Short-term effect of volume recruitment–derecruitment manoeuvre on chest-wall motion in Duchenne muscular dystrophy

Abstract

Because progressive respiratory muscle weakness leads to decreased chest-wall motion with eventual ribcage stiffening, the purpose was to compare vital capacity (VC) and contributions of chest-wall compartments before and after volume recruitment–derecruitment manoeuvres (VRDM) in Duchenne muscular dystrophy (DMD). We studied nine patients with DMD and VC lower than 30% of predicted. VRDM was performed using 15 insufflations–exsufflations of +30 to −30 cmH₂O. VC and three-dimensional chest-wall motion were measured, as well as oxygen saturation, transcutaneous partial pressure of carbon dioxide and the rapid shallow breathing index (respiratory rate/tidal volume) before (baseline) and immediately and 1 hour after VRDM. VC increased significantly immediately after VRDM (108% ± 7% of baseline, p = 0.018) but returned to baseline within 1 hour, and the rapid shallow breathing index increased significantly. The non-dominant side systematically increased immediately after VRDM (p = 0.0077), and in the six patients with abnormal breathing asymmetry (difference >10% of VC) at baseline, this asymmetry was corrected immediately and/or 1 hour after VRDM. VRDM improved VC and reduced chest-wall motion asymmetry, but this beneficial effect waned rapidly with respiratory muscle fatigue, suggesting that VRDM may need to be repeated during the day to produce lasting benefits.

Introduction

Respiratory insufficiency in patients with Duchenne muscular dystrophy (DMD) is due to a combination of respiratory muscle weakness and diminished pulmonary and chest-wall compliance, which decreases vital capacity (VC). Thus, VC is an accurate overall indicator of respiratory function in this disease.^1–6

The progressive decrease in respiratory system compliance may result from limited movement during spontaneous yawns and sighs, which normally prevent stiffening of the chest wall⁷ while also spreading surfactant over the air–fluid interface and re-inflating areas of atelectasis.⁸ Patients whose respiratory muscle weakness prevents them from fully performing volume recruitment–derecruitment manoeuvres (VRDM) may be most likely to benefit from mechanical insufflation–exsufflation to increase lung volume. In DMD, regular mechanical insufflation of the lungs was associated with slowing of the VC decline over time⁹ and, after 1 year, with an increase in VC.¹⁰ In addition to these long-term improvements, mechanical insufflation–exsufflation may also provide immediate benefits. Data on immediate effects of positive-pressure lung inflation in patients with restrictive respiratory disease are conflicting: two studies showed marked improvements in lung mechanics,^11,12 whereas two others found no effects on lung and chest-wall compliance.^13,14

To our knowledge, no study has evaluated the effects of VRDM in patients with DMD and severely restricted lung volumes. Here, our objective was to evaluate the short-term effects of mechanical insufflation–exsufflation on VC and the contribution to these effects of each chest-wall compartment as assessed by 3D motion analysis using optoelectronic plethysmography (OEP).

Materials and methods

The study was approved by our institutional review board and registered on ClinicalTrials.gov [#NCT 02236104]. Patients gave written informed consent before study inclusion.

Patients

The study was performed between March 2014 and June 2015 at the home ventilation department of the medical intensive care unit (ICU) at the Raymond Poincaré Teaching Hospital, Garches, France. We recruited nine consecutive patients with DMD seen at the post-ICU clinic for their annual follow-up visits, which routinely include patient education about cough-assist techniques. At each visit, the following were recorded routinely: VC,¹⁵ maximum expiratory pressure, maximum inspiratory pressure and sniff test measurements.^5,16 VC was measured with the same device and by the same technician (LF), as described below.

Inclusion criteria were documented DMD, age >18 years, haemodynamic stability, absence of acute bronchial congestion (respiratory tract infection) in the past month, use of noninvasive ventilation and VC lower than 30% of predicted.

Volume recruitment–derecruitment manoeuvres

VRDM was performed using the CoughAssist® mechanical insufflator–exsufflator (Philips Respironics, Murrysville, Pennsylvania, USA). On the day before testing, the patient was familiarized with the cough-assist techniques and device settings used for the study. For the study, with the patient supine and connected to the device via a sealed facemask, 15 insufflations–exsufflations were applied, each with +30 cmH₂O of inspiratory pressure for 2 s and −30 cmH₂O of expiratory pressure for 3 s, separated by 30-s intervals. Patients were instructed to allow production by the device of full inhalations and exhalations. The facemask (Laerdal Medical, Limonest, France) was firmly held on the patient’s face during the in–exsufflations to avoid leaks.

