Abstract
In patients with exacerbation of chronic obstructive pulmonary disease (ECOPD), the presence of respiratory acidosis is considered “a marker of severity”, identifying the need for ventilation treatment. The evaluation of ECOPD patients with hypercapnia, even without acidosis, may provide information on patients not commonly defined as “at risk”. We retrospectively assessed 407 hospitalised patients with ECOPD, divided into three groups at admission: patients with normocapnia (N = 176), hypercapnia (PaCO2 > 45 mmHg and pH ≥ 7.35, N = 126), and acidosis (PaCO2 > 45 mmHg and pH < 7.35, N = 105). Data on general, clinical, laboratory and microbiological characteristics, and on outcomes (mortality up to 3 years), were collected. Patients with hypercapnia and acidosis had similar functional and clinical characteristics to their normocapnic peers, but of greater severity. The mortality rate at 1 year was similar between hypercapnic and acidotic patients (24% and 25%), and higher than that recorded in the normocapnic group (14%). Similar trends were observed in mortality from 6 months to 3 years. The presence of hypercapnia and acidosis was significantly associated with an increased and independent risk of death at 1 year. A threshold value of PaCO2 ≥ 55 mmHg was associated with 1-year mortality. In a multivariate analysis considering the new threshold and compared to normocapnia, the presence of hypercapnia (PaCO2 ≥ 55), regardless of acidosis, was independently associated with a higher mortality risk at 1 year. In hospitalised ECOPD patients, hypercapnia itself, even when compensated, is associated with a worse prognosis up to an intermediate-term follow-up. The need to ventilate these patients should be evaluated, regardless of the presence or absence of acidosis.
Supplementary Information
The online version contains supplementary material available at 10.1007/s11739-026-04279-0.
Keywords: COPD exacerbations, Hospitalisation, Hypercapnia, Acidosis, Prognosis, Outcomes
Introduction
Chronic obstructive pulmonary disease (COPD) represents a significant and increasing challenge for health systems worldwide, due to its high impact on morbidity, disability, and mortality [1]. Multiple factors contribute to progressive skeletal muscle wasting in COPD patients [2], and in those who also have severe airway obstruction, the presence of alveolar hypoventilation may impair gas exchange, leading to hypercapnia [3]. The presence of hypercapnia, defined by an increase in the partial pressure of arterial carbon dioxide (PaCO2) >45 mmHg, is therefore a sign of ventilatory exhaustion and has not only pulmonary but also systemic impacts, including cardiovascular and musculoskeletal consequences [3]. In very severe COPD patients, the presence of hypercapnia is around 25% [4].
The exacerbation of COPD (ECOPD), interrupting the stable phase of the disease, worsens several outcomes, especially if patients need to be hospitalised [5]. In these patients, hypercapnia is primarily caused by mechanical alterations in the ventilation/perfusion (V/Q) ratio, resulting from airway resistance (bronchospasm), flow limitation, oedema of the airway wall and accumulation of mucus hypersecretion [3]. Moreover, the hyperactivity of the respiratory neural drive, characterised by a rapid and shallow breathing pattern, causes dynamic hyperinflation and pulmonary entrapment and an increase in the intrinsic positive end-expiratory pressure (iPEEP), which represents an additional mechanical load (work of breathing, WOB) which the respiratory system must overcome in order to guarantee respiratory gas exchanges [3]. All these factors, leading to increased oxygen demand and a reduction in muscle respiratory efficiency, may contribute to the development of hypercapnia during ECOPD [3, 6]. The presence of hypercapnia may be associated with respiratory acidosis, defined by a pH < 7.35 [6]. At admission due to ECOPD, the presence of acidosis is around 20% [7].
Bilevel non-invasive mechanical ventilation (NIMV), a key treatment for addressing iPEEP and WOB in ECOPD patients with hypercapnia and acidosis, was proposed in the 2017 European Respiratory Society (ERS) guidelines with strong evidence-based recommendations [8]. However, the same guidelines conditionally recommend that NIMV not be used in ECOPD patients with hypercapnia without acidosis [8]. In 2019, however, the ERS task force conditionally supported the long-term domiciliary use of NIMV in stable COPD with chronic hypercapnia to improve health outcomes [9].
In the recent definition and grading of ECOPD (the Rome proposal), the presence of acidosis was considered a marker of ECOPD, indicating severity, which differentiates it from moderate (characterised by hypercapnia without acidosis) and mild grades (considered the default) [10]. In terms of prognosis, however, the Rome classification may not be able to discriminate between patients with moderate or severe disease [11]. These findings brought into question whether hypercapnia without acidosis can be harmful to patients with ECOPD. The hypothesis of this study is that hypercapnia may play a predominant role in patients admitted for ECOPD. Therefore, the study aimed to evaluate the impact of hypercapnia at admission and the outcomes of hospitalised patients with ECOPD.
Methods
Study design and patients
We retrospectively analysed data of prospectively hospitalised ECOPD patients. The sampling method was systematic, and all patients with ECOPD admitted to the Pulmonology Department of Hospital Clínic of Barcelona (Spain) between May 2009 and March 2018 were considered for enrolment. COPD was confirmed by clinical history and spirometry performed during a stable phase at least 6 months prior to admission, when fixed airflow obstruction was evaluated after administration of a bronchodilator (salbutamol). COPD and ECOPD were defined according to the GOLD 2023 document [12], and the need for hospitalisation was established based on ECOPD severity in accordance with clinical signs and symptoms and the presence of potential indicators [13]. Acute respiratory failure was defined according to the values of arterial blood gas (ABG) measured at admission: the ratio of partial pressure of arterial oxygen to the fraction of inspired oxygen (PaO2/FiO2) <300 defined hypoxemia, while hypercapnia was defined when patients had a PaCO2 > 45 mmHg, with or without acidosis (pH < 7.35 and pH ≥ 7.35, respectively) [8]. Among admitted patients, we excluded those with concomitant chronic respiratory disease other than COPD (such as asthma or pulmonary fibrosis), as well as patients with community-acquired pneumonia (CAP) or acute heart failure, identified by clinical and radiological criteria. A consolidation on the thorax X-ray, accompanied by a higher inflammatory response as indicated by C-reactive protein (CRP) >12.9 mg/dL, was used as a criterion to identify CAP in patients with COPD and to differentiate it from ECOPD [14]. Moreover, we excluded patients in whom an ABG was not performed and those with metabolic acidosis (pH < 7.35 and PaCO2 ≤ 45 mmHg). The hospital’s Ethics Committee approved the study protocol (CEIC 2008/4106), and the study was conducted in accordance with good clinical practices and the Declaration of Helsinki. All enrolled patients provided informed consent.
