Intercept of minute ventilation versus carbon dioxide output relationship as an index of ventilatory inefficiency in chronic obstructive pulmonary disease.

BACKGROUND: Ventilatory inefficiency contributes to exercise intolerance in chronic obstructive pulmonary disease (COPD). The intercept of the minute ventilation (V˙ E) vs. carbon dioxide output (V˙ CO2) plot is a key ventilatory inefficiency parameter. However, its relationships with lung hyperinflation (LH) and airflow limitation are not known. This study aimed to evaluate correlations between the V˙ E/V˙ CO2 intercept and LH and airflow limitation to determine its physiological interpretation as an index of functional impairment in COPD.
METHODS: We conducted a retrospective analysis of data from 53 COPD patients and 14 healthy controls who performed incremental cardiopulmonary exercise tests (CPETs) and resting pulmonary function assessment. Ventilatory inefficiency was represented by parameters reflecting the V˙ E/V˙ CO2 nadir and slope (linear region) and the intercept of V˙ E/V˙ CO2 plot. Their correlations with measures of LH and airflow limitation were evaluated.
RESULTS: Compared to control, the slope (30.58±3.62, P<0.001) and intercept (4.85±1.11 L/min, P<0.05) were higher in COPDstages1-2, leading to a higher nadir (31.47±4.47, P<0.01). Despite an even higher intercept in COPDstages3-4 (7.16±1.41, P<0.001), the slope diminished with disease progression (from 30.58±3.62 in COPDstages1-2 to 26.84±4.96 in COPDstages3-4, P<0.01). There was no difference in nadir among COPD groups and higher intercepts across all stages. The intercept was correlated with peak V˙ E/maximal voluntary ventilation (MVV) (r=0.489, P<0.001) and peak V˙ O2/Watt (r=0.354, P=0.003). The intercept was positively correlated with residual volume (RV) % predicted (r=0.571, P<0.001), RV/total lung capacity (TLC) (r=0.588, P<0.001), peak tidal volume (VT)/FEV1 (r=0.482, P<0.001) and negatively correlated with rest inspiratory capacity (IC)/TLC (r=-0.574, P<0.001), peak VT/TLC (r=-0.585, P<0.001), airflow limitation forced expiratory volume in 1 s (FEV1) % predicted (r=-0.606, P<0.001), and FEV1/forced vital capacity (FVC) (r=-0.629, P<0.001).
CONCLUSIONS: V˙ E/V˙ CO2 intercept was consistently correlated with worsening static and dynamic LH, pulmonary gas exchange, and airflow limitation in COPD. The V˙ E/V˙ CO2 intercept emerged as a useful index of ventilatory inefficiency in COPD patients. 2021 Journal of Thoracic Disease. All rights reserved.

Introduction

Activity-related dyspnea is the defining complaint in patients with chronic obstructive pulmonary disease (COPD) (1). The limitations in activity and dyspnea are multifactorial. The development of lung hyperinflation (LH) plays an important role in the pathophysiologies of dyspnea and exercise intolerance (2). LH has a static and dynamic component. Static lung hyperinflation (sLH) is defined as functional residual capacity (FRC) equal to increased relaxation respiratory volume (Vr) resulting from loss of lung elastic recoil, while dynamic lung hyperinflation (dLH) when FRC exceeds Vr (3). dLH that occurs when the expiratory time becomes insufficient to allow the lung to achieve full exhalation yields an increased end-expiratory lung volume (EELV) during exercise (4). LH increases ventilatory workload and decreases inspiratory muscle pressure-generating capacity, despite some compensatory mechanisms (5). The diminished ventilatory capacity coupled with the increased ventilatory demand during exercise results in exercise intolerance.