Lung volume measurements

Measurements were performed with the patient supine, on three occasions, that is, immediately before the VRDM (baseline), immediately after the VRDM (post-VRDM) and 1 hour after the VRDM (1-hour post-VRDM). Flow was measured during a VC manoeuver using a spirometer (Micro 5000; Medisoft, Sorinnes, Belgium) and during spontaneous breathing using a Fleisch #1 pneumotachometer (Lausanne, Switzerland) connected to a differential pressure transducer (Validyne, Northridge, California, USA), connected in turn to a mouthpiece. The patient wore a nose clip. In addition, capillary oxygen saturation (SpO₂) was measured using a SOMNOscreen™ pulse oximeter (SOMNOmedics, Randersacker, Germany) and transcutaneous partial pressure of carbon dioxide (PtcCO₂) using a StandAloneDigital Monitoring System (SenTec, Therwil, Switzerland). The data were recorded at a sampling rate of 100 Hz with a digital computer system (MP 150; Biopac System, Santa Barbara, California, USA) and then used to derive breathing pattern variables.

Chest-wall motion assessments

Simultaneously, an OEP recording system (CX1; Codamotion System, Charnwood Dynamics Ltd, Rothley, Leicester, UK) was used to study chest-wall volume distribution and changes during breathing. As described by LoMauro et al.,¹⁷ 52 active markers were positioned on the anterior chest wall from the clavicles to the pubic bone. The three-dimensional coordinates of these markers were recorded at a sampling rate of 100 Hz. A fourth-order reverse and zero-phase forward Butterworth digital filter with a cut-off frequency of 3 Hz was applied to smooth the three-dimensional coordinates.¹⁸ Signals from the Biopac and motion capture systems were synchronized by the same computer clock using timestamps.

Outcomes

As previously done,^1–6 we chose the change in VC as the primary outcome measure. Changes were also evaluated in various chest-wall compartments (ribcage vs. abdomen and dominant side vs. non-dominant side). The non-dominant side was defined as the side contributing less than 50% of VC at baseline. Secondary outcomes were breathing pattern variables (tidal volume (VT), respiratory rate (RR) and rapid shallow breathing index (RR/VT)), SpO₂, PtcCO₂ and patients’ perception of respiratory difficulty assessed using the modified 10-point Borg scale.¹⁹

Statistical analysis

Friedman’s test was used for global comparisons of the three time points (baseline, post-VRDM and 1-hour post-VRDM). When the results showed a significant difference, pairwise comparisons were conducted using Wilcoxon’s test. When necessary, linear regression analysis was performed to evaluate correlations between two values. Values of p ≤ 0.05 were considered significant.

Results

Population

Table 1 lists the main characteristics of the nine consecutive patients included during the study period. All nine patients had surgery to correct scoliosis.

Table 1.

Characteristics of the nine patients with severe Duchenne muscular dystrophy.

Patient (#)	Age (years)	BMI (kg/m2)	VC (%)	MIP (cmH2O)	MEP (cmH2O)	Mechanical ventilation (h/24 h)	Arterial blood gas
							Evening before MV		Morning after MV
							pH	PaCO2 (mmHg)	pH	PaCO2 (mmHg)
1	22	28.0	14	15	24	8	7.33	70	7.42	50
2	20	23.7	25	24	39	12	7.43	38	7.51	33
3	18	14.2	16	25	23	10	7.37	49	7.38	47
4	19	14.9	15	23	22	4	7.36	48	NA	NA
5	20	9.3	11	20	22	16	7.36	47	7.40	43
6	23	20.5	11	26	15	8	7.41	42	7.40	42
7	30	11.0	7	7	8	12	7.38	55	7.49	42
8	20	11.5	25	36	28	10	7.42	42	7.46	30
9	21	15.9	18	34	23	8	7.40	40	NA	NA

BMI: body mass index; VC: vital capacity measured using a spirometer and expressed as the % of the predicted value; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; MV: mechanical ventilation; NA: not available.

