mmHg (Table 2). One study compared fluid replacement based on target Pmcf analogue compared to conventional treatment in continuous veno-venous hemodiafiltration [49]. Fluid replacement based on target Pmcf analogue led to significantly less fluid administration with stable cardiovascular variables (CVP, MAP, CO) and no complications. So, Pmcf analogue measurement adequately follows intravascular volume status in patients. Pmcf analogue measurements are automatic making it an attractive alternative to Pmcf hold and Pmcf arm.

More recently, the Pmcf analogue dynamics, measured with the Navigator™ device (Applied Physiology, Pty Ltd, Australia), were observed [20, 48, 51]. Patients were defined as responders with an increase in stroke volume or CO > 10% after 250 mL fluid administration. Pmcf analogue increased after fluid administration; however, baseline Pmcf analogue did not differ between responders and non-responders [20, 45, 48] (Table 2). This is contrary to results of another study [43] using Pmcf arm, possibly due to different fluid volume (250 vs. 500 mL) [42]. Although the driving pressure for VR (Pmcf CVP) was different between responders and non-responders, it showed low sensitivity (79%) and specificity (56%) to predict FLR [20, 48].

Precision and accuracy

Precision has not been assessed for Pmcf analogue (Table 4). Comparing measurement techniques revealed a lower Pmcf analogue value compared to Pmcf hold [16]. However, a significant regression of Pmcf analogue and Pmcf hold was observed enabling to adjust the Pmcf analogue value using calibration factor [5].
Table 4

Comparison of bedside Pmcf measurement techniques

 

Pmcf hold

Pmcf arm

Pmcf analogue

CO = (Pmcf CVP)/RVR

Pa = Pv

Pmcf = axCVP + bxMAP + cxCO

Applicability to a broad patient population

±

±

 

Restricted to fully sedated and mechanically ventilated patients

In theory applicable in all patients (sedated or awake) with an radial artery catheter

In theory applicable in all patients (sedated or awake)

 

Restricted to patients without a contraindication for inspiratory holds (such as COPD with bullae)

 

Continuous and accurate CO, MAP and CVP measurements needed

 

Continuous and accurate CO and CVP measurements needed

 

Not suitable in cardiac arrhythmia

 

Not suitable in cardiac arrythmia

  

Accuracy

+

+

 

Values interchangeable with Pmcf arm

Values interchangeable with Pmcf hold

Values significantly lower than derived with Pmcf hold

 

When sedated baroreflex probably of little influence

Dependent on time of measurement: > Pa and Pv equilibration. < hypoxia-induced vasodilatation

Pmcf analogue can be transformed to Pmcf hold values (constant error)

 

Mechanical ventilation may overestimate Pmcf value

Possible influence rapid cuff inflator on reflex mechanism altering Pmcf value in non-sedated patients. This is not studied

Mathematical coupling and the equation is based on assumptions that may not be generalizable to all patient populations in ICU

Precision

?

+

?

 

Not studied

No significant differences during repeated measurements. LSC for a single measurement is 14%

Not studied

Outcome operator independent

±

+

 

Inspiratory holds

Timing of measurement

CVP transducer position and CO measurement technique

 

CVP transducer position and CO measurement technique

  
 

Extrapolation of curve

  

Responding time

+

+

 

> 4 min

30–60 s

Fast, no exact times mentioned

Costs

+

+

 

Theoretically no extra devices needed than standard present in ICU

Rapid Cuff Inflator (Hokanson E20, Bellevue, Washington, USA) = 3000 euro

Navigator™ (Applied Physiology, Pty Ltd, Sydney, Australia)

   

Price unknown

Risk of complications

+

±

 

No complications reported in published studies. In theory:

No complications reported in published studies. In theory:

No complications reported in published studies. In theory:

 

Barotrauma from inspiratory holds

In sedated patients attention should be paid deflating the rapid cuff before hypoxemia-induced damage can occur

Complications associated with central venous catheters and CO measurement

 

Severe hemodynamic instability induced by inspiratory holds

In awake patients local pain could be caused by inflating the rapid cuff inflator

 
 

Complications associated with central venous catheters and CO measurement

  

CO cardiac output, CVP central venous pressure, RVR resistance to venous return, MAP mean arterial pressure, Pa arterial pressure, Pv venous pressure (the latter two measured in the arm)

Limitations

The mathematical model is based on CVP, MAP and CO measurements. As CVP values vary during ventilation, usually end-expiratory CVP-recordings can be used. Furthermore, CVP values depend on the position of the transducer. Accurate CO values are needed for this method. The limitation of Pmcf analogue is that the algorithm is based on a mathematical model with mathematical coupling between CO and Pmcf and fixed Csys and resistance parameters [5], therefore presumably not applicable for all patient populations or clinical conditions. We are unable to assess the availability of the Navigator™ for routine care.