The minute ventilation (V˙ E) vs. carbon dioxide production (V˙ CO2) relationship is a measure of the ventilatory efficiency at removing CO2 produced by the body. Early in exercise, V˙ E/V˙ CO2 decreases due to a lower in dead space ventilation (VD)/tidal volume (VT) ratio. When the lactic acidosis first develops, both V˙ E and V˙ CO2 increase curvilinearly, resulting in no further change in V˙ E/V˙ CO2 and PETCO2. (isocapnic buffering). The V˙ E/V˙ CO2 nadir is typically reached at isocapnic buffering period and then ventilation starts to increase to compensate for lactic acidosis at the respiratory compensation point (RCP) (6). The V˙ E/V˙ CO2 nadir was found to be highly reproducible in healthy subjects (7) and COPD patients (8). However, the V˙ E/V˙ CO2 nadir might underestimate ventilatory inefficiency if the descending curve is prematurely interrupted by lactic acidosis or a very short test duration (9). On the other hand, the V˙ E/V˙ CO2 ratio might be higher than the nadir due to the hyperventilation response to late-exercise acidosis in patients who can exercise beyond the RCP (10).

The V˙ E/V˙ CO2 slope has been used to assess disease progression and identify the presence of comorbidities (11-16). However, in many patients with moderate-to-severe COPD, the partial pressure of carbon dioxide (PaCO2) and mechanical constraints will predictably flatten the V˙ E/V˙ CO2 curve. In these patients, the V˙ E/V˙ CO2 slope might paradoxically decrease as the disease progresses if CO2 retention worsens during exercise. It is plausible that the V˙ E/V˙ CO2 nadir might be stable while the slope and intercept change in opposite directions despite COPD progression (10).

The V˙ E/V˙ CO2 intercept is a novel parameter derived from V˙ E vs. V˙ CO2 relationship by extrapolating a regression line during exercise, which theoretically equates to the VD (17) and cannot be constrained by dynamic mechanics (unlike the slope) or the test duration (unlike the nadir) (10,18). Indeed, V˙ E/V˙ CO2 = 863/[PaCO2 × (1 − VD/VT)]. If two periods—PaCO2 and the VD/VT ratio are constant during exercise, no intercept can appear. Otherwise, we can show an intercept during exercise because PaCO2 increases with greater exercise intensity, even in the absence of VD.

The V˙ E/V˙ CO2 intercept depends on the patterns of change in multiple parameters during exercise. The increased intercept in COPD patients might result from an altered breathing strategy such as increased breathing frequency to compensate for reduced VT secondary to greater mechanical constraints, and/or a progressive ventilation-perfusion mismatch in COPD patients (1). Thus, the V˙ E/V˙ CO2 intercept increases with greater disease severity in COPD patients, and it seems to be a particularly useful index for ventilatory inefficiency across the COPD severity continuum (10). However, the clinical implications of the V˙ E/V˙ CO2 intercept and its association with LH in COPD have not been formally examined except few study reported the relationship between intercept and dLH (19).

This study aimed to evaluate the relationship between the V˙ E/V˙ CO2 intercept and LH and airflow limitation in patients with COPD. We hypothesized that the V˙ E/V˙ CO2 intercept correlated well with measures of both LH and airflow limitation and could be a particularly useful index for ventilatory inefficiency in COPD.

We present the following article in accordance with the STROBE reporting checklist (available at http://dx.doi.org/10.21037/jtd-20-2725).

Methods

Study participants

This study was a retrospective analysis of incremental cardiopulmonary exercises data collected from Jan. 1, 2016 to Dec. 31, 2019 at the Respiratory Investigation Unit, Beijing Friendship Hospital, Capital Medical University (Beijing, China). The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). Participants were males and females aged ≥40 years with body mass index (BMI) of 18–35 kg/m2 and given informed consent before taking part in. The study was approved by the Institutional ethics committee of Beijing Friendship Hospital of Capital Medical University (No. 2018-P2-048-01) and the written informed consent was obtained from every participant. The 53 patients were current or ex-smokers (smoking history ≥10 pack-years) and had a well-established diagnosis of COPD (20) without asthma or other pulmonary diseases. Patients were required to have had no exacerbation in the preceding 6 weeks. Fourteen control subjects with no smoking history were in the same age range and they had no major orthopedic, neuromuscular, cardiac or metabolic diseases, to allow them to safely undertake the incremental exercise tests.

Pulmonary function tests

Each subject underwent resting spirometry (MasterScreen Body, CareFusion, Hoechberg, Germany), including inspiratory capacity (IC) assessment. Body plethysmography was performed to measure residual volume (RV), TLC and diffusing capacity of the lungs for carbon monoxide (DLCO). Patients took 400 µg albuterol by inhalation 20 min before testing. All pulmonary function tests fulfilled the American Thoracic Society/European Respiratory Society guidelines (21).

