Bedside Assessment of Lung Recruitability: Making Sense of the Recruitment-to-Inflation Ratio

Abstract

Determination of optimum PEEP levels remains an elusive goal. One factor is the recruitability of the lung, yet this is another difficult determination. Recently, a simple bedside technique, called the recruitment-to-inflation ratio, has been described and validated by comparison to the dual pressure-volume curve method. We describe the prior research and concepts of lung mechanics leading up to this metric and develop some background mathematics that help clinicians understand its meaning.

Introduction

A central concept in mechanical ventilation of the patient with ARDS is to prevent the constant opening and closing of alveoli that may be injurious to the lung (atelectrauma).¹ To that end, recruitment maneuvers to open alveolar units, followed by PEEP titration to keep them open throughout the respiratory cycle, have been a theoretically attractive ventilation strategy. Nonetheless, several randomized controlled trials that aimed to optimize recruitment in the intervention arm showed similar clinical outcomes to controls.^2,3 A signal for potential harm attributed to recruitment maneuvers was reported in a clinical trial.⁴

Defining Recruitability

The best way to define recruitability also remains controversial, and different views have been published.^3,5,6 Using computed tomography (CT), recruitability is measured as the amount of not-inflated tissue at a given pressure that reinflates at higher pressure.⁷ Alternative (ie, more practical) methods based on respiratory system compliance (C_RS) have been proposed. In older studies,^8,9 a prolonged exhalation (V_PE) to atmospheric pressure was used to estimate the lung volume gained by PEEP (Fig. 1). Raineri et al⁸ state that the gas volume of collapsed lung units recruited with PEEP can be calculated as the difference in lung volume between PEEP = 0 and PEEP > 0 for the same degree of static pressure of the respiratory system (eg, 20 cm H₂O) from the pressure-volume (P-V) curves obtained at different PEEP levels, assuming that the relaxation volume of the respiratory system does not change with PEEP.

Fig. 1.

By definition, compliance, C, is the ratio of volume change, ΔV, to the associated elastic pressure change, ΔP (eg, during slow-flow inflation). Volume change during inflation may be due to distention of previously open lung units or to recruitment of previously closed units. If partially collapsed lung units are recruited, they will yield a higher compliance (A) than when fully inflated (B). Hence, after expiration to 0 PEEP, a long linear segment of the pressure-volume curve with a relatively high compliance (A) can be explained by continuous recruitment. At about 30 cm H₂O, the 2 curves recorded from 0 PEEP (A) and from PEEP > 0 (B) tend to merge.⁹

According to Jonson et al,⁹ the change in effective compliance within tidal ventilation will depend on opposite factors: (1) the ongoing recruitment occurring during one inflation maneuver will tend to increase compliance; and (2) the effect of the established, time-dependent, recruitment obtained with PEEP will tend to decrease its value. Therefore, they conclude that measurement of effective compliance over the full tidal volume (V_T) is not a useful tool in setting PEEP.⁹

Put another way, Amato and Santiago⁷ explain that when 2 P-V curves are superposed on the same graph, with each curve starting from a different PEEP or end-expiratory lung volume (EELV) point if no recruitment occurred, the 2 curves should superpose. In contrast, if the curves separate, with more gas in the curve starting at higher PEEP, this is considered recruitment (essentially equivalent to the hysteresis observed between inspiratory and expiratory P-V loops).⁷

Although not described this way in the studies by Ranieri et al⁸ or Jonson et al,⁹ Amato and Santiago conclude that the whole volume collected during V_PE, minus the exhaled V_T, should be equal to inflation volume caused by PEEP. But if any residual difference is observed, it should be considered as recruited volume (V_rec).⁷

The Problem of Airway Closure

Note that use of P-V curves is complicated by the phenomenon of airway closure. Airway closure interrupts communication between the proximal airway opening and the distal alveolar and/or small airway structures. Therefore, lung inflation will only occur when closed airways are reopened. The pressure at the point of opening is called airway opening pressure (AOP). A method to measure AOP is given below. Chen et al¹⁰ report that in approximately one quarter to one third of ventilated subjects they studied with moderate to severe ARDS the airways were totally closed until a sufficiently high AOP. Presence of airway closure could make measurement of respiratory system mechanics inaccurate.¹¹