VC and contributions of chest-wall compartments

As shown in Figure 1, VC increased significantly from baseline to post-VRDM (Friedman, p = 0.0098; Wilcoxon, p = 0.018) but not to 1-hour post-VRDM (Wilcoxon, p = 0.29).

Figure 1.

Vital capacity at baseline, immediately after the VRDM (post-VRDM) and 1 hour after the VRDM. VRDM: volume recruitment–derecruitment manoeuvre.

Figure 2(a) and (b), illustrates the contributions of the ribcage and abdominal compartments to VC at baseline and after VRDM. VRDM had no significant effect on either compartment (Friedman: ribcage, p = 0.12 and abdomen, p = 0.17) and Figure 2(c) and (d) shows the contributions of the non-dominant and dominant sides to VC at baseline and after VRDM. On the non-dominant side, the change in volume was consistently increased immediately after VRDM (Friedman, p = 0.0043; Wilcoxon vs. baseline, p = 0.0077) but not to 1 hour after VRDM (Wilcoxon vs. baseline, p = 0.67). On the dominant side, in contrast, the change in volume was not increased immediately after VRDM (Friedman, p = 0.0498; Wilcoxon vs. baseline, p = 0.95) and was decreased 1 hour after VRDM (Wilcoxon vs. baseline, p = 0.028).

Figure 2.

Contributions of different compartments to VC, reported as percentage of baseline VC, at baseline, immediately after the VRDM (post-VRDM) and 1 hour after the VRDM (1-hour post-VRDM). (a) Displacement of ribcage volume; (b) displacement of abdominal volume; (c) displacement of the volume of the non-dominant side; (d) displacement of the volume of the dominant side. VC: vital capacity; VRDM: volume recruitment–derecruitment manoeuvre.

Abnormal asymmetry in chest-wall expansion (>10% difference in % of VC)¹⁸ was noted at baseline in patients #1, #3, #4, #6, #7 and #8 (Figure 2(c)). VRDM consistently corrected this abnormality (Figure 3).

Figure 3.

Difference in contribution to VC between the dominant and non-dominant sides at baseline, immediately after the VRDM (post-VRDM) and 1 hour after the VRDM (1-h post-VRDM). VC: vital capacity; VRDM: volume recruitment–derecruitment manoeuvre.

Secondary outcomes

Neither SpO₂ nor PtCO₂ changed significantly after VRDM. The Borg score for perceived breathing difficulty was not significantly modified by VRDM. VT remained unchanged, whereas both RR and the rapid shallow breathing index (RR/VT) increased significantly at 1-hour post-VRDM (Table 2).

Table 2.

Effect of the VRDM in the nine patients.

	Baseline	Post-VRDM	1-hour post-VRDM
MIP (cmH2O)	24 ± 10	22 ± 10	21 ± 9
MEP (cmH2O)	24 ± 9	22 ± 9	23 ± 9
VT (ml)	316 ± 61	310 ± 60	275 ± 27
RR (breaths/min)a,b	21 ± 4	19 ± 5	23 ± 4
RR/VT (breaths/min/L)a,b	66 ± 7	61 ± 5	84 ± 16
PtcCO2 (mmHg)	48 ± 6	47 ± 6	49 ± 6
SpO2 (%)	97 ± 2	97 ± 2	97 ± 2
Borg score	1.16 ± 0.96	1.33 ± 1.08	1.61 ± 1.36

VRDM: volume recruitment–derecruitment manoeuvre; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; VT: tidal volume; RR: respiratory rate; SpO₂: transcutaneous capillary oxygen saturation; PtcCO₂: transcutaneous partial pressure of carbon dioxide.

^aFriedman (p < 0.05) and Wilcoxon (p < 0.05) between baseline and post-VRDM; no other significant effects were observed.

^bFriedman (p < 0.05) and Wilcoxon (p < 0.05) between post-VRDM and 1-hour post-VRDM; no other significant effects were observed.

Discussion

This is the first systematic comparison of VC and chest-wall compartment distribution before and after mechanical VRDM in patients with DMD and severely restricted lung volumes. VC increased significantly immediately after the VRDM but returned to baseline within the first hour. Abnormal asymmetry in chest-wall expansion¹⁸ at baseline was noted in two-thirds of patients. VRDM routinely improved motion of the non-dominant side and thereby diminishing the asymmetry.

