Assessment of chest wall elastance in patients with pulmonary fibrosis waiting for lung transplantation, interest in donor size selection

Abstract

Some patients with pulmonary fibrosis (PF) can have severe and fixed chest wall retraction; others regain the shape of their original rib cage once the lungs are removed. These 2 possibilities determine the size of the lung graft to be allocated but are not predictable with classical respiratory tests or computed tomography (CT) scan. We first measured chest wall elastance (E_cw) with esophageal pressure on the day of transplantation (group 1) and then during pretransplant medical check-up, and used for donor selection (group 2). Twenty patients in group 1 had low pretransplantation actual total lung capacity/predicted total lung capacity (pTLC) ratio that was not correlated with E_cw. The amount of transplanted lung TLC_transplanted/pTLC was correlated to E_cw (R² = 0.43, p = 0.003). Patient with higher E_cw required lung resection and had more primary graft dysfunction. In group 2, 20 patients' E_cw measurements allowed for increase in TLC_transplanted/pTLC from 79 ± 20% to 93 ± 18%, p = 0.023 with only 2 lung resections. E_cw can be measured before transplantation to optimize size mismatch and lung resection.

Background

Some patients with pulmonary fibrosis (PF) can have severe and fixed chest wall retraction; others regain the shape of their original rib cage once the lungs are removed. A retracted chest cavity shrinks with significant contraction of the intercostal muscles, causing rib crowding over time. This contraction follows retraction due to lung fibrosis and can be reversible or not. The increase in the respiratory system elastance has been attributed to the lung component. However, some specific chest wall modifications may occur in some PF patients, leading to nonreversible chest retraction.¹ In that case, the stiffen chest wall should be associated with a high chest wall elastance (E_cw). Chest retraction will determine the size of the lung graft to be allocated but is not predictable with classical respiratory tests or computed tomography (CT) scan showing only low volume.²

Thus, the aim of this study was to measure, in a first group, the stiffness of the chest wall with E_cw and to predict the transplanted lung volume with or without resection, in comparison to pulmonary function tests and CT scan findings. In a second group, we assessed E_cw measurement in the medical pretransplant check-up to assess safety and effect on total lung capacity (TLC) donor selection.

Methods

Patients suffering from PF who underwent bipulmonary transplantation were studied. The study was approved by the local Ethics Committee (CER-BDX 2021-51), and all patients gave their informed consent.

In group 1, measurement of respiratory mechanics with E_cw started on the day of transplantation, under anesthesia just before incision. Surgeons selected lung donor size according to recipient CT scans and TLC but were not aware of E_cw during surgery.

In group 2, measurement of E_cw was done with pretransplant medical check-up, during gastric fibroscopy under general anesthesia. In that group, surgeons were aware of patients' E_cw on the waiting list. If E_cw was normal, pTLC ratio between the donor and recipient was close to 1, or even >1 (oversized). If E_cw > 10 cmH₂O/liter, donor aTLC was more likely closer to recipient aTLC.

Methods for E_cw measurement with esophageal balloon and CT scan scoring are resumed in the Online Supplements.

Statistical analysis

Data are expressed as mean ± standard deviation and median [interquartile range, IQR] for normally and non-normally distributed variables. Comparison pre- vs post-transplantation in group 1 or comparison between the 2 groups was carried out using the paired t-test or Wilcoxon. Correlations were performed using Pearson or Spearman test, a p value <0.05 was considered significant. Prism 6 (GraphPad Software, La Jolla, CA).

Results

Twenty consecutive patients in each group were studied. Patient’s characteristics regarding recipients and donors of the 2 groups are summarized in Table 1. Before transplantation, median aTLC and aTLC/pTLC ratio were low in both groups but actual forced vital capacity was significantly lower in group 2, p = 0.035.

Table 1.