Cardiopulmonary exercise test (CPET)

Symptom-limited incremental exercise testing was performed on an electronically braked cycle ergometer (ViaSprint, CareFusion, Hoechberg, Germany) with a pedaling rate of 60/min. After 3 min of rest and 3 min of unloaded pedaling, the work rate (WR) was increased by 5−15 W/min in a ramp fashion (5 W/min if FEV1 <1.0 L and 10 W/min if FEV1 ≥1.0 L for the COPD patients; 15 W/min for the control subjects, with repetition at 20 W/min if the peak WR was ≥200 W). Participants were asked to continue to exercise to the limit of tolerance, marked by the inability (despite encouragement) to maintain pedaling frequency or intolerable shortness of breath. Any participant with chest pain suggestive of ischemia, ventricular tachycardia and blood pressure (BP) ≥240/130 mmHg was prevented further exercise. Participants were continuously monitored with a 12-lead electrocardiogram and blood pressure by sphygmomanometer every 2 min.

Data collection

Respiratory gas exchange (V˙ E, V˙ O2, and V˙ CO2) and VT were measured breath-by-breath throughout the exercise testing. Serial measurements of these parameters were averaged at consecutive 30 s data. Arterial oxygen saturation was measured noninvasively by pulse oximetry (SpO2%). The V˙ E/V˙ CO2 nadir and peak V˙ E/V˙ CO2 were averaging the consecutive 30 s data points at lowest and peak time, respectively (7). The slope of the V˙ E/V˙ CO2 relationship was determined based on the V˙ E vs.V˙ CO2 plot (V˙ E on the y-axis and V˙ CO2 on the x-axis). A linear regression line was determined based on exercise data points, excluding data above the respiratory compensation point (7,22). The V˙ E/V˙ CO2 intercept was calculated by extrapolating the regression line to V˙ CO2 =0. Maximal voluntary ventilation (MVV) was calculated as FEV1 ×40. Peak VT/FEV1 was considered as emphysema factor (19) and peak VT/TLC as dLH parameter during exercise (23) while IC/TLC, RV % predicted and RV/TLC were used as sLH (24).

Statistical analysis

The sample size required was based on compare k means: one-way analysis of variance (ANOVA) pairwise, two-sided equality for α of 0.05, a power (1-β) of 0.8 to detect this difference by using Medcalc Statistical Software version 19.1.3 (Medcalc Software BV, Ostend, Belgium; https://www.medcalc.org; 2019). Values were reported as mean ± SD unless otherwise stated. P value <0.05 was considered significant in all analyses. Intraclass correlation coefficients were used to determine the level of between-investigator agreement in the calculation of the slope and intercept. Between-group comparisons were performed using one-way ANOVA with LSD post-hoc testing of significant variables. Pearson’s correlation coefficient (r) was used to assess the correlations between the ventilatory inefficiency parameters (V˙ E/V˙ CO2 intercept, slope and nadir) and peak V˙ E/MVV, peak V˙ O2/Watt, static and dynamic hyperinflation measures (RV % predicted, RV/TLC, IC/TLC and peak VT/TLC, peak VT/FEV1), and airflow limitation (FEV1% predicted and FEV1/FVC). All analyses were performed with SPSS Statistics 20.0 (IBM, Chicago, USA).

Results

Participant characteristics and resting spirometric measurements

As shown in , the COPDstages1-2 (n=35), COPDstages3-4 (n=18) and control (n=14) were well matched in terms of age, weight, and BMI. There are the expected decreases in FEV1% predicted, FEV1/FVC, IC % predicted, IC/TLC, DLCO % predicted and the expected increases in RV % predicted, RV/TLC, from COPDstages1-2 to COPDstages3-4 (P<0.001). shows demographics and selected resting pulmonary function variables in control and COPD patients.

