Introduction
Pulmonary vein stenosis (PVS) is a relatively infrequent yet detrimental complication of atrial fibrillation (AF) ablation thought to be related to intimal thermal or barotrauma of the pulmonary veins (PV). Early AF ablative techniques focusing specifically on the PV ostia resulted in PVS rates of over 40%. , However, with improved ablation techniques, the incidence of severe PVS has decreased and is now reported between 0.32% and 3.4%. , , Ascertaining the specific frequency of PVS, however, has become more difficult because current Heart Rhythm Society (HRS) consensus statements do not recommend routine postablation screening, and some providers now forgo surveillance imaging. These recommendations have resulted in the fact that when PVS occurs, the diagnosis is often delayed or missed. Symptomatic patients may come to attention months after the initial AF ablation procedure during which time the stenosis has continued to develop or may have even progressed to complete occlusion related to scarring and contracture of the PV itself, making interventions very difficult with associated with a less-than-optimal outcome.
Initial presentation and diagnostic workup
The diagnosis and management of PVS have been previously described in a large series of 124 patients, which demonstrates the challenge of diagnosing PVS. This challenge relates to the fact that patients with PVS often present with nonspecific symptoms, including dyspnea, fatigue, chest pain, cough, and even hemoptysis ( Table 23.1 ). Some patients present with findings consistent with bronchitis both clinically and radiographically—and are treated for this—with further delays to diagnosis. As many of these symptoms (except for hemoptysis) can be associated with the recurrence of AF or other nonspecific conditions, there may be a multimonth delay (median of over 4 months) between symptom onset and correct diagnosis. Once the diagnosis is suspected, a number of tests are usually performed. Conventional chest radiography may identify areas of infiltration, or infarction, or sometimes nodularity. Conventional transthoracic echocardiography (TTE) may not be diagnostic. The most accurate modality is dedicated computed tomography (CT) imaging, specifically CT angiography with contrast enhancement timed for opacification of the PVs before the index AF ablation ( Fig. 23.1 ). Careful preprocedure evaluation with CT includes analysis of multiple components, including the pulmonary parenchyma for signs of hemorrhage or infarction, detailed evaluation of the PVs (specifically the number of veins), the distribution, whether there is a common antrum of the left superior and left inferior PVs, the diameter of each vein, its branching pattern, if it is occluded, the length of the occlusion and the diameter of the veins distal to the occlusion, the severity of the stenosis, the presence of abnormalities which suggest thrombus, and the size of the right atrium (RA) and the location of the veins within it. For reasons that are still unclear, we note a higher incidence of left-sided PVS.
| Variable, n (%) | Patients ( n = 124) |
|---|---|
| Any symptom | 101 (83%) |
| Dyspnea at rest | 62 (67%) |
| Exertional dyspnea | 71 (69%) |
| Cough | 46 (45%) |
| Fatigue | 47 (45%) |
| Decreased exercise tolerance | 46 (45%) |
| Chest pain (pleuritic) | 23 (22%) |
| Chest pain (exertional) | 40 (38%) |
| Hemoptysis | 28 (27%) |
| Flulike | 10 (10%) |
| Diagnosis | |
| Time from ablation to symptom onset (months) | 4.0 ± 3.0 |
| Time from onset of symptoms to PVS diagnosis (months) | 4.4 ± 5.4 |
| Number of patients initially misdiagnosed, n (%) | 41 (35%) |
After diagnosis, patients typically should undergo transesophageal imaging to assess venous anatomy and pulsed-wave Doppler assessment, as well as ventilation/perfusion (V/Q) ratio, which generally shows defects corresponding to the affected PV ( Fig. 23.2 ). Cardiac magnetic resonance can also be used to evaluate PVS; however, this is less well clinically validated and awaits future evaluation and study. Once symptomatic PVS is identified, the time course of subsequent evaluation and treatment should be based on the severity of the stenosis. Very severe stenoses may progress rapidly to complete occlusion, which then renders attempted intervention considerably more difficult. We generally define severe PVS as greater than 75% PV obstruction as assessed by dedicated CT scan or angiography (if available) on preintervention imaging. If severe or subtotal stenoses are identified, invasive evaluation should be considered urgently. After the initial PVS intervention, patients are followed at a regular interval with repeat CT angiograms and V/Q scans guided by clinical symptoms.
Percutaneous intervention
Our practice is to preload patients with 325 mg chewable aspirin and 600 mg clopidogrel by mouth on the day of the procedure. Vitamin K antagonists can be continued up to the date of the procedure at a reduced dose to achieve an international normalized ratio (INR) less than 2. For severe PVS and elevated CHADS 2 -VAS c score (especially with prior stroke), one can consider bridging with low-molecular-weight heparin (LMWH). However, direct oral anticoagulants (DOACs) should be held for an appropriate period, as there is not only a risk of vascular injury with larger-bore sheaths in the femoral veins but also risk of cardiac perforation and subsequent pericardial effusion/tamponade. There is no role for bridging with LMWH and DOACs. Postprocedural antiplatelet and anticoagulation therapy will be discussed later in the chapter.