Discussion

We found three bedside techniques to measure Pmcf: Pmcf hold, Pmcf arm and Pmcf analogue. They were used to follow volumetric state and to study drug-induced hemodynamic changes in patients.

The interpretation of VR curves and Pmcf in clinical practice is subject to debate [5259]. The values found in heart-beating ICU patients are higher (14–33 mmHg) than in deceased ICU patients (12.8 ± 5.6 mmHg, mean ± sd), probably because of alteration of vasomotor tone after dying [53]. Furthermore, ICU patients often receive vasopressors which increase Pmcf and the study populations differed making it not one-to-one comparable. It is also speculated that the pressure described by Guyton is not measurable in heart-beating patients and the extrapolated pressure of the curve represents a different physiological parameter. Nevertheless, in two studies Pmcf arm was interchangeable with Pmcf hold [1630]. Furthermore, although Pmcf values may differ, the CVP values do as well, which may account for a similar driving pressure for VR. The reviewed studies illustrate the possible clinical benefits of using the bedside derived Pmcf values.

This review is limited since we were unable to pool the data because of the variety in used conditions and interventions. The 16 included studies were performed by only a few research groups with a limited amount of included patients. In most of the studies, each patient served as their own control since it is not clear what would be an appropriate outside control group.

Still, all studies testing the accuracy of Pmcf to follow intravascular changes and pharmacodynamics found significant results. Therefore, it is unlikely that a larger number of patients will show different outcomes. It is possible only positive studies were published, indicating publication bias. Pmcf values differ between the studies and have a wide range within studies (Table 2). Normal values for different patient populations need to be defined before Pmcf can be implemented into standard (ICU) care. The increase in Pmcf values after fluid administration depends on vascular redistribution, vasomotor tone and fluid loss into the interstitial space. Studies focusing on clinical decision-making based on Pmcf, driving pressure for VR, Vs or Csys have not yet been performed. Study designs need to be created to see if using these measures improves outcomes. Also, no precision studies examining Pmcf hold or Pmcf analogue exist yet.

Conclusions

Presently, three bedside Pmcf measurement techniques are available. All require invasive hemodynamic monitoring. Though Pmcf measures allow for more direct assessment of circulating blood volume, VR and Csys, studies are needed to determine cutoff values to allow Pmcf to trigger therapeutic interventions and to determine its value in clinical practice.

Abbreviations

CO: cardiac output; Csys: vascular compliance, CVP: central venous pressure; FiO2: fractional oxygen concentration; FLR: fluid loading responsiveness; ICU: intensive care unit; MAP: mean arterial pressure; RVR: resistance for venous return.

List of symbols

Pcc: critical closing pressure; Pmcf: mean circulatory filling pressure; Pra: right atrial pressure; VR: venous return; Vs: stressed volume; Vu: unstressed volume.

Declarations

Authors’ contributions

All authors contributed to the manuscript. MW, DV and BG designed the study. MW performed a systematic search of the literature. MW and DS independently screened articles for relevance and subsequently performed data extraction into predefined forms. Quality assessment of the included articles was also independently performed by MW and DS. MW, DV, BG and MP wrote the manuscript. JJ, EO and AV critically reviewed and revised the manuscript. All authors read and approved the final manuscript.

Competing interests

None of the authors have relevant conflict of interest present for any aspect of the submitted work. Denise Veelo performed consultancy work for Edwards Lifesciences, Hemologic and Merck outside the submitted work. Bart Geerts performed consultancy work for Edwards Lifesciences and Philips outside the submitted work. Michael Pinsky is a consultant for Cheetah Medical, Edwards Lifesciences, Exotstat Medical, LiDCO Ltd and Cyberonics outside the submitted work.

Ethics approval and consent to participate

Not applicable.