Table 1

Demographics and selected resting pulmonary function variables in control and COPD patients

Variables	Control (n=14)	COPDstages1-2 (n=35)	COPDstages3-4 (n=18)	ANOVA P value
Male, n (%)	12 (85.71)	31 (88.57)	16 (88.89)	–
Age (yr)	61±7.65	64.66±7.65	65.61±6.79	0.193
Height (cm)	167.29±4.84	168.09±7.6	170.33±5.67	0.377
Weight (kg)	70.5±9.88	68.43±10.49	66.88±8.1	0.595
BMI (kg/m2)	25.07±3.81	24.52±2.76	23.03±2.3	0.108
FEV1% predicted (%)	102.65±15.46	71.84±13.57***	39.87±7.4***###	0.000
FEV1/FVC (%)	81.74±6.67	61.05±6.89***	43.22±6.9***###	0.000
TLC % predicted (%)	92.3±11.07	98.33±12.21	103.12±16.76*	0.084
IC % predicted (%)	96.73±16.04	90.47±15.41	69.66±16.45***###	0.000
IC/TLC (%)	44.86±4.68	39.32±7.1*	30.14±7.39***###	0.000
RV % predicted (%)	97.66±17.91	134.58±30.73**	174.68±52.44***###	0.000
RV/TLC (%)	39.6±6.12	52.53±5.1***	62.71±8.61***###	0.000
DLCO % predicted (%)	87.2±8.83	72.32±13.84**	43.47±14.42***###	0.000

Data are presented as mean ± standard deviation. The * and # labeled for results of the post-hoc test. *, P<0.05 vs. control; **, P<0.01 vs. control; ***, P<0.001 vs. control; ###, P<0.001 COPDstages3-4 vs. COPDstages1-2. BMI, body mass index; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; IC, inspiratory capacity; TLC, total lung capacity; RV, residual volume; DLCO, diffusing capacity of the lung for carbon monoxide.

Exercise characteristics

shows the exercise variables at peak exercise in control and COPD patients. All subjects completed the exercise testing without complications. No one repeated the CPET and no one peak workload more than 200 W. Among them ranges of exercise duration were found from 6 to 10 min. Peak exercise capacity was progressively reduced from the control to COPDstages1-2 and COPDstages3-4 patients. Measures at peak exercise showed that the COPDstages3-4 patients had significantly reduced V˙ E, V˙ O2, V˙ O2% predicted, V˙ O2/HR, V˙ CO2and WR except peak V˙ E/MVV in comparison to control (P<0.001). COPDstages3-4 patients had significantly reduced V˙ O2% predicted, V˙ E, and WR while peak V˙ E/MVV increased in comparison to COPDstages1-2 (P<0.001). During exercise, the peak VT in COPDstages3-4 (1.3±0.2) was lower than in COPDstages1-2 (1.64±0.35), and significantly lower than control (1.79±0.22) (P<0.001). Considering emphysema factor, peak VT/FEV1 (%) in COPDstages3-4 (114±23.28) was higher than in COPDstages1-2 (84.72±17.18), and significantly higher than control (65.49±12.95) (P<0.001). There was also significant difference between COPDstages1-2 and COPDstages3-4 (P<0.001). Regarding the measures of dLH peak VT/TLC (%), compared to control (33.15±5.5), COPDstages1-2 (27.45±5.02, P<0.01) and COPDstages3-4 (19.88±4.78, P<0.001) patients all exhibited significant difference; there was also significant difference between COPDstages1-2 and COPDstages3-4 (P<0.001).