Definition of study groups
The study cohort was divided into three groups: patients with normocapnia (PaCO2 ≤ 45 mmHg), patients with compensated hypercapnia (pH ≥ 7.35 and PaCO2 > 45 mmHg, hence defined only as hypercapnia), and patients with respiratory acidosis (pH < 7.35 and PaCO2 > 45 mmHg, hence defined only as acidosis). Supplementary Fig. 1 displays a flow diagram of the study.
Measurements
We considered data on demographic variables, body mass index (BMI), smoking habit (never, current or former), with the number of packs/years, number of co-morbidities (Charlson index), dyspnoea grade measured by the modified Medical Research Council (mMRC) scale, and the use of long-term oxygen therapy (LTOT). The number of exacerbations occurring in the previous year was also recorded. Furthermore, respiratory rate and heart rate were assessed at admission. In addition to ABG data (pH, PaO2/FiO2, PaCO2, HCO3−-serum bicarbonate, and BE-base excess), other laboratory values (leucocytes, haemoglobin, CRP and creatinine) were recorded at admission and day 3. Moreover, the number of patients using systemic corticosteroids and antibiotics was recorded, as well as the number of positive sputum cultures, with the prevalence of microbiologic agents.
Primary outcome
The primary outcome was all-cause, 1-year mortality (since admission).
Secondary outcomes
Secondary outcomes were: (a) all-cause, in-hospital and at 30- and 90-day, 6-, 24- and 36-month mortality (since admission); (b) length of hospital stay (LHS); (c) the use of NIMV or invasive mechanical ventilation (IMV); (d) the need for intensive care unit (ICU) admission; (e) occurrence of new episodes of ECOPD and related readmissions within 30 days of discharge.
Statistical analysis
For categorical variables, we reported the number and percentage of patients by category, and for continuous variables the median (IQR-interquartile range). To compare categorical and continuous variables among study groups, we used the chi-squared test or the Freeman-Halton extension [15] of the Fisher exact test for contingency tables that were larger than 2 × 2, when appropriate, and the non-parametric Kruskal–Wallis test, respectively. We performed pairwise comparisons to compare different study groups using the Bonferroni method only for secondary outcomes. The assumption of normality was checked with Kolmogorov–Smirnov tests.
Survival curves for the study groups were obtained using the Kaplan–Meier method and compared using the Gehan–Breslow–Wilcoxon test [16]. We then analysed the association between the study groups and 1-year mortality using a Cox regression multivariable model [17] adjusting for potential confounders. The following variables, selected according to their potential clinical relevance, were included in the analysis: age, sex, body mass index (BMI), smoking habit, forced expiratory volume in the first second (FEV1), LTOT, Charlson Index and previous ECOPD. Regarding exacerbations, we retained the variable “number of previous ECOPD ≥ 2” for the multivariable analysis, while “previous hospitalisation/s for ECOPD” was excluded due to high collinearity with the former. Single collinearity in the multivariable analysis was evaluated using the Spearman correlation coefficient (r); multicollinearity was examined using the variance inflation factor (VIF), and variables with high collinearity (r >|± 0.50|) were excluded. Proportional hazard assumptions were tested using log-minus-log plots. Hazard ratios (HRs) and their 95% confidence intervals (CIs) were calculated. Assessment of discrimination in the multivariable model was measured using Harrell’s C and Somers’ D. Patients who were lost to follow-up were censored in survival analyses. We also performed appropriately adjusted analyses for secondary endpoints. Finally, the receiver operating characteristic (ROC) analysis, with the area under the curve (AUC), was used to identify the threshold of PaCO2 which was significantly stratified with the mortality at 1 year. A multivariable Cox regression model was repeated with the threshold of PaCO2 in order to assess factors associated with mortality at 1 year.
The level of significance was set at 0.05 (two-tailed). All analyses were performed using IBM SPSS version 26.0 (IBM Corp., Armonk, NY, USA).
Results
General, clinical, laboratory and microbiological results
We considered 407 patients with ECOPD, divided into three groups: patients with normocapnia (N = 176, 43%), hypercapnia (N = 126, 31%), and acidosis (N = 105, 26%). Patients with normocapnia showed significant differences in smoking (number of packs/year), FEV1 and FEV1/FVC, as well as LTOT, with respect to patients with hypercapnia and acidosis. Patients with acidosis had a higher prevalence of comorbidities compared to those with normocapnia. Table 1 displays the general characteristics.
Table 1.