Funding

None.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors’ Affiliations

(1)
Department of Anesthesiology, Academic Medical Center, Amsterdam, The Netherlands
(2)
Department of Intensive Care, Academic Medical Center, Amsterdam, The Netherlands
(3)
Laboratory of Experimental Intensive Care and Anesthesiology, Academic Medical Center, Amsterdam, The Netherlands
(4)
Department of Critical Care Medicine, University of Pittsburgh Medical Center, Pittsburgh, USA
(5)
Department of Intensive Care Medicine, Leiden University Medical Center, Leiden, The Netherlands

References

  1. Hainsworth R. The importance of vascular capacitance in cardiovascular control. News Physiol Sci. 1990;5:250–4.Google Scholar
  2. Maas JJ. Mean systemic filling pressure: its measurement and meaning. Neth J Crit Care. 2015;19:6–11.View ArticleGoogle Scholar
  3. Peters J, Mack GW, Lister G. The importance of the peripheral circulation in critical illnesses. Intensive Care Med. 2001;27:1446–58.View ArticlePubMedGoogle Scholar
  4. Gelman S. Venous function and central venous pressure: a physiologic story. Anesthesiology. 2008;108:735–48.View ArticlePubMedGoogle Scholar
  5. Parkin WG. Volume state control: a new approach. Crit Care Resusc. 1999;1:311–21.PubMedGoogle Scholar
  6. Jansen JR, Maas JJ, Pinsky MR. Bedside assessment of mean systemic filling pressure. Curr Opin Crit Care. 2010;16:231–6.View ArticlePubMedGoogle Scholar
  7. Bayliss WM, Starling EH. Observations on venous pressures and their relationship to capillary pressures. J Physiol. 1894;16:159–318.View ArticlePubMedPubMed CentralGoogle Scholar
  8. Guyton AC, Polizo D, Armstrong GG. Mean circulatory filling pressure measured immediately after cessation of heart pumping. Am J Physiol. 1954;179:261–7.PubMedGoogle Scholar
  9. Guyton AC, Lindsey AW, Abernathy B, Richardson T. Venous return at various right atrial pressures and the normal venous return curve. Am J Physiol. 1957;189:609–15.PubMedGoogle Scholar
  10. Versprille A, Jansen JR. Mean systemic filling pressure as a characteristic pressure for venous return. Pflugers Arch. 1985;405:226–33.View ArticlePubMedGoogle Scholar
  11. Pinsky MR. Instantaneous venous return curves in an intact canine preparation. J Appl Physiol Respir Environ Exerc Physiol. 1984;56:765–71.PubMedGoogle Scholar
  12. Hiesmayr M, Jansen JR, Versprille A. Effects of endotoxin infusion on mean systemic filling pressure and flow resistance to venous return. Pflugers Arch. 1996;431:741–7.View ArticlePubMedGoogle Scholar
  13. den Hartog EA, Versprille A, Jansen JR. Systemic filling pressure in intact circulation determined on basis of aortic vs. central venous pressure relationships. Am J Physiol. 1994;267:H2255–8.Google Scholar
  14. Yamamoto J, Trippodo NC, Ishise S, Frohlich ED. Total vascular pressure–volume relationship in the conscious rat. Am J Physiol. 1980;238:H823–8.PubMedGoogle Scholar
  15. Samar RE, Coleman TG. Measurement of mean circulatory filling pressure and vascular capacitance in the rat. Am J Physiol. 1978;234:H94–100.PubMedGoogle Scholar
  16. Maas JJ, Pinsky MR, Geerts BF, de Wilde RB, Jansen JR. Estimation of mean systemic filling pressure in postoperative cardiac surgery patients with three methods. Intensive Care Med. 2012;38:1452–60.View ArticlePubMedPubMed CentralGoogle Scholar
  17. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6:e1000100.View ArticlePubMedPubMed CentralGoogle Scholar
  18. Higgins JPT, Green S. Cochrane handbook for systematic reviews of interventions (version 5.1.0). 2011. http://handbook.cochrane.org. Accessed 16 March 2017.
  19. Wells G, Shea B, O’Connell J, Peterson J, Welch V, Losos M, et al. The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2009. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 20 Feb 2017.
  20. Gupta K, Sondergaard S, Parkin G, Leaning M, Aneman A. Applying mean systemic filling pressure to assess the response to fluid boluses in cardiac post-surgical patients. Intensive Care Med. 2015;41:265–72.View ArticlePubMedGoogle Scholar
  21. Maas JJ, Geerts BF, van den Berg PC, Pinsky MR, Jansen JR. Assessment of venous return curve and mean systemic filling pressure in postoperative cardiac surgery patients. Crit Care Med. 2009;37:912–8.View ArticlePubMedGoogle Scholar