Table 2

Selected variables at peak exercise in control and COPD patients

Variables	Control (n=14)	COPDstages1-2 (n=35)	COPDstages3-4 (n=18)	ANOVA P value
V˙ O2 (L/min)	1.77±0.34	1.37±0.27***	1.14±0.26***##	0.000
V˙ O2% predicted (%)	99±17.2	77.29±12.47***	60.94±12.25***###	0.000
V˙ O2/HR (mL/beat)	11.92±2.03	10.37±1.77**	9.39±1.49***	0.001
V˙ E (L/min)	59.21±13.93	53.71±11.02	40.94±8.61***###	0.000
V˙ CO2 (L/min)	2.1±0.44	1.63±0.34***	1.25±0.31***##	0.000
V˙ E/V˙ O2	33.37±4.43	39.49±6.17*	36.84±7.32	0.009
V˙ E/V˙ CO2	27.65±3.12	32.51±4.78**	32.48±6.08**	0.007
V˙ E/V˙ CO2 slope	24.75±3.07	30.58±3.62***	26.84±4.96##	0.000
V˙ E/V˙ CO2 nadir	26.9±2.92	31.47±4.47**	32.82±5.29***	0.001
V˙ E/V˙ CO2 intercept (L/min)	3.91±1.03	4.85±1.11*	7.16±1.41***###	0.000
V˙ E/MVV (%)	53.29±12.04	69.17±12.67**	89.52±17.7***###	0.000
WR (W)	140.79±25.16	107.26±24.92***	80.44±16.86***###	0.000
V˙ O2/Watt (mL/min/W)	12.65±1.6	12.97±1.33	14.14±1.51**##	0.007
peakVT (L)	1.79±0.22	1.64±0.35	1.3±0.2***##	0.000
peakVT/TLC (%)	33.15±5.5	27.45±5.02**	19.88±4.78***###	0.000
peakVT/FEV1 (%)	65.49±12.95	84.72±17.18***	113.93±23.28***###	0.000

Data are presented as mean ± standard deviation. The * and # labeled for results of the post-hoc test. *, P<0.05 vs. control; **, P<0.01 vs. control; ***, P<0.001 vs. control; ##, P<0.01 COPDstages3-4 vs. COPDstages1-2; ###, P<0.001 COPDstages3-4 vs. COPDstages1-2. V˙ O2, oxygen uptake; V˙ E, minute ventilation; V˙ CO2, carbon dioxide output; HR, heart rate; MVV, maximal voluntary ventilation; WR, work rate; VT, tidal volume; TLC, total lung capacity; FEV1, forced expiratory volume in 1 s.

Ventilatory inefficiency in COPD patients

The V˙ E/V˙ CO2 relationships were expressed in terms of the slope, nadir, and intercept. Compared to control (24.75±3.07), the slope was increased in COPDstages1-2 (30.58±3.62) and decreased in COPDstages3-4 (26.84±4.96). The nadir was increased in COPDstages1-2 (31.47±4.47, P<0.01) and almost stable in COPDstages3-4 (32.82±5.29, P<0.001) in comparison to control (26.9±2.92). There was no difference between COPDstages1-2 and COPDstages3-4. As for the intercept, the COPDstages1-2 and COPDstages3-4 had higher intercepts (4.85±1.11, P<0.05), and (7.16±1.41, P<0.001) respectively in comparison to control (3.91±1.03). There was significant difference between COPDstages3-4 and COPDstages1–2 (P<0.001). V˙ E/V˙ CO2 intercept increased across the severity spectrum of COPD. Among ventilatory inefficiency parameters (slope, nadir and intercept), only V˙ E/V˙ CO2 intercept exhibited correlation with peak V˙ O2/Watt (r=0.354, P=0.003) and peak V˙ E/MVV (r=0.489, P<0.001). shows measures of ventilatory inefficiency in control and COPD patients. shows V˙ E/V˙ CO2 intercept in correlation with peak V˙ E/MVV and peak V˙ O2/Watt in the entire study group.

Figure 1

Measures of ventilatory inefficiency in control and COPD patients. (A) V˙ E/V˙ CO2 intercept, (B) V˙ E/V˙ CO2 nadir, and (C) V˙ E/V˙ CO2 slope in control and COPD patients. (A) The COPDstages1-2 and COPDstages3-4 had higher intercepts in comparison to control. V˙ E/V˙ CO2 intercept increased across the severity spectrum of COPD (P<0.001). (B) The nadir was increased in COPD patients in comparison to control. There was no difference between COPDstages1-2 and COPDstages3-4. (C) Compared to control, the slope was increased in COPDstages1-2 (P<0.001) and then decreased in COPDstages3-4. There was no difference between COPDstages3-4 and control. COPD, chronic obstructive pulmonary disease; V˙ E, minute ventilation; V˙ CO2, carbon dioxide output.