Summary Equations

We formalize the above ideas in equations as follows: Starting with the V_PE, the exhaled V_T is subtracted along with the predicted inflation volume caused by a PEEP change (V_ΔPEEP, calculated as ΔPEEP multiplied by C_RS). Any residual difference should be considered as V_rec:

$$V_{PE}=V_T+V_{\Delta PEEP}+V_{rec}$$ (1)

So

$$V_{rec}=V_{PE}-V_T-V_{\Delta PEEP}=V_{PE}-V_T-\left(\Delta PEEP\times C_{RS}\right)$$ (2)

where ΔPEEP is the change in PEEP between 2 steady-state conditions (eg, from 15 cm H₂O to 5 cm H₂O) and C_RS is the baseline compliance of the respiratory system. Note that V_rec is considered to be the change in EELV (ΔEELV) associated with ΔPEEP.

However, according to Amato and Santiago,⁷ this method underestimates V_rec because V_PE is not enough to fully empty newly recruited units (ie, exhaled volume [V_E] may be underestimated because of some gas trapping behind unstable airways).¹² More recently, ΔEELV caused by ΔPEEP has been directly measured, either by helium dilution methods, by nitrogen washout techniques, or by electrical impedance tomography. After measuring ΔEELV between 2 PEEP levels, we subtract V_ΔPEEP, and the residual difference is then a better estimate of V_rec.⁷

Bedside measurement of lung recruitability to determine the severity of ARDS may be useful to individualize PEEP and protective lung strategies, to determine the need for a lung-recruitment maneuver, or even to assess the risk of impaired hemodynamic status.⁷ Availability of a robust, simple, and reproducible method at the bedside to determine lung recruitability (which is independent of ventilator brand) would be highly desirable to standardize the process and advance research in the field.

Understanding the Recruitment-to-Inflation Ratio

Calculating Recruitable Volume

Recently, Chen et al⁵ proposed a simple bedside technique (validated by comparison to the dual P-V curve method) (see Figs. 2 and 3).⁶ Chen et al⁶ write that the rationale is to detect a difference in lung volume (at a given elastic pressure) between 2 P-V curves, starting from 2 PEEP levels. If the higher PEEP results in recruitment, the difference between the 2 volumes at a given pressure will indicate V_PE.⁶ This is illustrated in Figure 2 (adapted from Fig. 1 from their paper).

Fig. 2.

Measurement of the recruited volume (ΔV_rec) and compliance of the recruited lung using the reference method (multiple pressure-volume [P-V] curves) in a representative patient with complete airway closure. Elastic P-V curves were obtained by low-flow (5 L/min) inflation. Airway opening pressure (AOP) was defined as the elastic airway pressure at which gas volume delivered to a patient became > 4 mL the volume compressed in an occluded circuit. The presence of AOP suggests complete airway closure because the initial part of the patient's P-V curve (red line) completely overlapped with the blocked circuit's P-V curve. ΔV_rec is the volume difference between 2 P-V curves of the patient, but ΔP_rec was the difference between higher PEEP and the AOP. Compliance of respiratory system above AOP was used as a surrogate for the compliance of the baby lung. C_rec = compliance of the recruited lung; P-V = pressure-volume; AOP = airway opening pressure.

Fig. 3.

Estimation of change in end-expiratory lung volume after (step 1) reducing the breathing frequency to avoid auto-PEEP; (step 2) sudden reduction of PEEP from 15 cm H₂O to 5 cm H₂O; (step 3) recording of exhaled volume (V_E) as a function of time, V_E(t) before and after the PEEP change (t₁ and t₂, respectively); and (step 4) recording plateau pressure at the low PEEP level. V_E = exhaled volume; t₁ = time when V_E is recorded for the breath before the PEEP reduction; t₂ = time when V_E is recorded after the breath with the PEEP reduction; P_plat = plateau pressure.