Patients are brought to the catheterization laboratory in the fasting state and maintained under light sedation throughout the procedure if intracardiac echocardiogram (ICE) is to be used, or general anesthesia should transesophageal echocardiography (TEE) be utilized. The choice of imaging should be operator and institution dependent; however, working and successful images can be obtained with both modalities complete with Doppler color and velocity measurements. Imaging is crucial not only to visualize and access the affected PVs but also to assist with the transseptal puncture. Biplane fluoroscopic imaging is preferred if available.
Vascular access is established utilizing two venous sheaths (8F and 9 to 11F) placed in the right and left femoral veins and a 4F arterial line either in the femoral or radial artery. ICE (we commonly use a four-direction, steerable 64-element 10F ICE catheter) is inserted in the left venous site. We usually prefer the larger, 10F system; however, a 9F system is available and can be found in 8F or 9F sizes, necessitating a long (30 cm) and one F size larger sheath. For example, an 8F ICE catheter would be best delivered through a 9F × 30 cm sheath. The ICE catheter not only can aid in transseptal puncture but is also used to identify the PV ostium and quantify the severity of PVS by direct measurement of the ostial diameter and Doppler assessment of blood flow velocity. In some patients, assessment of full right heart pressures is helpful—for example, in patients in whom symptoms may be related to pulmonary hypertension, or a noncompliant left atrium (LA). In occasional cases a subselective pulmonary angiogram with delayed imaging of left atrial emptying may be used if the relevant PV appears to be occluded.
Left atrial access and pulmonary vein engagement
Transseptal access is discussed in a previous chapter; however, it can be obtained in a variety of methods. Importantly, the use of ICE and TEE has allowed for crossing the septum in a specific location. As the PVs are quite posterior in the LA, an anterior transseptal puncture can be useful to approach both the left- and right-sided veins. For transseptal access, typically either an 8F transseptal sheath (St. Jude, SL1) is used to cross to the LA or an Agilis sheath, and wire position is established in the LA with an Amplatz stiff or Inoue wire placed through the sheath that was used to initially cross the septum. Once in the LA, heparin should be administered as needed to maintain an activated clotting time (ACT) between 250 and 350 seconds, generally at a dose of 200 units per kilogram. At this point, with the wire position firmly established in the LA, a steerable sheath such as an Agilis or, alternatively, Dexterity sheath can be introduced into the LA for more precise movement and positioning. In patients with a thickened fibrotic septum, if there are problems associated with needle entry into the LA, a Safe-Sept wire or radiofrequency delivered to the needle may be of value. Should the steerable sheath not readily cross the septum, dilation with either the Inoue dilator or even a coronary or peripheral set of wire/balloon can be used to readily gain access to the LA with the larger, steerable sheath. Certainly, a larger guiding catheter such as a 6F to 8F multipurpose catheter can be used to directly access and engage the affected PV. A conventional coronary guiding catheter is usually used in the event that intravascular ultrasound (IVUS) is needed to assess the lesion more carefully. Without the steerable sheath, however, this creates some difficulty in terms of support as well as maintaining access to the LA while switching out equipment. Either along or through the steerable sheath, we commonly use a 6F multipurpose catheter to intubate the affected PV, using clockwise maneuvering for posterior positioning and counterclockwise movements to position the sheath more anteriorly. Engagement with the orifice of the affected PV can be confirmed by brief contrast injection on fluoroscopic angiography or with ICE guidance using the venous jet.
Wire crossing
To cross the stenotic PV, one can use a 0.035″ Stork guidewire, similar diameter (0.035″) or pediatric (0.025″) hydrophilic Glidewire, a 0.018″ V-18 wire, or even a 0.014″ coronary wire, depending on the degree and character of the stenosis. Wire position across the stenosis should be confirmed with angiography in multiple planes as well as echocardiography (TEE or ICE). If no microchannel can be identified on CT scan, the ICE catheter can be used to interrogate the LA for a velocity representing flow. In addition, focused contrast injections could be used to try to identify a “beak” representing the site of occlusion. Often, despite a diagnosis of complete PV occlusion on CT, in up to 40% of the time there is a microchannel able to be crossed when the patient is taken to the catheterization laboratory. This places a premium on CT scanning as the initial diagnostic modality with subsequent hemodynamic catheterization with PV angiography to confirm complete occlusion. If no potential site or microchannel can be identified, the PV could be considered chronically occluded and the procedure aborted at this point because even with careful probing, there may be damage to the vein or LA itself.
Pulmonary vein intervention: Dilation and stenting
Once across the stenosis, one should not lose the wire position; it will be important to maintain access and upgrade the wire for more support to deliver balloons and stents. If possible, advance the multipurpose catheter into the PV for simultaneous PV/LA pressure measurements (see Fig. 23.3 ). LA pressure can either be obtained with a second catheter (4F multipurpose) in the LA or through the sidearm of the steerable catheter. In our series, higher-pressure gradients portend more difficult procedures requiring more devices and higher balloon/stent inflation pressures. Any pressure gradient between the PV/LA is abnormal and should be intervened upon if the patient is symptomatic. Along those lines, this is also an excellent postprocedural method—along with IVUS—to assess the intervention, as the post-balloon/stent gradient should be as low as possible. Higher residual pressure gradients also portend higher rates of restenosis.
