Atrial septal aneurysm, shunt size, and recurrent stroke risk in patients with patent foramen ovale
Selected in JACC by L. Biasco
This work aimed at assessing the respective influence of interatrial shunt size and presence of ASA on the risk of recurrent stroke in medically treated patients after an index cryptogenic ischemic stroke in patients with PFO without evidence of alternative causes.
References
Authors
Turc G, Lee JY, Brochet E, Kim JS, Song JK, Mas JL; CLOSE and DEFENSE-PFO Trial Investigators.
Reference
J Am Coll Cardiol. 2020 May 12;75(18):2312-2320.
Published
May 2020
Link
Read the abstractReviewer
My Comment
Why this study? – the rationale/objective
Percutaneous closure of a Patent Forame Ovale (PFO) is nowadays indicated for secondary prevention of cryptogenic ischemic stroke/TIA and systemic embolism in selected young patients in whom the clinical event has a high probability of being associated to a paradoxical embolism.
While catching a thrombus crossing the PFO is rare, this mechanism is hypothesised after exclusion of other possible causes. Traditionally, the presence of several anatomical characteristics such as PFO tunnel length and width, presence of an Atrial Septal Aneurysm (ASA) or evidence of large shunt evaluated at trans-esophageal echocardiography (TEE) have been regarded as high-risk features, giving strength to the possibility of a paradoxical ischemic event.
When encountered after a cryptogenic stroke, those characteristics were used to support a higher likelihood of right to left shunt thus justifying the mechanistic role of PFO. In particular, the presence of ASA and of a large shunt have been specified as inclusion criteria in more recent randomised trials, probably also impacting on the shown positive effect of interventional treatment. Nonetheless, due to the presumptive nature of a paradoxical embolic event and the low risk of long term recurrences, choosing wisely remains challenging.
When pooling together data derived from different trials, the number needed to treat to prevent 1 recurrent stroke at 1 year after percutaneous PFO closure remains well above 100 (NNT 131; 95% CI, 101–400 according to Turc et al J Am Heart Assoc. 2018). Given these premises, criteria to define patients at higher risk that might more effectively benefit from a percutaneous approach are obviously needed.
The aim of this work was to assess the respective influence of interatrial shunt size and presence of ASA on the risk of recurrent stroke in medically treated patients after an index cryptogenic ischemic stroke in patients with PFO without evidence of alternative causes.
How was it executed? – the methodology
Pooled analysis of 898 individual patient data derived from medically treated patients with cryptogenic stroke and associated PFO enrolled in 2 observational studies (Mas JL at al. NEJM 2001 and Lee JY et al. Am Heart J 2010) and medical treatment arms of 2 randomized clinical trials (CLOSE and DEFENSE-PFO) performed by the same groups in France and South Korea. Primary outcome was set as time to recurrent ischemic stroke.
All patients were evaluated with a standardized TEE protocol and classified into 4 categories according to the presence/absence of a large shunt (defined as passage of more than 30 microbubbles in the left atrium within 3 cardiac cycles after opacification of the right atrium) or ASA (defined as an excursion of the septum primum ≥ 10 mm).
After evaluating the association between baseline variables and recurrent ischemic stroke at univariable analysis, the impact of anatomical characteristics to time to recurrent stroke was assessed by multivariable analysis through a mixed-effects Cox regression approach. Then, the individual impact of shunt size and presence of ASA, after correcting for clinical characteristics, was also evaluated.
What is the main result?
