The data obtained indicates that normoxic patients with COPD who desaturate within the first minute of starting the 6MWT have a greater risk of mortality compared to NED or ND patients. Survival in the latter two groups tends to overlap.
This conclusion coincides, in part, with the results of previous studies (3,5,6) which found that exercise-induced desaturation (EID) was a mortality risk factor in patients with COPD. However, we have added another mortality risk factor, early desaturation, which has an evidently worse prognosis.
We initially carried out a comparative analysis between desaturators and non-desaturators as well and also found a significant difference, with greater mortality in desaturators. Our objective, however, was to analyze survival in ED patients and upon analyzing the survival curves, we found accumulated survival in years and saw that the ED group had reduced survival (Figure 2), with the estimated survival time being significantly lower in this group of patients with respect to the NED and ND groups (5.9, 7.5, and 9.6 years, respectively). The mortality risk for the ED group was comparatively 3.5 times higher than the ND group (Table 2). We found a marked difference in mortality in the group of patients with early desaturation, 73% compared to 38% in the non-early desaturators group. This percentage is nearly double and translates to an almost 4-year shorter survival among ED patients. In light of these results, the greatest mortality among desaturators seems to be in the early desaturators group, a finding that has not been evaluated in previous studies.
When we included factors modifying mortality risk (age, BMI and CO diffusion) (Table 3) in the multivariate analysis, early desaturation continued to show higher risk compared to the ND group (HR 2.5; CI 95% 1.6-3.8). No difference in risk was found in patients classified as NED compared to the ND patients that we used as a reference group (HR 1.03; CI 95% 0.62-1.69). The multivariate analysis maintains observed significance for early desaturation.
It is important to remember that hypoxemia observed in patients with COPD is caused by alterations in the ventilation/perfusion relationship as a result of airflow obstruction and the presence of emphysema, and it can be exacerbated during sleep and exercise. Chronic hypoxemia can result in the development of pulmonary hypertension and polycythemia, can increase underlying systemic inflammation and can also favor musculoskeletal dysfunction, which will impact quality of life and exercise tolerance, as well as increase the risk of cardiovascular events and death (7).
We also know that the 6MWT is a submaximal effort test, which evaluates functional exercise capacity and which is routinely used in managing patients with COPD. Unlike other lung function tests, it provides a global view of the patient’s exercise tolerance and does not require costly apparatuses like cardiopulmonary exercise testing (8). At the same time, it is the most sensitive test to detect oxygen desaturation in patients with COPD (9).
We used the 6MWT in our study in order to show that the hypoxemia that appears with exercise can have different consequences depending on how long it takes to appear. Although desaturators are believed to have greater alterations in gas exchange in the ventilation/perfusion relationship and in diffusion, other physiopathological phenomena can occur in early desaturators that would explain this greater repercussion.
After discovering the EID phenomenon in patients with COPD, some authors described predictive factors such as decreased DLCO and SpO2 (10,11,12,13), a low FEV1 or PaO2 (14,15) with different cut-off points (16), and associations with different physiopathological phenomena such as yearly FEV1 decline, dynamic hyperinflation, desaturation during sleep, and scores like the BODE index (17,18,19,20,21).
The objective of this study was not to analyze predictive factors for desaturation nor clinical situations associated with desaturation in the ED and NED groups. However, given the mortality observed in the ED group, there is a possibility of finding another predictive factor, as well as other associated clinical situations.
In a previous study conducted by our group in 67 patients with COPD concerning the association between EID and desaturation that occurs during daily activities using 24-hour pulse oximetry, we found that patients who desaturated 3.30 minutes or more after starting the walk test had a 100% probability of not desaturating during daily activities, while those who desaturated within the first minute also desaturated during their daily activities. We referred to the latter group as early desaturators (22).
However, the first studies that analyzed the 6MWT in patients with COPD placed more value on the distance walked than on oxygen desaturation during the test (23, 24), although some later studies revealed that both oxygen desaturation and distance walked were prognostic factors (25). This led some authors to combine both factors in their analyses (26, 27,28).
We also observed that ED patients had a greater need for long-term home oxygen than the group of NED patients at 5 years of follow-up (29). This reveals the importance of the ED patient group, which was associated with desaturation during daily activities and was also predictive of chronic respiratory distress or a need for long-term home oxygen therapy.
On the other hand, recent studies have found that EID was a risk factor for mortality and exacerbation (3), was associated with a greater degree of emphysema in CAT (computerized axial tomography) (30), and these patients had a 2.4-2.7 times greater risk of death (5,6,31) and a higher prevalence of atrial fibrillation (32) compared to non-desaturators.
In our study, we did not record exacerbations but did register comorbidity using the Charlson index, without finding significant differences between the groups analyzed. However, we did find that obesity was associated with early desaturation in another recent study (33). We did not expect to find such a large increase in mortality in the ED group and such a discreet increase in the NED group. The fact that early and not late desaturation is associated with desaturation during daily activities could likely explain this striking increase in mortality. The presence of intermittent hypoxia sustained over time, which potentially induces the physiopathological effects of chronic hypoxemia, as well as the more sedentary lifestyle of these patients and the associated nighttime hypoxemia (21) can potentially explain this increase found in mortality. This, along with the fact that survival curves overlapped for the NED and ND groups in the analysis, leads us to conclude that early desaturators have greatest mortality among desaturators, with a hazard ratio of 2.5 in the multivariate analysis compared to non-desaturators (Fig. 2 and 3).
We can note that in the walk test, it is important not only to record the meters walked, but also the time it takes to desaturate, in order to determine whether or not patients desaturate early, since late desaturation has a lower mortality that is similar to that observed in non-desaturators.
Furthermore, these observations suggest that many studies in patients with COPD that have analyzed EID without taking the time until desaturation into account could have led to misguided conclusions with respect to mortality or the response to different treatments. Therefore, in survival studies following oxygen therapy in patients with COPD and exercise-induced desaturation, it should be determined whether or not they are early desaturators, since the response to oxygen therapy could also be different and improve survival.
Our study has some limitations: the total number of patients is not very high (319), the percentage of women is not relevant, and it has focused on analyzing mortality and not other factors that can also be important (such as a decline in respiratory function or risk of exacerbations). However, it does emphasize early desaturation as a phenomenon to keep in mind as a prognostic factor. Future multicenter studies with a large sample size are needed to confirm these findings and also analyze the possible response to oxygen therapy in these patients.