Main findings
We present an observational study of 105 Marfan patients with a median 7.76-year follow up period from baseline. It is the first study focusing on the primary endpoint cardiovascular death in relation to sleep apnea in this rare disease. Our results demonstrate that sleep apnea is associated with an increased risk for cardiovascular death, independent of the patients’ BMI or age. There was no effect on the secondary endpoints 1) aortic events in general, 2) acute aortic events, or 3) the proximal or distal aortic growth rate, probably due to the high number of patients with prior aortic surgery.
Sleep apnea and mortality
Our study establishes an association of sleep apnea with cardiovascular mortality in Marfan syndrome. There may be several explanation for this finding.
First, sleep apnea in general, leads to an increased risk of cardiovascular disease, including difficult-to-control blood pressure, coronary artery disease, congestive heart failure, arrhythmias and stroke (11). It is known that patients with sleep apnea have a higher prevalence of atrial and ventricular arrhythmia compared to the general population (21). Additionally, an increased risk of cardiovascular mortality was reported in patients with severe sleep apnea (22). The negative effects of sleep apnea on the cardiovascular system in MFS are probably similar to non-Marfan patients, and a reasonable explanation for the increased risk for cardiovascular death in Marfan patients with sleep apnea. In a cross-sectional study, Muiño-Mosquera et al. confirmed that Marfan patients with sleep apnea tend to have higher systolic blood pressures, larger distal aortic diameters and a higher prevalence of ventricular arrhythmias (7). These differences were, however, not significant after adjusting for confounders (7).
Second, MFS is not only an aortic disease, but has several manifestations with possible risks for cardiovascular death, including ventricular arrhythmias and myocardial involvement (10). Due to great achievements in medical and surgical diagnostics and therapy (including PPPM strategy), death from other cardiovascular causes, excluding aortic events, have become more visible and treatable in the last years (23). Our results are in line with these findings, showing 20% deaths of cardiovascular causes beside aortic events.
In our study, we observed high rates of mild sleep apnea, which may indicate an early state of sleep apnea in this rather young population. Yet, we were able to show that sleep apnea is a significant risk factor for cardiovascular death. Even in case of an increase of AHI from 5 to 15/h, there is an 22% increase in risk for cardiovascular death. This novel finding demonstrates the relevance of sleep apnea screening in this group of patients as an important adjunct to the management of Marfan syndrome. Based on these results, further studies are needed to evaluate different therapy options for sleep apnea in this group of patients, preferably as multicenter trials.
Sleep apnea and aortic events
The results of our study did not show a significant effect of sleep apnea or the AHI-score on aortic events or the proximal or distal aortic growth rate in patients with Marfan syndrome. Interestingly, further analyses of the subgroups of patients with and without prior aortic surgery showed, that there was a weak effect of the AHI-score on aortic events in patients without prior surgery. Due to the low number of patients in the subgroup, we did not reach the level of significance. In contrast to this subgroup, there was absolutely no effect in patients with prior aortic surgery. Therefore, we believe that the high number of patients with prior aortic surgery (42%) might have influenced the aortic event rate during follow-up and therefore be one explanation for the non-significant effect of the AHI-score on the secondary outcome.
Limited data is published regarding this important topic. Sowho et al. showed that a high risk for sleep apnea (detected by a composite survey score) was associated with aortic enlargement and a threefold increased risk of having prior aortic root replacement in patients with Marfan syndrome (24). Kohler et al. compared Marfan patients with and without sleep apnea and observed a significantly shorter aortic event free survival in sleep apnea patients (9). They showed an association between the AHI-score and aortic events independent of the patients’ BMI. However, this association was no longer significant after adjusting for further covariates like age, gender, baseline aortic diameter, systolic blood pressure and antihypertensive medication (9).
Regarding the general population beside Marfan syndrome, several studies indicate that sleep apnea alone elevates the risk for aortic dissection, aortic dilatation and aortic rupture (25–27), whereas others did not confirm these findings (28). It was also observed that the duration of an oxygen saturation < 90% influenced the sizes of the ascending aorta and the main pulmonary artery, showing greater dimensions in patients with sleep apnea (29).
Summing up, whether sleep apnea leads to an increased risk for aortic events in MFS remains unclear and needs further investigation in larger cohorts, preferably in patients without prior aortic surgery or intervention.
Predictive, preventive, and personalized medicine
The PPPM proposes, implements, and supports the need of paradigm shift from reactive medical services to predictive, preventive, and personalized medicine concepts of health (30). Care of patients with Marfan syndrome has been roughly in line with this concept for years, as it aims to prevent patients from life threatening complications. Additional to guideline recommendations for preventive surgery, treatment of patients is a personalized concept, which includes living conditions and family planning of the individual. Due to the autosomal dominant heredity, PPPM is not limited to the individual patient but includes the whole family (30).
Our results demonstrate sleep apnea to be predictive for cardiovascular death. Therefore, early sleep apnea screening may serve as an additional predictive diagnostic measure in a complex and multifactorial setting of this inherited connective tissue disease. Further studies are needed to evaluate treatment of sleep apnea with regard to secondary and tertiary prevention of cardiovascular complication in this rare disease.
Limitations
Few aspects may weaken reproducibility in other populations. First, our study was observational with varying observation times across individuals. Second, besides age and BMI, other confounders may exist. Third, sleep apnea screening was offered to all patients, regardless of symptoms. Nevertheless, several patients did not participate, which may have excluded an important proportion of patients leading to selection bias of the study cohort. Additionally, the power of the comparisons was low due to the small number of patients and the low number of events. The effects on the secondary endpoints might have been clearer in a larger group of patients. However, the effect of sleep apnea on the primary endpoint cardiovascular mortality was seen clearly.
Interpretation and outlook
Sleep apnea shows to be independently predictive of cardiovascular death in patients with Marfan syndrome. Even with only mild sleep apnea the impact on mortality is significant. Therefore, early screening of sleep apnea may be an important adjunct to the PPPM management of Marfan syndrome.
Take-Home Points
Research Question: To examine the effect of sleep apnea on cardiovascular death and on aortic events in patients with Marfan syndrome.
Results: Sleep apnea shows to be independently predictive of cardiovascular death in patients with Marfan syndrome.
Interpretation: These findings highlight the necessity of early sleep apnea screening in MFS patients as an important adjunct to the PPPM management of the disease.