General characteristics of study groups
| Variables | Total cohort (N = 407) |
Patients with normocapnia (N = 176) |
Patients with hypercapnia (N = 126) |
Patients with acidosis (N = 105) |
p-value |
|---|---|---|---|---|---|
| Age, years | 72 [14] | 73 [15] | 72 [12] | 71 [13] | 0.611 |
| Male | 323 (80) | 135 (77) | 97 (77) | 91 (87) | 0.111 |
| BMI, kg/m2 | 27.4 [6.7] | 27.3 [6.3] | 27.5 [6.6] | 27.5 [7.7] | 0.928 |
| Smoking habit | 0.914 | ||||
| Never | 14 (3) | 5 (3) | 5 (4) | 4 (4) | |
| Current | 163 (40) | 70 (40) | 48 (38) | 45 (43) | |
| Former | 230 (57) | 101 (57) | 73 (58) | 56 (53) | |
| Packs/year | 56 [40] | 50 [30.5] | 57.5 [50] | 60 [40] | 0.012a,c |
| FEV1, % of predicted | 40.5 [24] | 50 [28] | 36 [22] | 38 [14.5] | <0.001b,d |
| FEV1/FVC | 47 [23] | 52 [23.5] | 47 [23] | 43 [19.5] | 0.001a,d |
| mMRC ≥2 | 172 (61) | 66 (52) | 66 (72) | 40 (63) | 0.742 |
| GOLD 2023 stages | 0.970 | ||||
| A | 73 (23) | 38 (27) | 19 (18) | 16 (22) | |
| B | 98 (30) | 42 (30) | 34 (32) | 22 (30) | |
| E | 152 (47) | 62 (44) | 54 (50) | 36 (49) | |
| Charlson index ≥2 | 232 (59) | 86 (51) | 77 (63) | 69 (68) | 0.015c |
| Ischaemic heart disease | 38 (9) | 19 (11) | 9 (7.2) | 10 (9.5) | 0.563 |
| Congestive heart disease | 61 (15) | 21 (12) | 22 (18) | 18 (17) | 0.333 |
| Diabetes | 88 (22) | 31 (18) | 31 (25) | 26 (25) | 0.244 |
| ECOPD ≥2e | 96 (24) | 40 (23) | 34 (27) | 22 (21) | 0.588 |
| ECOPD hospitalisations ≥1e | 131 (32) | 51 (29) | 49 (39) | 31 (30) | 0.182 |
| Domiciliary triple therapy | 128 (37) | 60 (39) | 34 (32) | 34 (39) | 0.466 |
| LTOT | 111 (27) | 29 (16) | 45 (36) | 37 (35) | <0.001a,c |
Data are presented as the number of patients (percentage) or medians [interquartile range]. Percentages are calculated for non-missing data. In bold significant values.
a and bp < 0.05 and p < 0.001 for comparison between patients with normocapnia and hypercapnia, respectively;
c and dp < 0.05 and p < 0.001 for comparison between patients with normocapnia and acidosis, respectively;
e occurring in the preceding year.
BMI indicates body mass index, FEV1 Forced expiratory volume in the 1 st second, FVC Forced vital capacity, mMRC Modified Medical Research Council, GOLD Global Initiative for Chronic Obstructive Lung Disease, ECOPD Exacerbation of the chronic obstructive pulmonary disease, LTOT Long-term oxygen therapy.
In the clinical and laboratory variables evaluated at admission (Supplementary Table 1), ABG measures differed among the three groups in pH, PaCO2, and BE, while PaO2/FiO2 and HCO3− differed between patients with normocapnia and those with hypercapnia and acidosis. Again, at admission, CRP levels differed between hypercapnic patients (higher) and acidotic patients (lower), while the creatinine value was higher in patients with acidosis than in those with normocapnia and hypercapnia. At day 3, pH, PaCO2, HCO3− and BE differed between patients with normocapnia and those with hypercapnia and acidosis. The prevalence of patients treated with systemic corticosteroids was higher in hypercapnic and acidotic groups than in the normocapnic group.
Among microbiological variables (Supplementary Table 2), the prevalence of patients having Pseudomonas aeruginosa infection was higher in the hypercapnia group, while that of Streptococcus pneumoniae was higher in the normocapnia group.
Outcomes
Primary
The all-cause mortality rate at 1 year was higher in patients with hypercapnia (24%) and acidosis (25%) than in those with normocapnia (14%), as well as the survival time. The Kaplan–Meier curves (Fig. 1) showed significant differences between the three study groups (Gehan–Breslow–Wilcoxon test, p = 0.024), particularly between patients with normocapnia and those with acidosis.
Fig. 1.
Kaplan–Meier survival curves at 1 year. ap = 0.063 for comparison between patients with normocapnia and hypercapnia; p = 0.040 for comparison between patients with normocapnia and acidosis; p > 0.999 for comparison between patients with hypercapnia and acidosis
The ROC curve of PaCO2 (Supplementary Fig. 2) shows the best threshold of 55 mmHg associated with mortality at 1 year (AUC 0.625; SE 0.036; 95%CI 0.553 to 0.696; p = 0.001). The Kaplan–Meier curves at 1 year (Fig. 2) showed a significant difference between patients with a PaCO2 value of 55 mmHg or more and those with a value less than 55 mmHg (p < 0.001 on the Gehan–Breslow–Wilcoxon test).
Fig. 2.
Kaplan–Meier survival curves at 1 year using the PCO2 cut-off of 55 mmHg
Secondary
Differences in mortality were observed between patients with normocapnia (7.5%) and those with acidosis (17%) (at 6 months), between normocapnic (21%) and both hypercapnic and acidosis groups (34% and 37%) (at 2 years), and between normocapnic and hypercapnic (31% vs 46%) (at 3 years).
The LHS was shorter in patients with normocapnia (median, 7 days) than in those with hypercapnia and acidosis (8 and 9 days, respectively). The prevalence of patients treated with NIMV and IMV was significantly higher in acidosis than in the other two groups. The prevalence of patients admitted to the ICU showed a significant sequential increase between normocapnic, hypercapnic and acidosis, while the prevalence of patients having a new ECOPD within 30 days of discharge was higher in acidotic patients than in their normocapnic peers.
All study outcomes are reported in Table 2.
Table 2.