The bedside technique is a comparison of the measured ΔEELV during a PEEP reduction maneuver and the predicted ΔEELV in the absence of any recruitment. The difference is the V_PE, symbolized by ΔV_rec and calculated as:⁶

$$\Delta V_{rec}=measured\Delta EELV-predicted\Delta EELV$$ (3)

Figure 3 illustrates the procedure for gathering data to calculate ΔV_rec and is adapted from the supplemental material in the paper by Chen et al.⁶ But Equation 3 was not derived mathematically in that paper. Chen et al⁶ simply state that ΔEELV is measured with a flow meter during a PEEP reduction maneuver, recording the expired V_T (Figure E2 in their online supplement). The predicted ΔEELV is calculated under the assumption with constant compliance, without any change in aerated lung units. Thus, the difference between the measured ΔEELV and the predicted ΔEELV is the recruitment caused by the higher PEEP.⁶

Unfortunately, Chen et al⁶ do not explain how the measurements of ΔV_E (Fig. 3) are used to calculate the measured ΔEELV. Furthermore, Equation 3 does not agree with Equation 2 because it leaves out V_T. This can be confusing. Therefore, we derive an equation for measured ΔEELV as follows, where V_E = exhaled volume; t₁ = time when V_E is recorded for the breath before the PEEP reduction; t₂ = time when V_E is recorded after the breath with the PEEP reduction; EELV = end-expiratory lung volume; and functional residual capacity (FRC) = EELV at PEEP = 0. We begin with the definition of EELV as FRC plus V_T; then it follows that the measured ΔEELV is due to the change in V_E between 2 breaths because FRC remains constant. Note that we use the general term V_E in the equations below instead of V_T because the V_E after the PEEP reduction is not a V_T:

$$\Delta EELV=FRC+\Delta V_E$$ (4)

The ΔEELV is measured as the difference between EELV at the higher PEEP minus the EELV at the lower PEEP as shown in Figure 3

$$measured\Delta EELV=EELV\left(t_2\right)-EELV\left(t_1\right)$$ (5)

Substituting definition of EELV into Equation 5, we get:

$$measured\Delta EELV=\left[FRC+V_E\left(t_2\right)\right]-\left[FRC+V_E\left(t_1\right)\right]$$ (6)

$$=V_E\left(t_2\right)-V_E\left(t_1\right)=\Delta V_E$$ (7)

This result indicates that the ΔEELV is simply the measured change in exhaled V_T.

Example

$$V_E\left(t_2\right)=600mL$$

$$V_E\left(t_1\right)=400mL$$

$$measured\Delta EELV=V_E\left(t_2\right)-V_E\left(t_1\right)=600-400=200mL$$

This result is consistent with the description given by Chen et al⁵ in an earlier paper where they say that the difference in expired V_T values between the expired V_T after decreasing PEEP (ie, at t₂) and the breath before changing PEEP (ie, at t₁) is referred to as the total change in lung volume from high to low PEEP.

Note that the volume exhaled at t₁, V_E(t₁), is the set V_T at PEEP_high (assuming no leaks). The volume exhaled at t₂, V_E (t₂), is the set V_T plus the volume due to the PEEP reduction (ie, the predicted ΔEELV) plus any V_rec. Hence, the difference between these 2 volumes, the measured ΔEELV, is composed of the predicted ΔEELV and V_rec. It follows that to calculate V_rec we must now subtract the predicted ΔEELV from the measured ΔEELV, and the result is Equation 3.

In patients without airway closure, the predicted ΔEELV is simply a product of C_RS and ΔPEEP, where C_RS is calculated at PEEP_low and ΔPEEP = PEEP_high – PEEP_low (assuming no auto-PEEP because of the low breathing frequency):

$$predictedEELV=C_{RS}\times \Delta PEEP$$ (8)

$$=\left(\frac{set{V}_T}{P_{platatPEE{P}_{low}}-PEE{P}_{low}}\right)\times \left(PEE{P}_{high}-PEE{P}_{low}\right)$$