Figure 2

Correlations between V˙ E/V˙ CO2 intercept and peak V˙ E/MVV and peak V˙ O2/Watt in the entire study group. (A) V˙ E/V˙ CO2 intercept vs. peak V˙ E/MVV (r=0.489, P<0.001); (B) V˙ E/V˙ CO2 intercept vs. peak V˙ O2/Watt (r=0.354, P<0.01). MVV, maximal voluntary ventilation; V˙ E, minute ventilation; V˙ CO2, carbon dioxide output; V˙ O2, oxygen uptake.

Correlation of ventilatory inefficiency with LH and airflow limitation

shows the correlation coefficients of the ventilatory inefficiency parameters and measures of LH, airflow limitation. The relationships between the measures of ventilatory inefficiency and LH and airflow limitation were assessed in COPD patients. The V˙ E/V˙ CO2 intercept was correlated with the measures of sLH and dLH (), rest IC/TLC (r=−0.574, P<0.001), RV % predicted (r=0.571, P<0.001), RV/TLC (r=0.588, P<0.001), peak VT/FEV1 (r=0.482, P<0.001) and peak VT/TLC (r=−0.585, P<0.001) respectively while the V˙ E/V˙ CO2 slope not. The V˙ E/V˙ CO2 nadir was also correlated with rest IC/TLC (r=−0.35, P=0.004), RV % predicted (r=0.383, P=0.001) RV/TLC (r=0.431, P<0.001), and peak VT/TLC (r=−0.503, P<0.001).

Table 3

Correlation coefficients for ventilatory inefficiency parameters with measures of LH and airflow limitation

Variables	V˙ E/V˙ CO2 slope			V˙ E/V˙ CO2 nadir			V˙ E/V˙ CO2 intercept
	r	P value		r	P value		r	P value
Peak VT/TLC (%)	−0.148	0.232		−0.503	0.000		−0.585	0.000
Peak VT/FEV1	−0.253	0.036		0.195	0.113		0.482	0.000
IC/TLC (%)	−0.006	0.962		−0.35	0.004		−0.574	0.000
RV % predicted (%)	0.146	0.242		0.383	0.001		0.571	0.000
RV/TLC (%)	0.191	0.121		0.431	0.000		0.588	0.000
FEV1/FVC (%)	−0.167	0.178		−0.434	0.000		−0.629	0.000
FEV1% predicted (%)	−0.064	0.609		−0.368	0.002		−0.606	0.000
V˙ E/MVV (%)	0.181	0.142		0.424	0.000		0.489	0.000
V˙ O2/Watt (mL/min/W)	−0.058	0.639		0.213	0.083		0.354	0.003

VT, tidal volume; IC, inspiratory capacity; TLC, total lung capacity; RV, residual volume, FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; V˙ E, minute ventilation; MVV, maximal voluntary ventilation; V˙ O2, oxygen uptake.

A similar pattern of results was found in correlation to airflow limitation. The V˙ E/V˙ CO2 intercept showed correlation with FEV1% predicted (r=−0.606, P<0.001) and FEV1/FVC (r=−0.629, P<0.001) while V˙ E/V˙ CO2 slope not. V˙ E/V˙ CO2 nadir was also correlated with FEV1% predicted (r=−0.368, P=0.002) and FEV1/FVC (r=0.424, P<0.001).

Discussion

The main finding of this study was that the V˙ E/V˙ CO2 intercept was consistently correlated with worsening static and dynamic LH and increasing airflow limitation in COPD. This suggests that it could be a useful index for ventilatory inefficiency during incremental exercise in COPD.

