Health-related quality of life and long-term prognosis in chronic hypercapnic respiratory failure: a prospective survival analysis

Background: Health-related quality of life (HRQL) is considered as
an important outcome parameter in patients with chronic diseases.
This study aimed to assess the role of disease-specific HRQL for
long-term survival in patients of different diagnoses with chronic
hypercapnic respiratory failure (CHRF). Methods: In a cohort of 231
stable patients (chronic obstructive pulmonary disease (COPD), n =
98; non-COPD (obesity-hypoventilation syndrome, restrictive
disorders, neuromuscular disorders), n = 133) with CHRF and current
home mechanical ventilation (HMV), HRQL was assessed by the
disease-specific Severe Respiratory Insufficiency (SRI)
questionnaire and its prognostic value was prospectively evaluated
during a follow-up of 2-4 years, using univariate and multivariate
regression analysis. Results: HRQL was more impaired in COPD (mean
+/- SD SRI-summary score (SRI-SS) 52.5 +/- 15.6) than non-COPD
patients (67.6 +/- 16.4; p < 0.001). Overall mortality during
28.9 +/- 8.8 months of follow-up was 19.1% (31.6% in COPD, 9.8% in
non-COPD). To identify the overall role of SRI, we first evaluated
the total study population. SRI-SS and its subdomains (except
attendance symptoms and sleep), as well as body mass index (BMI),
leukocyte number and spirometric indices were associated with
long-term survival (p < 0.01 each). Of these, SRI-SS, leukocytes
and forced expiratory volume in I s (FEV(I)) turned out to be
independent predictors (p < 0.05 each). More specifically, in
non-COPD patients SRI-SS and most of its subdomains, as well as
leukocyte number, were related to survival (p < 0.05), whereas
in patients with COPD only BMI and lung function but not SRI were
predictive. Conclusion: In patients with CHRF and HMV, the
disease-specific SRI was an overall predictor of long-term survival
in addition to established risk factors. However, the SRI
predominantly beared information regarding long-term survival in
non-COPD patients, while in COPD patients objective measures of the
disease state were superior. This on one hand highlights the
significance of HRQL in the long-term course of patients with CHRF,
on the other hand it suggests that the predictive value of HRQL
depends on the underlying disease.