Example

$$set{V}_T=400mL$$

$$PEE{P}_{high}=15cm{H}_{2}O$$

$$PEE{P}_{low}=5cm{H}_{2}O$$

$$P_{platatPEE{P}_{low}}=25cm{H}_{2}O$$

$$C_{RS}=\left(\frac{set{V}_T}{P_{platatPEE{P}_{low}}-PEE{P}_{low}}\right)=\left(\frac{400}{25-5}=20mL/cm{H}_{2}O\right)$$

$$predicted\Delta EELV=C_{RS}\times \Delta PEEP=20\times 10=200mL$$

In patients with airway closure, C_RS is measured not at PEEP_low but rather above the point of AOP:

$$predicted\Delta EELV=C_{RS}\times \Delta P$$ (9)

$$=\left(\frac{set{V}_T}{P_{platatPEE{P}_{low}}-AOP}\right)\times \left(PEE{P}_{high}-AOP\right)$$

Example

$$set{V}_T=400mL$$

$$PEE{P}_{high}=15cm{H}_{2}O$$

$$AOP=8cm{H}_{2}O$$

$$P_{platatPEE{P}_{low}}=25cm{H}_{2}O$$

$$C_{RS}=\left(\frac{set{V}_T}{P_{platatPEE{P}_{low}}-AOP}\right)=\left(\frac{400}{25-8}=23.5mL/cm{H}_{2}O\right)$$

$$predicted\Delta EELV=C_{RS}\times \Delta PEEP=23.5\times 7=164.5mL$$

Piraino¹³ has described the above procedure differently, but again no equations were given for calculating V_rec. However, he has provided a web-based calculator to facilitate calculation of V_rec from bedside measurements (https://crec.coemv.ca. Accessed January, 16, 2023). To understand this online calculator, recall that the total volume released during the ΔPEEP maneuver (measured ΔEELV) has 3 components: the exhaled V_T at PEEP_high plus the volume released by decreasing PEEP from high to low (predicted ΔEELV, accounting for any AOP) plus any unknown volume recruited at PEEP_high (V_rec). This is shown in Figure 4. The equations for predicted ΔEELV are the same as above. However, the equation for V_rec is now like Equation 2:

$$\Delta V_{rec}=released\Delta V-V_T-predicted\Delta EELV$$ (10)

Fig. 4.

Estimation of change in recruited volume after (step 1) reducing the breathing frequency to avoid auto-PEEP, (step 2) sudden reduction of PEEP from 15 cm H₂O to 5 cm H₂O, (step 3) recording of released volume after the PEEP change, and (step 4) recording plateau pressure at the low PEEP level. The predicted change in end-expiratory lung volume is calculated using Equation 8 or Equation 9. P_plat = plateau pressure; V_T = tidal volume; ΔV = released volume; V_rec = recruited volume.

Determining Airway Opening Pressure

AOP in Equation 9 above is determined from graphical analysis of a slow-flow (quasi-static) P-V curve as shown in Figure 5.¹⁴

Fig. 5.

Determination of airway opening pressure (AOP). Low-flow inflation pressure-volume (P-V) curves without (A, green solid line) and with (B, blue solid line) complete airway closure. Red dashed line represents P-V curve of patient circuit (C = 2.4 mL/cm H₂O). Blue dash-dotted line marks AOP. In (A), P-V curve of the patient and circuit separate immediately. In (B), the initial part of the P-V curve is very flat, and superimposed on the curve of the circuit, suggesting closed airways. Above AOP (15 cm H₂O), the curves separate. The slope of the patient P-V curve increases, suggesting lungs are open. AOP = airway opening pressure.

Rationale for Basing V_rec on ΔEELV

Chen et al⁶ state that the rationale for the single-breath experimental method for determining V_rec (based on calculations of ΔEELV) “…has a similar rationale as the multiple P-V curve method….” However, no explanation for this is given.

Looking at Equation 8, we see that the predicted ΔEELV is simply ΔPEEP multiplied by baseline compliance. Furthermore, the measured ΔEELV is also a function of ΔPEEP. The change in volume associated with ΔPEEP is assumed to be due to 2 main factors: a change in compliance and/or a change in the open lung units (ie, recruitment). Indeed, the change in open lung units is apparently more important because the compliance seems to drop with the increase in PEEP. This is seen in Figure 2 and is reflected in a comment by Jonson et al⁹ “…the effect of the established, time-dependent, recruitment obtained with PEEP will tend to decrease its value, as shown in this study.” Hence, we may conclude that the difference between measured and predicted ΔEELV is essentially the difference in volume between 2 P-V curves or ΔV_rec (Fig. 2), and this is confirmed by the results of the study by Chen et al.⁶

