Obesity is the most prevalent metabolic disease in the Western world. It is associated with high rates of morbidity and mortality, and has been recognized by the World Health Organization as a serious health, social and economic problem.1 It produces significant changes in the physiology of the respiratory system that can give rise to a wide spectrum of clinical manifestations, ranging from secondary dyspnea and restrictive ventilatory defect to hypercapnic respiratory failure characteristic of obesity-hypoventilation syndrome (OHS).2–4

OHS was first described in 19565 and is currently defined as a condition characterized by obesity, hypercapnia and respiratory changes during sleep, in the absence of any other disease that could explain respiratory failure.6 Despite the passage of time, the pathogenesis of this entity has still not been fully explained, although it seems to have a multifactorial origin in which altered ventilatory mechanics, muscle dysfunction secondary to obesity and ventilation control play an important role.7,8 Many authors consider it as simply a progressive form of the obstructive sleep apnea–hypopnea syndrome (OSAHS) in some obese subjects.9 However, although respiratory disorders during sleep are a defining condition for OHS, they are not always associated with OSAHS10 and it remains unknown which subgroup will progress the best. The prevalence of OHS in the general obese population is not precisely defined, but in obese patients with OSAHS it ranges between 10% and 20%.11,12 Studies in hospitalized obese patients have shown greater prevalence, also revealing increased mortality rates in this group.13

Since its introduction for the treatment of OSAHS, continuous positive airway pressure (CPAP) has also become a treatment for OHS, with non-invasive mechanical ventilation (NIV) being reserved for cases that do not respond adequately to positive pressure.14–16 Numerous groups initially rejected the use of NIV as standard treatment for OHS, but it was shown that in this group of patients similar results were obtained to those observed in patients with respiratory failure due to chest wall dysfunction,17 both in acute situations or in long-term home use.18 NIV to treat night-time OHS produces clinical and functional improvement with favorable changes in blood gases, polyglobulia, respiratory function and ventilatory response to hypercapnia. These changes lead to a reduction in the number and duration of hospital admissions and visits to emergency rooms, and an improvement in quality of life and survival.19,20 However, few studies have evaluated long-term survival, and behavioral differences between patients with and without OSAHS have not been analyzed, even though it has been shown that the latter can be treated safely with CPAP.21

The aim of this study was to determine the differences in progress between patients with OHS and those with OHS and OSAHS receiving home mechanical ventilation (HV), and to identify prognostic factors for survival.

In this study, HV was found to be effective in the treatment of patients with OHS. It was associated with blood gas and functional improvement and with increased survival. Results are better in subjects with OHS–OSAHS, although the differences between these 2 groups are not significant and the only predictive factor for survival was good FVC at outset.

Historically, ventilatory support for patients with OHS was first used with concomitant OSAHS. CPAP was used initially and, if this was insufficient to control hypoventilation, positive pressure ventilation was applied using either bilevel pressure or volume-controlled ventilators.14–16 Since then, long-term NIV results have been analyzed in numerous cohort studies of OHS patients,18,24–27 all of which have reported improvement in blood gas levels, and OHS is now among the indications given in guidelines and recommendations for the use of HV.17 To date, only one clinical trial has been published in which the results of NIV are compared with conventional oxygen therapy. This is a small randomized study in 36 patients with very mild hypercapnia and a follow-up period of only one month. However, despite these limitations, improvement in sleep structure and gas exchange were greater than in the control group.28

All NIV cohort studies in patients with OSH show an improvement in gas exchange. This was also found in this study, in which correction of hypercapnia was seen from the beginning of treatment, an improvement that persisted for as long as 10 years in both OSH and OSH-OSAHS patients. This improvement has been attributed to multiple mechanisms, including recovery of sensitivity to hypercapnia, respiratory muscle rest and improvement in ventilatory mechanics.8,19,20

In patients with OHS, the use of NIV is associated with an increase in respiratory minute volume (VE) and occlusion pressure in response to hypercapnia.24,29 A previous study performed by our group in a small series of patients also found that the use of NIV in patients with OHS was associated with an increase in the VE/PaCO2 and P0.1/PaCO2 slope,24 but these variables were not available in most of the patients in the cohort analyzed here, so these findings cannot be confirmed. Changes in PaO2, in addition to reflecting variations in ventilation, may also be related to an improvement in the level of pulmonary hypertension and right ventricular overload that frequently develop in these patients, and, according to some recent studies, may be reversed with the use of HV.30,31 Two important aspects to remember when considering the improvement in blood gas levels are good ventilatory support during administration and adherence to treatment. Our working procedures include nocturnal arterial blood gas monitoring for evaluating the efficacy of the ventilation, so all patients were subject to nocturnal monitoring, both at the start of ventilation and during regular check-ups, the results of which were used to adjust the ventilator parameters. Treatment adherence is known to be a prerequisite for normalizing blood gas levels, with reports indicating an association between the number of hours of ventilation and the degree of PaO2 reduction, and only around 4.5h of ventilation can achieve significant blood gas level improvements.32 The patients in this study received more than 4h of ventilatory support, a factor that also contributed to the good results obtained.