Abnormalities in the V˙ E/V˙ CO2 relationship were present across the spectrum of COPD severity. Increases in both the V˙ E/V˙ CO2 nadir and slope were associated with lower maximal exercise capacity in COPD patients (10,25-27). In our study, compared to control, COPDstages1–2 had a higher slope and nadir. For healthy subjects who can tolerate high levels of exercise, the V˙ E/V˙ CO2 nadir and ratio at the anaerobic threshold were usually very similar (7). As the lactate threshold may not always be identified, particularly in clinical populations with low exercise capacity (28), the V˙ E/V˙ CO2 nadir seems a more accurate indication of ventilatory efficiency than the V˙ E/V˙ CO2 at the lactate threshold. However, the V˙ E/V˙ CO2 nadir, might underestimate ventilatory efficiency for early lactic acidosis or an excessively short test duration (9). The results showed patients with COPDstages1-2 had a higher slope and nadir, while COPDstages3-4 had a lower slope, and a stable nadir. The V˙ E/V˙ CO2 intercept, by definition, cannot be constrained by worsening mechanics (as the slope) or test duration (as the nadir). It should be recognized that including the data points after the RCP will necessarily increase the computed slope and decrease the computed intercept. Furthermore, including the data points after the RCP does not accurately reflect V˙ E vs. V˙ CO2 plot profile. In present study, we excluded the data above the RCP for linear regression to calculate the slope and intercept. The V˙ E/V˙ CO2 intercept was higher across all stages in current study. This is in line with a retrospective study reported by Neder of 316 patients with a large range of resting pulmonary function (FEV1 =12−148% predicted) (10). A lower slope, stable nadir, and a higher intercept in more advanced COPD stage likely reflected the opposite changes in V˙ E/V˙ CO2 slope and intercept.

There is mounting evidence that ventilatory inefficiency parameters are powerful prognostic predictors in COPD patients with comorbidity. A retrospective study of 145 COPD patients undergoing surgery for non-small cell cancer (NSCLC) showed that a V˙ E/V˙ CO2 slope >34 predicted mortality after lung resection surgery (29). A high-intensity pulmonary rehabilitation programs could improve ventilatory efficiency in patients with COPD undergoing lung resection for NSCLC (V˙ E /V˙ CO2 slope from 32.0±2.8 to 30.1±4.0, P<0.01) (30). As for the V˙ E/V˙ CO2 nadir, Neder et al. reported that a value >34 in combination with resting hyperinflation predicted mortality in COPD (31). Importantly, a series of studies demonstrated that the V˙ E/V˙ CO2 intercept (cutoff values ranging from 2.64−4.07 L/min) might discriminate COPD from heart failure (32,33).

Ventilatory inefficiency increases ventilatory demand and exercise capacity limitation in COPD patients. The increase of peak V˙ E/MVV across all stages reflected the worsening breathing reserve in COPD. Interestingly, the V˙ E/V˙ CO2 nadir and intercept were highly positively related to the peak V˙ E/MVV in the current study. V˙ O2-work rate relationship describes how much O2 is utilized by exercise subject in relation to the quantity of external work performed. Among the V˙ E/V˙ CO2 slope, nadir and intercept, only V˙ E/V˙ CO2 intercept was positively related to the peak V˙ O2/Watt (). This was consistent with Neder’s reports that V˙ E/V˙ CO2 intercept was correlated with the peak metabolic stress (V˙ O2peak) (10).

Two other independent studies found correlations between the V˙ E/V˙ CO2 nadir and emphysema severity on high-resolution computed tomography scans in COPD patients with largely preserved FEV1 (34,35). Teopompi reported that patients with the ratio of peak VT/FEV1 ≥1 had a predominant pulmonary emphysema profile, thereby showing a higher degree of airflow obstruction (19). We found the V˙ E/V˙ CO2 intercept was positively related to peak VT/FEV1 and particularly the ratio of peak VT/FEV1 in COPDstages3-4 was more than 1 (1.13±0.23). The consequence of pulmonary parenchyma destruction and loss of elastic recoil is static hyperinflation. Emphysema and development of expiratory flow limitation promote progressive air trapping with an increase in the EELV and a decrease in IC. RV was also increased in emphysema/COPD due to both loss of elastic recoil and premature closure of the small airways (3,36,37). In expiratory flow-limited patients, EELV was a continuous dynamic variable dependent on expiratory duration and breathing pattern. DH refers to this temporary and variable increase in EELV. As ventilation increases and expiratory duration decreases, there is not enough time to allow EELV to decline to its baseline resting value (38).