  22. Maas JJ, de Wilde RB, Aarts LP, Pinsky MR, Jansen JR. Determination of vascular waterfall phenomenon by bedside measurement of mean systemic filling pressure and critical closing pressure in the intensive care unit. Anesth Analg. 2012;114:803–10.View ArticlePubMedPubMed CentralGoogle Scholar
  23. Keller G, Desebbe O, Benard M, Bouchet JB, Lehot JJ. Bedside assessment of passive leg raising effects on venous return. J Clin Monit Comput. 2011;25:257–63.View ArticlePubMedGoogle Scholar
  24. de Wit F, van Vliet AL, de Wilde RB, Jansen JR, Vuyk J, Aarts LP, et al. The effect of propofol on haemodynamics: cardiac output, venous return, mean systemic filling pressure, and vascular resistances. Br J Anaesth. 2016;116:784–9.View ArticlePubMedGoogle Scholar
  25. Maas JJ, Pinsky MR, de Wilde RB, de Jonge E, Jansen JR. Cardiac output response to norepinephrine in postoperative cardiac surgery patients: interpretation with venous return and cardiac function curves. Crit Care Med. 2013;41:143–50.View ArticlePubMedGoogle Scholar
  26. Helmerhorst HJ, de Wilde RB, Lee DH, Palmen M, Jansen JR, van Westerloo DJ, et al. Hemodynamic effects of short-term hyperoxia after coronary artery bypass grafting. Ann Intensive Care. 2017;7:20.View ArticlePubMedPubMed CentralGoogle Scholar
  27. Persichini R, Silva S, Teboul JL, Jozwiak M, Chemla D, Richard C, et al. Effects of norepinephrine on mean systemic pressure and venous return in human septic shock. Crit Care Med. 2012;40:3146–53.View ArticlePubMedGoogle Scholar
  28. Guerin L, Teboul JL, Persichini R, Dres M, Richard C, Monet X. Effects of passive leg raising and volume expansion on mean systemic pressure and venous return in shock in humans. Crit Care. 2015;19:411.View ArticlePubMedPubMed CentralGoogle Scholar
  29. Maas JJ, Geerts BF, Jansen JR. Evaluation of mean systemic filling pressure from pulse contour cardiac output and central venous pressure. J Clin Monit Comput. 2011;25:193–201.View ArticlePubMedGoogle Scholar
  30. Maas JJ, Pinsky MR, Aarts LP, Jansen JR. Bedside assessment of total systemic vascular compliance, stressed volume, and cardiac function curves in intensive care unit patients. Anesth Analg. 2012;115:880–7.View ArticlePubMedGoogle Scholar
  31. Fessler HE, Brower RG, Wise RA, Permutt S. Effects of positive end-expiratory pressure on the gradient for venous return. Am Rev Respir Dis. 1991;143:19–24.View ArticlePubMedGoogle Scholar
  32. Hedenstierna G. Pulmonary perfusion during anesthesia and mechanical ventilation. Minerva Anestesiol. 2005;71:319–24.PubMedGoogle Scholar
  33. Jellinek H, Krenn H, Oczenski W, Veit F, Schwarz S, Fitzgerald RD. Influence of positive airway pressure on the pressure gradient for venous return in humans. J Appl Physiol. 2000;88:926–32.View ArticlePubMedGoogle Scholar
  34. Berger D, Moller PW, Weber A, Bloch A, Bloechlinger S, Haenggi M, et al. Effect of PEEP, blood volume and inspiratory hold maneuvers on venous return. Am J Physiol Heart Circ Physiol. 2016;311:H794–806.View ArticlePubMedGoogle Scholar
  35. Borst C, Karemaker JM. Time delays in the human baroreceptor reflex. J Auton Nerv Syst. 1983;9:399–409.View ArticlePubMedGoogle Scholar
  36. Peters JK, Lister G, Nadel ER, Mack GW. Venous and arterial reflex responses to positive-pressure breathing and lower body negative pressure. J Appl Physiol. 1997;82(6):1889–96.View ArticlePubMedGoogle Scholar
  37. Sato M, Tanaka M, Umehara S, Nishikawa T. Baroreflex control of heart rate during and after propofol infusion in humans. Br J Anaesth. 2005;94:577–81.View ArticlePubMedGoogle Scholar
  38. Ebert TJ. Sympathetic and hemodynamic effects of moderate and deep sedation with propofol in humans. Anesthesiology. 2005;103:20–4.View ArticlePubMedGoogle Scholar
  39. Lennander O, Henriksson BA, Martner J, Biber B. Effects of fentanyl, nitrous oxide, or both, on baroreceptor reflex regulation in the cat. Br J Anaesth. 1996;77:399–403.View ArticlePubMedGoogle Scholar