Study outcomes
| Variables | Patients with normocapnia (N = 176) |
Patients with hypercapnia (N = 126) |
Patients with acidosis (N = 105) |
p-value |
|---|---|---|---|---|
| Primary outcome | ||||
| Mortality at 1 year | 24 (14) | 30 (24) | 26 (25) | 0.032 |
| Survival time, days | 337 [325 to 349] | 310 [291 to 330] | 307 [285 to 328] | 0.024c |
| Secondary outcomes | ||||
| In-hospital mortality | 5 (2.8) | 5 (4) | 4 (3.8) | 0.832 |
| Mortality at 30 days | 4 (2.3) | 6 (4.8) | 5 (4.8) | 0.372 |
| Mortality at 90 days | 10 (5.7) | 13 (10) | 10 (9.5) | 0.287 |
| Mortality at 6 months | 13 (7.5) | 20 (16) | 18 (17) | 0.023c |
| Survival time, days | 173 [169 to 179] | 164 [156 to 172] | 164 [156 to 173] | 0.027c |
| Mortality at 2 years | 34 (21) | 42 (34) | 37 (37) | 0.006a,c |
| Survival time, days | 638 [608 to 669] | 565 [520 to 610] | 562 [512 to 613] | 0.004a,c |
| Mortality at 3 years | 49 (31) | 53 (46) | 44 (46) | 0.015a |
| Survival time, days | 916 [865 to 966] | 787 [715 to 859] | 778 [698 to 857] | 0.006a,c |
| Length of hospital stay, days | 7 [4] | 8 [4] | 9 [6] | 0.001a,d |
| NIMV | 4 (2.3) | 22 (18) | 73 (70) | <0.001c,f |
| IMV | 3 (1.7) | 2 (1.6) | 12 (12) | <0.001c,f |
| ICU admission | 5 (2.9) | 9 (7.2) | 44 (43) | <0.001a,c,e |
| New ECOPD after dischargeg | 21 (12) | 25 (21) | 25 (25) | 0.023c |
| Readmission for ECOPDg | 4 (2.3) | 2 (1.6) | 3 (2.9) | 0.837 |
Data are presented as the number of patients (percentage) or medians [interquartile range], except for survival time, which is calculated as the mean estimate [95%CI]. Percentages are calculated for non-missing data. In bold significant values.
a and bp < 0.05 and p < 0.001 for comparison between patients with normocapnia and hypercapnia, respectively; c and dp < 0.05 and p < 0.001 for comparison between patients with normocapnia and acidosis, respectively;
e and fp < 0.05 and p < 0.001 for comparison between patients with hypercapnia and acidosis, respectively;
g within 30 days of discharge.
NIMV and IMV indicate non-invasive and invasive mechanical ventilation, respectively, ICU Intensive care unit, ECOPD Exacerbation of the chronic obstructive pulmonary disease.
Factors associated with mortality at 1 year
The multivariable Cox regression model (Table 3), with normocapnic patients as the reference group, showed a significant and increased mortality risk for those with hypercapnia (HR 1.91; 95%CI 1.09 to 3.35) and those with acidosis (HR 2.14; 95%CI 1.21 to 3.78). Moreover, older patients (≥80 years), overweight patients, and patients with more than two previous ECOPD episodes also had an increased risk of death at 1 year.
Table 3.
Multivariable Cox regression model assessing factors associated with mortality at 1 year, evaluated with the three study groups
| Variables | HR | 95%CI | p-value |
|---|---|---|---|
| Study groups | |||
| Patients with normocapnia | 1 | – | – |
| Patients with hypercapnia | 1.91 | 1.09 to 3.35 | 0.023 |
| Patients with acidosis | 2.14 | 1.21 to 3.78 | 0.009 |
| Age | |||
| <65 years | 1 | – | – |
| 65 to 79 years | 1.89 | 1.00 to 3.59 | 0.051 |
| ≥80 years | 3.37 | 1.63 to 6.97 | 0.001 |
| BMI | |||
| <18 kg/m2 | 0.85 | 0.26 to 2.82 | 0.790 |
| 18 to 24 kg/m2 | 1 | – | – |
| 25 to 29 kg/m2 | 0.48 | 0.26 to 0.92 | 0.026 |
| ≥30 kg/m2 | 0.81 | 0.43 to 1.51 | 0.497 |
| Previous ECOPD ≥2 | 3.15 | 2.00 to 4.95 | <0.001 |
Harrell’s C statistic is 0.71 and Somers’ D statistic is 0.42.
HR indicates hazard ratio, CI Confidence interval, BMI Body mass index, ECOPD Exacerbation of the chronic obstructive pulmonary disease.
The multivariable Cox regression model, using the threshold of 55 mmHg for PaCO2 (Table 4), showed that patients with PaCO2 ≥ 55 mmHg either with acidosis (HR 1.95; 95%CI 1.12 to 3.39) or without (HR 2.87; 95%CI 1.60 to 5.14) had an increased risk of death at 1 year with respect to those without acidosis and with PaCO2 < 55 mmHg. In the same model, again, older patients and those with more than two previous ECOPD, together with the presence of LTOT, also had a higher mortality risk.
Table 4.
Multivariable Cox regression model assessing factors associated with mortality at 1 year, evaluated according to the presence of respiratory acidosis and the cut-off of 55 mmHg of PaCO2
| Variables | HR | 95%CI | p-value |
|---|---|---|---|
| Respiratory acidosis and hypercapnia | |||
| Patients without acidosis (pH ≥ 7.35) and with PaCO2 <55 mmHg | 1 | – | – |
| Patients without acidosis (pH ≥ 7.35) and with PaCO2 ≥55 mmHg | 2.87 | 1.60 to 5.14 | <0.001 |
| Patients with acidosis (pH < 7.35) and with PaCO2 <55 mmHg | 1.97 | 0.69 to 5.61 | 0.202 |
| Patients with acidosis (pH < 7.35) and with PaCO2 ≥55 mmHg | 1.95 | 1.12 to 3.39 | 0.019 |
| Age | |||
| <65 years | 1 | – | – |
| 65 to 79 years | 1.67 | 0.88 to 3.17 | 0.119 |
| ≥80 years | 2.96 | 1.44 to 6.05 | 0.003 |
| Presence of LTOT | 2.37 | 1.46 to 3.86 | 0.001 |
| Previous ECOPD ≥2 | 2.67 | 1.69 to 4.23 | <0.001 |
Harrell’s C statistic is 0.74 and Somers’ D statistic is 0.48.