Calculating the Recruitment-to-Inflation Ratio

If we can easily estimate the recruited lung volume from a ΔPEEP maneuver, then what more do we need? The paper by Chen et al⁶ does not address this issue. Their study showed a range of V_rec values from 0–400 mL. Practically, we need to index V_rec somehow because its value will depend not only on the degree of lung atelectasis but also total lung size (related to ideal body weight). This is why specific compliance (compliance normalized for FRC) is used in pediatrics to account for the wide range of body sizes. But measurement of FRC is generally not practical at the bedside. Instead, Chen et al⁶ have proposed the idea of the recruitment-to-inflation ratio(R/I ratio), which is the ratio of the compliance of recruited lung (after change to PEEP_high) to the baseline tidal compliance at PEEP_low. Conceptually, this is similar to normalizing compliance with FRC because compliance at PEEP_low is affected by body size as well as lung condition. Indeed, Chen et al⁶ actually say “the R/I ratio is calculated by normalizing the C_rec to the C_rs at PEEP_low or above AOP.” It follows that the lower the ratio, the lower the potential for recruiting lung tissue. Let's see how this might be justified:

As noted above and in the paper by Jonson et al,⁹ increasing PEEP may recruit lung units, but it also may decrease compliance, and hence setting PEEP by tidal compliance is unreliable.^6,9 Indeed, examination of Figure 2 shows that compliance at PEEP_high is < compliance at PEEP_low (by visual inspection of the P-V slope at PEEP_high vs PEEP_low). Therefore, if we were to test the hypothesis that the lungs of this patient were recruitable based on an improved tidal compliance at PEEP_high, we would conclude that they were not, whereas the R/I ratio (1.67) indicated they were.

The novelty of the R/I ratio technique is that it is based on what is conceptualized as the compliance of the recruited lung (C_rec), not compliance associated with the V_T at either high or low PEEP (because that could be unreliable as discussed above). This recruitable part of the ARDS lung is in contrast to an assumed non-recruitable part and a “baby lung part”¹⁵ (ie, that part that remains aerated at PEEP_low or FRC⁶).

Note that C_rec is a theoretical construct rather than a mechanical characteristic of the respiratory system. There is no way to evaluate the actual C_rec separate from the other lung tissue. C_rec is just a ratio of a volume change to a pressure change and thus happens to have the same units as compliance. Chen et al¹⁴ conceive C_rec as the volume that can be regained by each cm H₂O of ΔPEEP.

After calculating ΔV_rec from Equation 3, calculation for C_rec is simple. For patients without airway closure, the equation is:

$$C_{rec}=\frac{\Delta V_{rec}}{\Delta P_{rec}}=\frac{\Delta V_{rec}}{PEE{P}_{high}-PEE{P}_{low}}$$ (12)

Example

$$V_{rec}=200mL$$

$$PEE{P}_{high}=15cm{H}_{2}O$$

$$PEE{P}_{low}=5cm{H}_{2}O$$

$$C_{rec}=\left(\frac{V_{rec}}{PEE{P}_{high}-PEE{P}_{low}}\right)=\left(\frac{200}{15-5}=20mL/cm{H}_{2}O\right)$$

For patients with AOP > PEEP_low, the equation is:

$$C_{rec}=\frac{\Delta V_{rec}}{\Delta P_{rec}}=\frac{\Delta V_{rec}}{PEE{P}_{high}-AOP}$$ (13)

Example

$$V_{rec}=200mL$$

$$PEE{P}_{high}=15cm{H}_{2}O$$

$$AOP=8cm{H}_{2}O$$

$$C_{rec}=\left(\frac{V_{rec}}{PEE{P}_{high}-AOP}\right)=\left(\frac{200}{15-8}=28.6mL/cm{H}_{2}O\right)$$

But if a hypothesis test for recruitability based on a change in tidal compliance (between PEEP_high and PEEP_low) is assumed to be unreliable, then with what do we compare C_rec? Chen et al⁶ say only that “By comparing C_rec with the compliance of the baby lung, one might predict the likelihood of the distribution of volume between the recruited lung (recruitment) and the baby lung (inflation/hyperinflation).” How is this any better than comparing compliance at PEEP_high versus PEEP_low?