With regard to changes in ventilatory mechanics, the literature is contradictory, and although the study published by Heinemann et al.26 showed that NIV in OSH patients was associated with an increase in inspiratory capacity and total lung capacity, this has not been confirmed by other authors.20,25 Although lung volumes were not measured in this study, a sustained increase in FVC was found. Other factors cannot be ruled out, but similarly to Heinemann et al., high support pressures were used in this study, which could have assisted in resolving possible microatelectasis, thus improving respiratory mechanics. Muscle rest provided by NIV may also contribute to the improvement in lung volumes,33 but since pressures were not measured in this study no conclusions could be drawn in this regard.

The most important aspect of this study is the impact of HV on mortality. Few studies have analyzed the survival of patients with OHS treated with HV for periods longer than 5 years.28,34 Our results confirm the findings of those studies, with survival rates of over 90% in the first year and 70% in the fifth. We also had good results for 8- and 10-year survival, data that has not yet been reported by other researchers. Unlike other authors,27,34 we did not find that the degree of hypoxemia or hypercapnia or pH levels had a prognostic value for patient survival, but the difference is probably due to the fact that our patients were very closely assessed before inclusion in an HV program, i.e., their situation was already stable.

None of the previous studies analyzed the impact of OSAHS associated with OHS on patient survival. In our study, after the first year of ventilation, patients with OSAHS showed a clear trend toward improved survival, although this was not statistically significant. Since patients with OHS initially had a higher level of hypoventilation than those with OHS–OSAHS, correction of airway instability in the latter may have a more favorable effect on gas exchange, but differences in other fundamental physiological aspects cannot be ruled out.

One important and hitherto unreported factor analyzed is the impact of changes in ventilatory mechanics on the clinical course. In this study, the only factor with predictive value for survival was FVC: the lower the FVC, the less chance of survival. It can be assumed that in patients with more compromised ventilation at the time of diagnosis, gas exchange alterations would already have caused irreversible structural changes that would explain their poorer progress, even after an initial improvement in blood gas levels. Moreover, as data on the cause of death are not available, we cannot rule out that patients with greater restriction did not also have greater associated comorbidity.

Interestingly, although most of the OHS patients in the series published to date also had OSAHS, these subjects were a minority in our study. This may be because our population had a greater level of hypercapnia that those of other studies, and patients were referred for respiratory failure and not because of suspected sleep apnea.

One of the most important limitations of this study is the lack of data related to the impact of comorbidity on the clinical course and survival of patients, since, although follow-up was prospective, this information was not included when the statistical analysis plan was designed. Nor is it possible to analyze which type of ventilator would be more effective or, more importantly, whether some patients could have been treated from the outset with CPAP instead of NIV.

Although some authors claim that in extremely obese patients the use of volume-cycled ventilators may be necessary, all published series have reported satisfactory results with bilevel ventilators with backup respiratory rate for avoiding central apnea events35; the efficacy of this technology is supported by our results. It has been suggested recently that the use of bilevel volume-assured pressure support may have additional advantages. However, the available data are contradictory and although 2 clinical trials aimed at evaluating this technology have been performed, very few patients were included and results cannot be considered conclusive.36,37 As for the question of whether patients with OSAHS could have been effectively treated with CPAP, all these patients were included in the HV program according to protocol, so this aspect could not be analyzed. What was observed was that, after correction of initial hypoventilation, some patients could be switched to CPAP; this finding has already been published by our group in a previous study38 and was also reported in the series published by Pérez de Llano et al.21 The problem is that, although treatment algorithms recommend starting CPAP therapy in stable, acidosis-free patients, in most studies treatment is initiated with BiPAP, whether in the acute or stable phase. Only after blood gas levels have improved is a switch to CPAP considered in patients with OSAHS. To date, only one short-term randomized clinical trial in a limited number of patients with an inclusion criterion of only mild PaO2 and PaCO2 alteration has been published. The authors conclude that both treatments are equally effective.39 Enough data is currently available to support the recommendation that patients with mild gas exchange alterations associated with moderate or severe OSAHS begin treatment with CPAP, while BiPAP would be the treatment of choice for patients with clearly predominant nocturnal hypoventilation.4,8,20,40 Additional clinical trials are necessary to position one or the other technique.

To conclude, our results confirm that the use of HV in OHS patients with or without OSAHS is very effective for correcting gas exchange alterations and improving ventilatory mechanics, and survival rates can exceed those reported in other populations classically considered candidates for this therapy. Two questions, however, remain: which is the ideal patient population for this treatment, and in which patients can CPAP from outset assure the same level of safety and efficacy, assuming clinical stability in both cases.

The authors state that they have no conflict of interests.