Comparison Between Group 1 and Group 2

	Group 1 (N = 20)	Group 2 (N = 20)	p
Age (year)	58.6±6.9	52.5±15.3	0.109
Weight (kg)	78.5±12.6	66.5±13.4	0.006
Height (cm)	172.3±7.5	167.9±7.1	0.073
BMI	26.5±4.4	23.3±3.5	0.015
aTLC (L)	3.3 [3.1-4.0]	2.9 [2.2-3.6]	0.027
pTLC (L)	6.6 [6.3-7.0]	6.5 [5.7-6.7]	0.113
aTLC/pTLC (%)	52,5 [47.2-37.0]	44.0 [37.0-61.0]	0.131
aFVC (ml)	2.2 [1.7-3.0]	1.6 [1.3-2.0]	0.035
aFVC/pFCV (%)	58.5 [42.5-66.5]	43.0 [34.0-56.0]	0.049
VT (ml)	397.4±50.1	384.6±50.8	0.213
VT (ml.kg−1 PBW)	5.3 [4.6-5.6]	5.8 [5.5-6.7]	0.004
Pplat (cmH2O)	21.0 [19.0-26.0]	22.0 [20.0-25.0]	0.653
PEEP (cmH2O)	5.0 [4.0-7.0]	5.5 [5.0-8.2]	0.220
DP (cmH2O)	16.0 [14.0-21.0]	16.0 [13.5-18.5]	0.780
Crs (ml.cmH2O−1)	24.2±6.4	22.5±6.5	0.451
EL (ml.cmH2O−1)	25.0 [19.0-48.0]	39.5 [32.5-47.5]	0.093
Ecw (ml.cmH2O−1)	10.7 [7.7-15.5]	6.8 [5.0-8.6]	0.001
Ecw/Ers	0.29 [0.17-0.42]	0.16 [0.12-0.19]	0.009
pTLC donor (liter)	5.6±1.1	5.7±1.1	0.686
TLC transplanted (liter)	5.1±1.1	5.6±1.1	0.127
TLC transplanted/pTLC (%)	79±20	93±18	0.023
Lung resection n (%)	N = 6 (30%)	N = 2 (10.5%)	0.235
PaO2/FiO2 donor (mm Hg)	399±68	420 ±89	0.424
Ischemic time (min)	417 [372-441]	420 [360-540]	0.210
ICU length of stay (days)	20.5 [12.0-47.8]	15.0 [11.0-25.0]	0.190
PGD 3 n (%)	N = 3 (15.0%)	N = 3 (15.8%)	0.999
Mortality at 3 months n (%)	N = 2 (10%)	N = 0 (0%)	0.487

Abbreviations: aFVC, actual forced vital capacity; aTLC, actual total lung capacity; C_RS: respiratory system compliance; DP, driving pressure; E_CW, chest wall elastance; E_L, lung elastance; E_RS, respiratory system elastance; PEEP, positive end expiratory pressure; pFVC, predicted forced vital capacity; PGD, primary graft dysfunction; P_plat, end inspiratory plateau pressure; pTLC, predicted total lung capacity.

Data are expressed as median [Q1-Q3] if the distribution was not normal with a Mann-Whitney test, and mean ± standard deviation if distribution was normal with a Student's t-test.

For quantitative variables with a Fisher exact test was used.

In group 1, 6 PF patients (30%) had an E_cw above 15 cmH₂O/liter. Before transplantation, E_cw was not correlated with aTLC and aTLC/pTLC ratio (R² = 0.05, p = 0.378 and R² = 0.06, p = 0.311; respectively). After transplantation, mean TLC_transplanted/pTLC was 79.0 ± 20.0% and was significantly correlated with E_cw (R² = 0.47, p = 0.0009), with TLC_transplanted = pTLC × (−3.13 × E_cw + 114)/100 (TLC in liter and E_cw in cmH₂O/liter, Figure 1).

Figure 1.

Chest wall elastance correlations with the pre-operative ratio aTLC/pTLC and the ratio of transplanted lung volume/predicted Total Lung Capacity. The data are from patients of group 1: E_cw was measured the day of lung transplantation after donor selection. Panel A: This figure illustrates the lack of correlation between the wide range of chest wall elastance (ECW) in PF patients and their significantly reduced aTLC/pTLC ratio. Panel B: This figure illustrates the significant correlation between chest wall elastance (ECW) and the amount of lung volume transplanted represented by the ratio of transplanted donor TLC with recipient predicted TLC in %. In blue patients with full lung donor transplanted, in red patients with lung resection. aTLC, actual total lung capacity; pTLC, predicted total lung capacity; TLC, total lung capacity.

The decrease of functional vital capacity between the day of diagnosis and the last pulmonary functional test was not correlated to E_CW (R² = 0.22, p = 0.372). Mean time between first respiratory symptoms and lung transplantation was 1,709 [IQR: 925-2,104] days and was not correlated with E_cw (R² = 0.22; p = 0.383).

Mean total fibrosis score on CT scan was 261 ± 154 with a mean aorta-sternum distance of 21 ± 9 mm and a mean lung height of 16 ± 2 cm, it was not correlated with E_cw.