Tidal compliance at PEEP_low is a reasonable baseline if we assume it reflects the relatively normal portion of the ARDS lung.¹⁶ However, tidal compliance at PEEP_high might reflect overdistention of the baby lung as well as compliance change due to any recruitment. The assumption is that C_rec is a better reflection of the compliance of recruited lung, although, as Chen et al⁶ point out, “Increasing PEEP in high recruiters does not guarantee the absence of hyperinflation.”

Logically, if C_rec is a better index than tidal compliance at PEEP_high, then a simple comparison of C_rec to tidal compliance at PEEP_low should yield a reasonable normalization of C_rec (as mentioned above) with more reliable results than comparing C_RS at PEEP_high versus PEEP_low.

Therefore, for patients without airway closure, the R/I ratio is defined as

$$R/I=\frac{C_{rec}}{C_{RS}atPEE{P}_{low}}$$ (14)

For patients with complete airway closure, the R/I ratio is defined as

$$R/I=\frac{C_{rec}}{C_{RS}aboveAOP}$$ (15)

At present, no ventilator can automatically calculate the R/I ratio. Therefore, we have created a Microsoft Excel–based calculator to facilitate bedside calculations (https://1drv.ms/x/s!AuFakBJODC3DhOlLq39LseLtOgQkUw?e=LfjCpg) along with forms that can be printed if calculations need to be performed manually.

All of the important equations from above are summarized in Table 1.

Table 1.

Summary of Important Equations

Interpreting the Recruitment-to-Inflation Ratio

Chen et al⁶ clearly state that C_rec and R/I ratio are new concepts and no predictive thresholds exist yet. They arbitrarily chose to use the median value of the R/I ratio (0.5) to define lung recruitability (range 0–2.0). Therefore, “high recruiters” were defined by an R/I ratio ≥ 0.5 and “low recruiters” were defined by an R/I ratio < 0.5.⁶ This logical but arbitrary threshold would need to be validated and fine-tuned for clinical use in prospective studies.

Discussion

The jury is still out on whether we should recruit lung that appears recruitable. Recently, a Bayesian network meta-analysis by Dianti and colleagues¹⁷ concluded that application of higher PEEP was superior to lower PEEP in improving 28-d mortality in ARDS. Notwithstanding, the 18 trials that contributed to the analysis varied widely in the adopted high and low PEEP ranges and use and method of recruitment maneuvers. This observation lays bare the need to standardize the definition of recruitability and the methods to detect and quantitate its presence.¹⁸ Therefore, a method that is relatively simple to implement at the bedside that directly estimates recruitment independent of ventilator capabilities would be more than welcome. The R/I ratio promises to fit the bill.

Before incorporation in clinical practice, R/I ratio reproducibility and accuracy require further study. For instance, in an established lung model, Cour and colleagues¹⁹ demonstrated clinically important variability in R/I ratio assessment between 5 different ICU ventilators. Clinically useful cutoff values for R/I should be validated in mechanistic studies that employ accepted standards for recruitment, such as CT scan or electrical impedance tomography.

Although prospective data are lacking in its usefulness, tidal pressure, P_T, (also known as driving pressure) is a simpler bedside measurement that reflects C_RS at a given EELV and V_T. Lower P_T has been linked with favorable clinical outcomes in retrospective studies.²⁰ Nonetheless, P_T, as aforestated, may not be an accurate representation of recruitment since changes in this measurement cannot distinguish between improvement in compliance via inflation of already recruited lung units versus recruitment. P_T contains less information than the R/I ratio and, therefore, may not have the same level of precision or utility. Future clinical studies should compare ventilatory strategies employing P_T measurement versus R/I ratio in determining optimal PEEP. These studies should include control groups that reflect standard of care, that is, low V_T ventilation.

In conclusion, R/I ratio is theoretically sound and relatively accurate for obtaining bedside measurement of recruitability that awaits validation in larger studies.