Studies reported that both sLH and the degree of dLH were associated with the development of dyspnea and exercise intolerance in COPD patients (39,40). Assuming TLC stability, the resting IC and inspiratory reserve (IRV) showed the operating position of VT relative to the TLC. The smaller the resting IC, the shorter the exercise time before VT plateaus and dyspnea abruptly escalates (41). A 4-year longitudinal study reported that significant reductions in peak V˙ O2 and V˙ E were related to a decreased resting IC (42). Both rest IC/TLC and RV/TLC in patients with COPD reflected the degree of sLH and also the functional reserve. Rest IC/TLC was also found to be a valuable and independent predictor of all-cause and respiratory mortality in COPD compared with that of the BODE (body mass index, airflow obstruction, dyspnea, exercise performance) index (43). The present study showed that the V˙ E/V˙ CO2 intercept was more strongly correlated with a series of highly relevant parameters, rest IC/TLC (r=−0.574, P<0.001), RV % predicted (r=0.571, P<0.001) and RV/TLC (r=0.588, P<0.001), and V˙ E/V˙ CO2 nadir with rest IC/TLC (r=−0.35, P=0.004), RV % predicted (r=0.383, P=0.001), RV/TLC (r=0.431, P<0.001). The V˙ E/V˙ CO2 slope had no correlation with sLH parameters.

A progressive increase in EELV, while IC decreases was associated with dyspnea and exercise intolerance in COPD during exercise (44). Serial measurements of IC to detect its changes were reported as a classic way to identify dynamic hyperinflation (39,40,45). However, the study participants had to be familiar with the maneuvers, and IC measurements also had to be standardized by researchers (46). Dynamic IC measurement is not recommended for ramp-pattern protocols where VT cannot steadily proceed to perform the IC maneuver. However, the ramp-pattern protocol is a widely used for incremental testing (46). Elevated EELV can substantially constrain VT expansion at higher exercise intensities. It is logical that COPD patients reach a VT plateau and have similar minimal inspiratory reserve volume. Chuang et al. investigated peak VT/TLC as a convenient new marker of DH and reported a cutoff of 0.27 (23). In this study, we used peak VT/TLC instead of ∆ (peak-rest) IC as the index of dLH. Among ventilatory inefficiency parameters (slope, nadir, and intercept), we found that only the V˙ E/V˙ CO2 intercept exhibited a better correlation to peak VT/TLC (r=−0.585, P<0.001). To our knowledge, this is the first study to describe the relationship between ventilatory inefficiency parameters and rest IC/TLC, RV/TLC and peak VT/TLC as static and dynamic LH index. Interestingly, the V˙ E/V˙ CO2 intercept was also correlated with worsening pulmonary airflow limitation as indicated by FEV1/FVC (r=−0.629, P<0.001) and FEV1% predicted (r=−0.606, P<0.001). This was consistent with Teopompi’s reports, expressed as FEV1% predicted (r=−0.482, P<0.001) (19) and FEV1/VC in subjects with COPD (r=−0.38, P=0.009) (47).

A limitation of our study is the small number of subjects although well matched proportions of female patients in three groups (2/14 in control, 4/35 in COPDstages1-2, and 2/18 in COPDstages3-4, respectively). We did not correlated the V˙ E/V˙ CO2 slope, nadir and intercept with dyspnea in current study. Notably, our findings of a significantly lower V˙ E/V˙ CO2 slope in COPDstages3-4 compared with COPDstages1-2 might be due to changes in VD/VT, which we did not calculation in present study. We believe that the increased ventilatory inefficiency associated with LH might be more pronounced in patients with more advanced COPD. However, in the absence of a true criterion test for ventilatory inefficiency during exercise, we relied on a set of variables that are indirect markers of pulmonary gas-exchange disturbances. We also recognize that variables related to disease phenotypes and test factors (e.g., duration) affect different strategies to reflect ventilatory inefficiency. Therefore, we can only infer, not establish the mechanism underlying the ventilatory inefficiency in COPD patients during exercise. Further an additional study with VD/VT and PaCO2 measurements is needed.

Conclusions

V˙ E/V˙ CO2 intercept increased across the severity spectrum of COPD patients and consistently correlated with exercise capacity. Increases in V˙ E/V˙ CO2 intercept were consistently correlated with worsening static and dynamic LH and airflow limitation. A hitherto underappreciated variable, the V˙ E/V˙ CO2 intercept, was found to be a particularly useful index of ventilatory inefficiency during incremental exercise in COPD patients. Further comprehensive studies are promised to corroborate these preliminary results.

The article’s supplementary files as