In group 1, patients with lung resection had a similar median pretransplantation aTLC/pTLC ratio but had significantly higher median E_cw (18.2 [IQR: 10.7-18.9] vs 10.2 [IQR: 7.3-12.7] cmH₂O/liter, respectively; p = 0.0002), Table 1.

Eight patients required a clamshell thoracotomy and they had an E_cw of 14.0 [11.2-17.4] cmH₂O/liter, the other 12 patients had an antero-lateral bithoracotomy with an E_cw of 9.3 [7.0-12], p = 0.07.

In group 2, E_cw was measured without any complication. Median E_cw was significantly lower than in group 1, p = 0.001 and was not correlated to aTLC/pTLC ratio.

After transplantation, mean TLC_transplanted/pTLC was significantly higher than in group 1, p = 0.023, Figure 2. Seven patients received an oversized lung with TLC_transplanted/pTLC > 100%, none of them had a compartment syndrome, and 2 patients had lung resection.

Figure 2.

Comparison of the ratio of transplanted lung volume/predicted total lung capacity between groups 1 and 2. This figure illustrates the amount of lung volume transplanted, represented by the ratio of transplanted donor TLC with recipient predicted TLC in %. In group 1, E_cw was measured on the day of lung transplantation after donor selection and in group 2, E_cw was measured during pretransplant medical check-up and used for donor selection. With E_cw measurement before lung transplantation (group 2), more patients had an oversized donor ratio >100% (green area) and less had lung resection (red dot). E_CW, chest wall elastance; pTLC, predicted total lung capacity; TLC, total lung capacity.

If we pool the 2 groups, patients with primary graft dysfunction (PGD) 3 had higher median E_cw 12.8 [10.5-20.2] n = 5 vs 7.7 [6.7-12.5] n = 35 cmH₂O/liter, p = 0.031; whereas median aTLC/pTLC was not different: 53.3 [46.8-71.9] vs 49.2 [40.0-60.7]%, p = 0.222.

Discussion

To our knowledge, this study is the first to demonstrate that the common increase of the elastance of the whole respiratory system in patients with lung fibrosis can also be attributed to the chest wall component and not only the fibrotic lungs, with a large range of E_cw. Interestingly, patients with high E_cw measurements were more likely to have smaller transplanted lung, lung resection with lobar transplantations, more Clamshell incisions, and more PGD.

E_cw appeared to be the sole measurement being able to obtain this crucial information, since patients with high and low E_cw were not different in terms of pulmonary diseases, pulmonary functional tests, and CT scan assessments. Indeed, TLC and CT scan will find low lung volume because of fibrosis but it will not give any information about chest wall stiffness. During transplantation, for a same low aTLC, some chest cavity can get back to normal size while others keep a chest retraction, and the physiological difference will come from the pleural pressure. So the question is, what is the pleural pressure for a given low TLC measured with CT scan and pulmonary function tests.

Patients with lung resection had extremely high E_cw as previously described in severe morbidly obese acute respiratory distress syndrom patients with a mean E_cw of 17.2 ± 1.7 cmH₂O/liter.³ Lastly, E_cw was not correlated to the duration of lung disease, which cannot explain chest cavity retraction.

In group 2, donor lung size selection was optimized as the amount of transplanted lung was significantly higher than in group 1, with less lung resection.

One of the limitations is that unmasking chest stiffness will come from pleural pressure measurement, which is invasive. However, it is a simple method used by intensivist and feasible in selected patients.

Conclusion

There is a wide range of E_cw in patients with end-stage PF. Unlike medical history, preoperative TLC, and CT scan assessment, E_cw correlates to the volume of the chest cavity once the fibrotic lungs have been removed. It can be measured safely during the pretransplant medical check-up to optimize donor lung size selection.

CRediT authorship contribution statement

I hereby declare that I'm the corresponding author of the manuscript. I have not used any sources other than those listed in the bibliography and identified as references. I further declare that I have not submitted this manuscript to any other journal. All authors are accountable for all aspects of this study. The manuscript is approved by all authors and tacitly by the responsible authorities where the work was carried out. H.R. designed the study. H.R. and M.T. wrote the manuscript. All coauthors collected data in their field for the study (pneumologists, radiologists, anesthesiologists, surgeons, and intensivists).

Disclosure statement

The authors of this study have no conflict of interest.

We would like to thank Erwan Floch, PharmD, from Newmed Publishing Services for reviewing the manuscript.

This study was supported solely by the Department of Thoraco-Abdominal Anesthesiology and Critical Care.

Appendix A. Supplementary data

Supplementary material

.

Supplementary material

.

Supplementary material

.

Supplementary material

.