  40. Anderson RM. Appendix: clinical determination of mean cardiovascular pressure. In: Anderson RM, editor. The gross physiology of the cardiovascular system. 2nd ed. Tucson: Racquet Press; 1993. p. 61–2.Google Scholar
  41. Aya HD, Rhodes A, Fletcher N, Grounds RM, Cecconi M. Transient stop-flow arm arterial-venous equilibrium pressure measurement: determination of precision of the technique. J Clin Monit Comput. 2016;30:55–61.View ArticlePubMedGoogle Scholar
  42. Aya HD, Rhodes A, Chis Ster I, Fletcher N, Grounds RM, Cecconi M. Hemodynamic effect of different doses of fluids for a fluid challenge: a quasi-randomized controlled study. Crit Care Med. 2017;45:e161–8.View ArticlePubMedGoogle Scholar
  43. Geerts BF, Maas J, de Wilde RB, Aarts LP, Jansen JR. Arm occlusion pressure is a useful predictor of an increase in cardiac output after fluid loading following cardiac surgery. Eur J Anaesthesiol. 2011;28:802–6.View ArticlePubMedGoogle Scholar
  44. Kawalilak CE, Johnston JD, Olszynski WP, Kontulainen SA. Least significant changes and monitoring time intervals for high-resolution pQCT-derived bone outcomes in postmenopausal women. J Musculoskelet Neuronal Interact. 2015;15:190–6.PubMedPubMed CentralGoogle Scholar
  45. Betik AC, Luckham VB, Hughson RL. Flow-mediated dilation in human brachial artery after different circulatory occlusion conditions. Am J Physiol Heart Circ Physiol. 2004;286:H4428.View ArticleGoogle Scholar
  46. Hatib F, Jansen JR, Pinsky MR. Peripheral vascular decoupling in porcine endotoxic shock. J Appl Physiol. 2011;111:853–60.View ArticlePubMedPubMed CentralGoogle Scholar
  47. Stern DH, Gerson JI, Allen FB, Parker FB. Can we trust the direct radial artery pressure immediately following cardiopulmonary bypass? Anesthesiology. 1985;62:557–61.View ArticlePubMedGoogle Scholar
  48. Cecconi M, Aya HD, Geisen M, Ebm C, Fletcher N, Grounds RM, et al. Changes in the mean systemic filling pressure during a fluid challenge in postsurgical intensive care patients. Intensive Care Med. 2013;39:1299–305.View ArticlePubMedGoogle Scholar
  49. Parkin G, Wright C, Bellomo R, Boyce N. Use of a mean systemic filling pressure analogue during the closed-loop control of fluid replacement in continuous hemodiafiltration. J Crit Care. 1994;9:124–33.View ArticlePubMedGoogle Scholar
  50. Crozier TM, Wallace EM, Parkin WG. Haemodynamic assessment in pregnancy and pre-eclampsia: a Guytonian approach. Pregnancy Hypertens. 2015;5:177–81.View ArticlePubMedGoogle Scholar
  51. Aya HD, Ster IC, Fletcher N, Grounds RM, Rhodes A, Cecconi M. Pharmacodynamic analysis of a fluid challenge. Crit Care Med. 2016;44:880–91.View ArticlePubMedGoogle Scholar
  52. Brengelmann GL. A critical analysis of the view that right atrial pressure determines venous return. J Appl Physiol. 2003;94:849–59.View ArticlePubMedGoogle Scholar
  53. Repessé X, Charron C, Fink J, Beauchet A, Deleu F, Slama M, et al. Value and determinants of the mean systemic filling pressure in critically ill patients’. Am J Physiol Heart Circ Physiol. 2015;309:H1003–7.View ArticlePubMedGoogle Scholar
  54. Brengelmann GL. Letter to the editor: comments on “Value and determinants of the mean systemic filling pressure in critically ill patients”. Am J Physiol Heart Circ Physiol. 2015;309:H1370–1.View ArticlePubMedGoogle Scholar
  55. Repessé X, Vieillard-Baron A. Reply to Letter to the editor: comments on ‘Value and determinants of the mean systemic filling pressure in critically ill patients’. Am J Physiol Heart Circ Physiol. 2015;309:H1372–3.View ArticlePubMedGoogle Scholar
  56. Beard DA, Feigl EO. Understanding Guyton’s venous return curves. Am J Physiol Heart Circ Physiol. 2011;301:H629–33.View ArticlePubMedPubMed CentralGoogle Scholar
  57. Teboul JL. Mean systemic filling pressure: we can now estimate it, but for what? Intensive Care Med. 2013;39:1487–8.View ArticlePubMedGoogle Scholar
  58. Parkin G. Re: mean systemic filling pressure: we can now estimate it but for what? Intensive Care Med. 2014;40:139.View ArticlePubMedGoogle Scholar
  59. Teboul JL. Mean systemic filling pressure: we can now estimate it, but for what? Response to comment by Parkin. Intensive Care Med. 2014;40:140.View ArticlePubMedGoogle Scholar

Copyright

© The Author(s) 2018

Advertisement