HR indicates hazard ratio, CI Confidence interval, LTOT Long-term oxygen therapy, ECOPD Exacerbation of the chronic obstructive pulmonary disease.
Discussion
In patients with COPD, the presence of hypercapnia is a marker of functional severity and progressive systemic involvement [3]. In hospitalised patients with ECOPD, hypercapnia is considered a marker of severity only if associated with acidosis [10]; this is the only selection criterion for NIMV treatment [8]. We demonstrate that hypercapnia at admission, even if not associated with acidosis, may worsen the prognosis of ECOPD patients both during hospitalisation and in the 3-year follow-up after discharge. Moreover, the threshold at admission of PaCO2 above 55 mmHg, rather than 45, allows better identification of patients “at risk” for a worse prognosis, regardless of the presence or absence of acidosis.
May hypercapnia without acidosis be a marker of ECOPD severity?
A severe airflow limitation is a significant risk factor associated with hypercapnic failure in ECOPD patients [18], causing profound mechanical ventilatory alterations such as the presence of iPEEP and increased WOB which can be adequately addressed using NIMV [3, 8]. However, in ECOPD patients, there is solid evidence regarding the effect of the use of NIMV on the mortality rate of acidotic patients, but not in hypercapnic patients without acidosis [8], which conceptually defines these compensated patients as less severe than those with acidosis. Along these lines, the new Rome classification categorises the ECOPD hypercapnic as moderate, applying the definition of “severe ECOPD” only to acidotics [10]. Our findings challenge these considerations, and some additional clarifications are needed.
First, the limited evidence profile of NIMV efficacy in hypercapnic non-acidotic ECOPD stems from only a few studies (and patients), and these studies also present some methodological concerns [8]. Keenan’s study [19], for example, a randomised-controlled trial (RCT), considering 52 patients only and powered for a reduction of LHS and not for mortality, considered the application of bilevel NIMV for a few hours a day and not continuously for a more reasonable time. Additionally, the mean initial settings provided in that study likely resulted in low inspiratory pressures and had a less pronounced effect. Physiologically, high-intensity pressure ventilation is more effective than low-intensity in decreasing elevated PaCO2, reducing inspiratory effort, and alleviating dyspnoea [20]. Moreover, patients ventilated with high-intensity pressures may have significant advantages in terms of the need for endotracheal intubation (ETI) [21] and short-term mortality [22]. Finally, Keenan’s study considered patients with a relatively low mean PaCO2 (50 mmHg), indicating mild to moderate ventilatory impairment, which may show slight improvement with NIMV during hospitalisation.
Second, the evaluation of the efficacy of NIMV on hypercapnic ECOPD has been primarily assessed by mortality outcomes [8], while more solid findings may be obtained by considering ETI [8, 23]. In fact, a prospective multicentre RCT conducted in 19 Chinese hospitals recorded that, in a subgroup of ECOPD ventilated patients (N = 151) who were not acidotic, there was a reduction in reporting meeting criteria for ETI (3% vs 11%), similar to patients with acidosis (7% vs 27%) [23].
Third, in the context of ECOPD, acidosis is an intercurrent phenomenon related to individual and temporal factors of metabolic compensation, primarily achieved through renal retention of bicarbonate [24], and therefore not closely associated with the severity of ventilatory lung failure. Patients with acute kidney injury (AKI), a common complication occurring in 21% of ECOPD [25] impairing the kidney’s capacity to compensate for acidosis, exhibit several worse outcomes with respect to patients without AKI,, including higher mortality rates at 30 days (14% vs 4%), ETI (54% vs 14%), and more severe hypercapnic acidosis [26]. Conversely, the development of sufficient metabolic compensation and adequate renal function significantly decreases mortality [27]. The measure of hypercapnia, in contrast to acidosis, reflects a mechanical lung alteration only, without any metabolic interference.
Fourth, a retrospective analysis evaluating the outcomes of 347 hospitalised patients with ECOPD and categorised according to the Rome classification, found a higher mortality risk after 3 years pf follow-up when patients presented a severe grade (acidosis) or moderate grade (hypercapnia only) than those with mild grade (no hypercapnia). The study highlighted the prognostic role of hypercapnia, regardless of acidosis [11].
Finally, in a real-life scenario, very recent evidence indicates that the use of NIMV for ECOPD in patients with hypercapnia achieves an overall rate of success (need for ETI and/or all-cause in-hospital death) similar to that of patients with acidosis (23% vs 20%) [28]. The evidence about the role of the compensated hypercapnic patients should be updated.
What role may hypercapnia play in the chronicity of the disease?
In COPD patients, the presence of hypercapnia may be persistent, also after ECOPD [3]. A long-term study of patients with stable COPD showed shorter survival in patients with chronic hypercapnia than in those with normocapnia (5 vs 6.5 years) [29]; of note, measures of severe lung impairment, related to alveolar hypoventilation, such as PaCO2 (HR = 1.03; 95%CI 1.01 to 1.04) and of airflow limitation such as FEV1% of predicted (HR = 0.98; 95%CI 0.97 to 0.99), together with age (HR = 1.04; 95%CI 1.01 to 1.08) were considered independent prognostic risk factors [29]. Similarly, in our study the presence of hypercapnia may represent a sign of a more severe COPD, as confirmed by the significant differences in terms of airflow obstruction (FEV1/FVC), severity of airflow limitation (FEV1% predicted), and in the prevalence of patients requiring LTOT (Table 1). Therefore, a more severe COPD may have a worse prognosis. We report a higher mortality rate for hypercapnic patients, up to 3 years (Table 2), with hypercapnia or acidosis both increasing the mortality risk at 1 year (Table 3); interestingly, in our model, older age was a negative prognostic factor [29].