Authors analysed data derived from 898 patients with a mean age of 45,3±11,8 years and median ROPE score of 7. After a mean follow up of 3.8 years, 47 (5,2%) recurrent strokes were observed leading to an incidence rate per 100 person-years of 1.3, 95% CI: 1.0 to 1.7. When stratified according to each anatomical feature, the incidence rate of recurrent events per 100 person-years were:
- ASA with large PFO: 2.4 (95% CI: 1.6 to 3.8)
- ASA with non-large PFO: 2.7 (95% CI: 1.3 to 5.5)
- large PFO without ASA: 0.6 (95% CI: 0.4 to 1.1)
- non-large PFO without ASA: 1.3 (95% CI: 0.7 to 2.3)
At univariable analysis, anatomical characteristics, age and hypertension were associated with an increased likelihood of recurrence, while oral anticoagulation exerted a protective effect. After adjusting for clinical variables (age, hypertension and oral anticoagulation) only presence of ASA (adjusted HR: 3.27; 95% CI: 1.82 to 5.86; p < 0.0001), but not shunt size (average adjusted HR across studies: 1.43; 95% CI: 0.50 to 4.03; p = 0.50) was associated with recurrence of stroke.
Critical reading and the relevance for clinical practice
Following the publication of the early PFO trials (CLOSURE I, RESPECT and PC) the interest towards percutaneous closure was somehow mitigated and its potential role as a preventive measure for recurrent cryptogenic stroke substantially questioned.
Recent availability of the long term follow up of RESPECT trial as well as the results of REDUCE, CLOSE and DEFENSE-PFO showed how length of risk exposure and careful patient’s selection completely change the knowledge about a specific disease/condition. Positive results of trials favoring percutaneous closure have again increased interest and referrals for percutaneous closure. In our practice, the number of patients referred for interventional evaluation or treatment after a cryptogenic stroke has increased, nonetheless the ratio between treated patients and prevented events remained persistently high, stressing the need for adequate criteria to select among all candidates those at the highest risk.
The rationale of the article published by Turc et al. in JACC is represented by the need to discern in patients with PFO the relative impact of ASA and shunt size on stroke recurrence in order to carefully select for closure those in whom the paradoxical mechanism is more likely the determinant of clinical event as well as those who probably might benefit the most from an interventional approach.
In their large, pooled population of medically treated patients after an index stroke, a significant association between presence of ASA and ischemic recurrences was recognized. On the other hand, presence a large shunt did not show any association with recurrent event. Unfortunately, whether their findings will significantly impact on current patient’s selection still has to be proven.
What this study adds to current knowledge is that ASA represents a predictor of increased risk for recurrences after an index event in medically treated patients. However, this finding cannot be directly translated in practice assuming that patients with ASA might more extensively benefit from PFO closure. In a certain way, the same holds true for the lack of association with large shunt.
The neutral impact of shunt size on risk of recurrent ischemic events is somehow counterintuitive. Presence of a large shunt has been regarded as an high-risk anatomical feature and used as an inclusion criteria in CLOSE and DEFENSE-PFO while being extremely frequent both in REDUCE and RESPECT trials. From a clinical perspective, larger shunts are often considered as an adjunctive evidence in supporting PFO closure. Surely a lot depends on the way a large shunt is detected and classified. Current TEE based evaluation of shunt size has several known limitations such as the potential incorrect quantification of bubbles crossing the septum due to improper alignment as well as difficulties in properly obtaining a Valsalva manoeuvre. Data derived from a more objective quantification of shunt size by means of transcranial Doppler might help in adding knowledge to this field.
Another consideration that should be made relates to the role of ASA in cryptogenic stroke. While its impact in increasing the risk of recurrent strokes has now been ascertained, its mechanism remains somehow unclear. In fact, it is still uncertain whether this increased risk relates to an increased likelihood of paradoxical embolism through a (even small) PFO or if its presence might increase the risk of local thrombosis and, thus, systemic embolization. What is known, is that, whatever the case, stabilization of ASA by percutaneous implant of a percutaneous device, would solve both problems.
In an area of uncertainty such as the evaluation and treatment of young patients with cerebral ischemic events of unknown origin, we are all eager to obtain certainties. Unfortunately, as for its intrinsic nature, cryptogenic stroke remains somehow elusive.
No comments yet!