To understand the baseline asymmetry in chest-wall expansion, we retrospectively examined the chest radiographs of the study patients. No spinal or pulmonary changes potentially responsible for asymmetry were identified (data not shown). The ability of VRDM to diminish the asymmetry did not support a permanent anatomical abnormality as the cause of the asymmetry. Moreover, as the mediastinum is mobile between the two sides of the chest cavity, there is probably no simple correlation between chest expansion asymmetry and lung inflation asymmetry. To elucidate the interactions between the lungs and chest wall under normal and pathological conditions, Lin et al. developed a model consisting of two lungs, which could have different properties, enclosed in a chest wall and separated by a compliant mediastinum.^20,21 They evaluated situations of asymmetric lung injury with symmetric chest-wall expansion.²¹ Because the mediastinum was compliant, it was displaced by pressure differences between the two sides of the chest cavity, allowing transmission of these pressures. Thus, inflation of one lung tended to increase the pleural pressure on the other side, limiting inflation of the other lung and promoting asymmetric lung expansion, whereas chest-wall expansion remained symmetric. Consistent with these findings, a study involving OEP and computed tomography after single-lung transplantation for emphysema showed asymmetric ventilation between the two lungs with symmetric changes in volume of the two hemithoraces.²² To our knowledge, no study has investigated volume changes in the two hemithoraces when atelectasis is present. We therefore do not know whether, contrary to unilateral emphysema, asymmetric ventilation is associated with asymmetric ribcage motion. The effect of asymmetric ventilation probably depends on disease severity and therefore on the mechanical properties of the different components of the model developed by Lin et al.²¹ Situations characterized by major modifications in local lung compliance, such as severe atelectasis, can probably result in asymmetric chest-wall expansion. However, such a mechanism is unlikely in our patients, given that their chest radiographs showed no evidence of unilateral atelectasis. This is in accordance with a previous study in which computed tomography imaging of the lung failed to detect atelectasis in patients with chronic respiratory muscle weakness and reduced lung distensibility.²³ This study concluded that lung compliance reduction might be related to alterations in the elasticity of lung tissue. Finally, this asymmetry, which decreased after VRDM, suggests an asymmetric stiffening of the ribcage, which can be partially reversed after VRDM. The mechanism of this stiffness is probably multifactorial. It could be related in part to ankylosis of costosternal and costovertebral joints that might depend on the position immediately before the evaluation. Ribcage muscle or connective tissue elements abnormalities and residual spinal deformities, remaining after surgical correction of scoliosis, could also reduce regional chest-wall compliance. Nevertheless, independently from the potential explanatory mechanisms, these results suggest that VRDM can transiently reverse regional ribcage stiffness and then could improve the efficiency of physiotherapy and airway clearance; considering that the chest-wall stiffening previously described in patients with chronic respiratory muscle weakness⁷ could be modified by VRDM, it could therefore improve mobilization during assisting cough manoeuvres making them more efficient.

The VC improvement was lost within 1 hour of spontaneous breathing after VRDM. This loss was associated with an increase in the rapid shallow breathing index, suggesting that the patients were unable to remain stable without mechanical ventilation. Although the maximal pressures were not significantly modified, the patients probably experienced respiratory muscle fatigue, which may have contributed to the VC decline, in combination with the gradual decline in the beneficial effect of VRDM. This explanation is consistent with the severe respiratory insufficiency in the study patients, all of whom had VC values <30% of predicted and five of whom had severe hypercapnia (>46 mmHg) at the end of the day just before starting nocturnal mechanical ventilation. Most of the study patients had been referred to our hospital to start extending their nocturnal mechanical ventilation to part of the day. Furthermore, in eight of the nine patients, the abdominal contribution to VC was less than 30%, confirming the severe muscle weakness, in accordance with previous reports of a decrease in the abdominal contribution to breathing during the progression of DMD.¹⁷ The only patient whose VC did not improve immediately after VRDM was the patient with the most severe respiratory acidosis at the end of the day, suggesting that the failure of the VRDM was ascribable to limited respiratory self-sufficiency.

Conclusion

This study provides further evidence that VRDM can be beneficial in DMD patients with severe restrictive syndrome and should be considered as a component of usual care. The VC increase immediately after VRDM may explain the long-term beneficial effects of two daily VRDM sessions reported previously.¹⁰ The short-lived effect of VRDM suggests a need for repeating the sessions at closer intervals during the day.