Although in a smaller cohort, a pilot study considering Asian ECOPD patients found a similar risk between hypercapnic and acidotic patients at admission in terms of life-threatening events and mortality risk [30]. However, in the regression model in that study PaCO2 was included as a continuous measure, rather than using a specific threshold (55 mmHg), which helps identify patients “at risk” for a hypothetical treatment, regardless of acidosis. Interestingly, in our regression model (Table 4), the presence of acidosis, if associated with a relatively small increase of PaCO2 (<55 mmHg), was not a prognosis risk factor (HR 1.97; 95%CI 0.69 to 5.61), suggesting again the greater role of the severe and chronic ventilatory impairment than of the intercurrent metabolic acidosis.
Can we therefore change the prognosis of hypercapnic patients, even if they do not have acidosis?
The domiciliary use of NIMV in hypercapnic COPD has been demonstrated in a recent meta-analysis to be associated with a lower risk of mortality than all-cause hospital admission, and lower need for ETI [31]. Although a prospective Dutch study reported that the improvement in hypercapnia raised survival in chronically ventilated patients [32], other reports indicate that the survival effect may be particularly evident when NIMV is targeted at reducing PaCO2 levels, indicating the effect of different ventilatory parameters on the efficacy of this treatment in COPD patients [33–35]. Moreover, a recent meta-analysis involving 1482 patients confirmed that the chronic use of NIMV significantly reduces mortality, especially in a subgroup analysis of COPD patients with higher baseline levels of PaCO2 (>55 mmHg)[36]. Similarly, our regression model (Table 4) reports a significant mortality risk in patients with a level of PaCO2 greater than 55 mmHg.
Significant differences in the change in PaCO2 were found in patients ventilated with inspiratory pressure levels of at least 18 cm H2O, in patients who used a ventilator for at least 5 h per night, and in patients with a baseline PaCO2 level of at least 55 mmHg[37]. Similar studies have confirmed the importance of setting ventilatory pressures at a high level, thereby improving arterial blood gas levels [34, 38]. To enhance the efficacy and tolerability of NIMV, it may be beneficial to provide strategies for optimising treatment.
Strengths and limitations
Major strengths of the study include the originality of our approach, the selection criteria used for enrolling patients, and the potential to propose a threshold for PaCO2, which helps identify patients with a worse prognosis. As for the limitations, we should mention the missing data on patients using domiciliary NIMV, which means that we cannot rule out the possibility that some hypercapnic patients at admission may have been chronically hypercapnic. However, this aspect may also be present in patients with acidosis and would then be diluted between the two study groups. Moreover, the prevalence of patients with LTOT as a sign of disease severity and a proxy for chronic need for NIMV may help address this issue, and appears similar between hypercapnia and acidosis.
In conclusion, we demonstrate that in hospitalised patients with ECOPD, hypercapnia at admission may be associated with a worse outcome regardless of the presence of acidosis. Considering a threshold of PaCO2 of 55 mmHg may help identify patients who have a worse prognosis and, therefore, are potentially eligible for NIMV in the acute setting, thereby expanding the indication of ERS guidelines and leading the way for the design of a specific study to demonstrate its efficacy. In this context, the severity and chronicity of ventilatory impairment might be more relevant than an intercurrent acidosis.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thank all collaborators involved in data collection and are grateful for the opportunity to conduct the current study.
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Giulia Sartori, Alberto Fantin, Filippo Sartori, and Albert Gabarrús. The first draft of the manuscript was written by Giulia Sartori, Alberto Fantin, Ernesto Crisafulli, and Antoni Torres. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This research did not receive any specific funding from agencies in the public, commercial, or not-for-profit sectors.
Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Conflict of interests
The authors declare that they have no competing interests regarding the contents reported in this manuscript.
Ethical approval and consent to participate
The Ethics Committee of the hospital (Hospital Clínic of Barcelona, Spain) approved the study protocol (CEIC 2008/4106). The study was conducted in accordance with good clinical practices and the Declaration of Helsinki. All enrolled patients provided informed consent.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Adeloye D, Song P, Zhu Y, Campbell H, Sheikh A, Rudan I, Unit NRGRH (2022) Global, regional, and national prevalence of, and risk factors for, chronic obstructive pulmonary disease (COPD) in 2019: a systematic review and modelling analysis. Lancet Respir Med 10:447–458. 10.1016/S2213-2600(21)00511-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Henrot P, Dupin I, Schilfarth P, Esteves P, Blervaque L, Zysman M, Gouzi F, Hayot M, Pomiès P, Berger P (2023) Main pathogenic mechanisms and recent advances in COPD peripheral skeletal muscle wasting. Int J Mol Sci. 10.3390/ijms24076454 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Csoma B, Vulpi MR, Dragonieri S, Bentley A, Felton T, Lázár Z, Bikov A (2022) Hypercapnia in COPD: causes, consequences, and therapy. J Clin Med. 10.3390/jcm11113180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dreher M, Neuzeret PC, Windisch W, Martens D, Hoheisel G, Gröschel A, Woehrle H, Fetsch T, Graml A, Köhnlein T (2019) Prevalence of chronic hypercapnia in severe chronic obstructive pulmonary disease: data from the HOmeVent registry. Int J Chron Obstruct Pulmon Dis 14:2377–2384. 10.2147/COPD.S222803 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Crisafulli E, Torres A (2019) Hospitalized acute exacerbations of chronic obstructive pulmonary disease: which patients may have a negative clinical outcome? Expert Rev Respir Med 13:691–693. 10.1080/17476348.2019.1635459 [DOI] [PubMed] [Google Scholar]
- 6.MacIntyre NR (2023) Acute hypercapnic respiratory failure in COPD. Respir Care 68:973–982. 10.4187/respcare.10560 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Roberts CM, Stone RA, Buckingham RJ, Pursey NA, Lowe D, NCOPDRaOPi Group (2011) Acidosis, non-invasive ventilation and mortality in hospitalised COPD exacerbations. Thorax 66:43–48. 10.1136/thx.2010.153114 [DOI] [PubMed] [Google Scholar]
- 8.Rochwerg B, Brochard L, Elliott MW, Hess D, Hill NS, Nava S, Navalesi P, Antonelli M, Brozek J, Conti G, Ferrer M, Guntupalli K, Jaber S, Keenan S, Mancebo J, Mehta S, Raoof S (2017) Official ERS/ATS clinical practice guidelines: noninvasive ventilation for acute respiratory failure. Eur Respir J. 10.1183/13993003.02426-2016 [DOI] [PubMed] [Google Scholar]
- 9.Ergan B, Oczkowski S, Rochwerg B, Carlucci A, Chatwin M, Clini E, Elliott M, Gonzalez-Bermejo J, Hart N, Lujan M, Nasilowski J, Nava S, Pepin JL, Pisani L, Storre JH, Wijkstra P, Tonia T, Boyd J, Scala R, Windisch W (2019) European Respiratory Society guidelines on long-term home non-invasive ventilation for management of COPD. Eur Respir J. 10.1183/13993003.01003-2019 [DOI] [PubMed] [Google Scholar]
- 10.Celli BR, Fabbri LM, Aaron SD, Agusti A, Brook R, Criner GJ, Franssen FME, Humbert M, Hurst JR, O’Donnell D, Pantoni L, Papi A, Rodriguez-Roisin R, Sethi S, Torres A, Vogelmeier CF, Wedzicha JA (2021) An updated definition and severity classification of chronic obstructive pulmonary disease exacerbations: the Rome proposal. Am J Respir Crit Care Med 204:1251–1258. 10.1164/rccm.202108-1819PP [DOI] [PubMed] [Google Scholar]
- 11.Crisafulli E, Sartori G, Huerta A, Gabarrús A, Fantin A, Soler N, Torres A (2023) Association between Rome classification among hospitalized patients with COPD exacerbations and short-term and intermediate-term outcomes. Chest 164:1422–1433. 10.1016/j.chest.2023.07.021 [DOI] [PubMed] [Google Scholar]
- 12.Agustí A, Celli BR, Criner GJ, Halpin D, Anzueto A, Barnes P, Bourbeau J, Han MK, Martinez FJ, Montes de Oca M, Mortimer K, Papi A, Pavord I, Roche N, Salvi S, Sin DD, Singh D, Stockley R, López Varela MV et al (2023) Global initiative for chronic obstructive lung disease 2023 report: GOLD executive summary. Eur Respir J. 10.1183/13993003.00239-202337770089 [Google Scholar]
- 13.Crisafulli E, Barbeta E, Ielpo A, Torres A (2018) Management of severe acute exacerbations of COPD: an updated narrative review. Multidiscip Respir Med 13:36. 10.1186/s40248-018-0149-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Crisafulli E, Manco A, Ferrer M, Huerta A, Micheletto C, Girelli D, Clini E, Torres A (2020) Pneumonic versus nonpneumonic exacerbations of chronic obstructive pulmonary disease. Semin Respir Crit Care Med 41:817–829. 10.1055/s-0040-1702196 [DOI] [PubMed] [Google Scholar]
- 15.Freeman GH, Halton JH (1951) Note on an exact treatment of contingency, goodness of fit and other problems of significance. Biometrika 38:141–149 [PubMed] [Google Scholar]
- 16.Miké V, Stanley KE (1982) Statistics in medical research: methods and issues, with applications in cancer research. Wiley, New York [Google Scholar]
- 17.Collett D (2003) Modelling survival data in medical research. Chapman & Hall/CRC, Boca Raton [Google Scholar]
- 18.Grünewaldt A, Fritsch N, Rohde G (2022) Hypercapnic failure in acute exacerbated COPD patients: severe airflow limitation as an early warning signal. J Clin Med. 10.3390/jcm12010258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Keenan SP, Powers CE, McCormack DG (2005) Noninvasive positive-pressure ventilation in patients with milder chronic obstructive pulmonary disease exacerbations: a randomized controlled trial. Respir Care 50:610–616 [PubMed] [Google Scholar]
- 20.Luo Z, Cao Z, Li Y, Jin J, Sun W, Zhu J, Zhao N, Liu J, Wei B, Hu Y, Zhang Y, Ma Y, Wang C (2022) Physiological effects of high-intensity versus low-intensity noninvasive positive pressure ventilation in patients with acute exacerbation of chronic obstructive pulmonary disease: a randomised controlled trial. Ann Intensive Care 12:41. 10.1186/s13613-022-01018-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Luo Z, Li Y, Li W, Nie Q, Shi Y, Wang J, Ji Q, Han X, Liu S, Li D, Wang S, Li Z, Jia D, Ge H, Xu P, Feng Z, Li F, An F, Tai N, Yue L, Xie H, Jin X, Liu H, Dang Q, Zhang Y, Sun L, Wang J, Huang H, Chen L, Ma Y, Cao Z, Wang C, Fang W, Zheng D, Zhou X, Liu X, Liu Y, Liu L, Yang C, Xu Z, Lin R, Yang J, Peng Y, Li J, Yu C, Ling X, Ming S, Xu X, Hou X, Xue X, Ma H, Shen Y, Yang X, An Y, Mo T, Sang X, Shi H, Zhang H, Zhang P, Chen X, Han L, Wang Z, Guo Y, Xiang Yi, Feng J, Qiao H, Chen F, Wu C, Abula A, Tong J, Qin Y (2024) Effect of high-intensity vs low-intensity noninvasive positive pressure ventilation on the need for endotracheal intubation in patients with an acute exacerbation of chronic obstructive pulmonary disease: the HAPPEN randomized clinical trial. JAMA 332:1709–1722. 10.1001/jama.2024.15815 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Shahid H, Ahmed M, Sami SZ, Kumar S, Farooq J, Asif I, Hanif ZM, Nazir A (2025) High-intensity versus low-intensity non-invasive ventilation in acute exacerbations of chronic obstructive pulmonary disease: a systematic review and meta-analysis. Respir Med 247:108277. 10.1016/j.rmed.2025.108277 [DOI] [PubMed] [Google Scholar]
- 23.Collaborative Research Group of Noninvasive Mechanical Ventilation for Chronic Obstructive Pulmonary Disease (2005) Early use of non-invasive positive pressure ventilation for acute exacerbations of chronic obstructive pulmonary disease: a multicentre randomized controlled trial. Chin Med J (Engl) 118:2034–2040 [PubMed] [Google Scholar]
- 24.Bruno CM, Valenti M (2012) Acid-base disorders in patients with chronic obstructive pulmonary disease: a pathophysiological review. J Biomed Biotechnol. 10.1155/2012/915150 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wan X, Chen D, Tan Y, Ma M, Zhang F, Liu Z, Chen Y, Shao W, Cao C (2020) Incidence, risk factors, and prognostic implications of acute kidney injury in patients with acute exacerbation of COPD. Int J Chron Obstruct Pulmon Dis 15:1085–1092. 10.2147/COPD.S238343 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Marcy F, Goettfried K, Enghard P, Piper SK, Kunz JV, Schroeder T (2024) Impact of AKI on metabolic compensation for respiratory acidosis in ICU patients with AECOPD. J Crit Care 83:154846. 10.1016/j.jcrc.2024.154846 [DOI] [PubMed] [Google Scholar]
- 27.Ucgun I, Oztuna F, Dagli CE, Yildirim H, Bal C (2008) Relationship of metabolic alkalosis, azotemia and morbidity in patients with chronic obstructive pulmonary disease and hypercapnia. Respiration 76:270–274. 10.1159/000131707 [DOI] [PubMed] [Google Scholar]
- 28.Pisani L, Corsi G, Carpano M, Giancotti G, Vega ML, Catalanotti V, Nava S (2023) Clinical outcomes according to timing to non invasive ventilation initiation in COPD patients with acute respiratory failure: a retrospective cohort study. J Clin Med 12(18):5973. 10.3390/jcm12185973 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Yang H, Xiang P, Zhang E, Guo W, Shi Y, Zhang S, Tong Z (2015) Is hypercapnia associated with poor prognosis in chronic obstructive pulmonary disease? A long-term follow-up cohort study. BMJ Open 5:e008909. 10.1136/bmjopen-2015-008909 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lun CT, Tsui MS, Cheng SL, Chan VL, Leung WS, Cheung AP, Chu CM (2016) Differences in baseline factors and survival between normocapnia, compensated respiratory acidosis and decompensated respiratory acidosis in COPD exacerbation: a pilot study. Respirology 21:128–136. 10.1111/resp.12652 [DOI] [PubMed] [Google Scholar]
- 31.Wilson ME, Dobler CC, Morrow AS, Beuschel B, Alsawas M, Benkhadra R, Seisa M, Mittal A, Sanchez M, Daraz L, Holets S, Murad MH, Wang Z (2020) Association of home noninvasive positive pressure ventilation with clinical outcomes in chronic obstructive pulmonary disease: a systematic review and meta-analysis. JAMA 323:455–465. 10.1001/jama.2019.22343 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Raveling T, Bladder G, Vonk JM, Nieuwenhuis JA, Verdonk-Struik FM, Wijkstra PJ, Duiverman ML (2018) Improvement in hypercapnia does not predict survival in COPD patients on chronic noninvasive ventilation. Int J Chron Obstruct Pulmon Dis 13:3625–3634. 10.2147/COPD.S169951 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Liao H, Pei W, Li H, Luo Y, Wang K, Li R, Xu L, Chen X (2017) Efficacy of long-term noninvasive positive pressure ventilation in stable hypercapnic COPD patients with respiratory failure: a meta-analysis of randomized controlled trials. Int J Chron Obstruct Pulmon Dis 12:2977–2985. 10.2147/COPD.S148422 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Köhnlein T, Windisch W, Köhler D, Drabik A, Geiseler J, Hartl S, Karg O, Laier-Groeneveld G, Nava S, Schönhofer B, Schucher B, Wegscheider K, Criée CP, Welte T (2014) Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial. Lancet Respir Med 2:698–705. 10.1016/S2213-2600(14)70153-5 [DOI] [PubMed] [Google Scholar]
- 35.Zhou L, Li X, Guan L, Chen J, Guo B, Wu W, Huo Y, Zhou Z, Liang Z, Zhou Y, Tan J, Chen X, Song Y, Chen R (2017) Home noninvasive positive pressure ventilation with built-in software in stable hypercapnic COPD: a short-term prospective, multicenter, randomized, controlled trial. Int J Chron Obstruct Pulmon Dis 12:1279–1286. 10.2147/COPD.S127540 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Wu Z, Luo Z, Ge J, Jin J, Cao Z, Ma Y (2022) Baseline level and reduction in PaCO2 are associated with the treatment effect of long-term home noninvasive positive pressure ventilation in stable hypercapnic patients with COPD: a systematic review and meta-analysis of randomized controlled trials. Int J Chron Obstruct Pulmon Dis 17:719–733. 10.2147/COPD.S344962 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Struik FM, Lacasse Y, Goldstein RS, Kerstjens HA, Wijkstra PJ (2014) Nocturnal noninvasive positive pressure ventilation in stable COPD: a systematic review and individual patient data meta-analysis. Respir Med 108:329–337. 10.1016/j.rmed.2013.10.007 [DOI] [PubMed] [Google Scholar]
- 38.Struik FM, Sprooten RT, Kerstjens HA, Bladder G, Zijnen M, Asin J, Cobben NA, Vonk JM, Wijkstra PJ (2014) Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study. Thorax 69:826–834. 10.1136/thoraxjnl-2014-205126 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.