Hereditary hemorrhagic telangiectasia is an autosomal dominant disorder characterized by vascular malformations, and many clinical complications are related to pulmonary arteriovenous malformations (PAVMs) because they provide direct right-to-left shunts. Paradoxical emboli through these shunts are a well-recognized cause of transient ischemic attack, stroke, and cerebral abscess. The aim of this study was to determine whether paradoxical emboli through PAVMs could manifest as cardiac ischemic events. The study included a single-center population of 98 patients with PAVMs. Eighty-four had undergone PAVM embolotherapy, and the remaining 14 patients had PAVMs too small to require embolization. Patients were interviewed by telephone and surveyed regarding their cardiopulmonary symptoms and histories of cardiac diagnoses. We found that 6 patients (which is 18% of patients with symptomatic PAVMs, n = 33, and 6% of the total cohort, n = 98) reported that they had experienced typical angina pectoris-like chest pain or had a myocardial infarction before PAVM embolotherapy. Five patients had had a cardiac catheterization, 4 had normal coronary arteries, and 1 had a single artery occlusion. In conclusion, we suggest that in patients with untreated PAVMs, cardiac ischemia can occur because of a paradoxical embolus through PAVMs to a coronary artery.
Hereditary hemorrhagic telangiectasia (HHT), also known as Osler-Weber-Rendu syndrome, is an autosomal dominant disorder characterized by mucocutaneous telangiectases and/or visceral vascular malformations. A key vascular malformation is the pulmonary arteriovenous malformation (PAVM), for it can cause life-threatening manifestations. Although estimates vary, ≤50% of patients with HHT have PAVMs. A PAVM provides a direct right-to-left shunt, approximately the diameter of the feeding artery, which in effect bypasses the pulmonary capillary bed, allowing the passage of paradoxical emboli. Such emboli most commonly present as stroke, transient ischemic attack, or brain abscess, and treatment of PAVMs with embolotherapy is considered the standard of care to prevent such events. Although cerebral events are the most widely recognized, it stands to reason that paradoxical emboli traveling to other sensitive vascular beds may also occur in patients with PAVMs. Other types of right-to-left shunts have been described in the published studies as the cause of acute coronary syndromes (ACS) due to a paradoxical embolus lodging in a coronary artery. We sought to determine if PAVMs, as right-to-left shunts, might similarly result in cardiac ischemia due to paradoxical emboli to the coronary arteries.
Methods
This study was approved by the Institutional Review Board and was carried out in full compliance with the Health Care Portability and Accountability Act. Patients were identified from a quality assurance database of all patients with PAVMs seen at our institution, which is recognized as a Hereditary Hemorrhagic Telangiectasia (HHT) Center of Excellence, during the period of August 2002 to January 2010. Ninety-eight patients with imaged PAVMs were included, and 80 of those patients had a diagnosis of definite HHT and 8 had a diagnosis of probable HHT. The clinical diagnosis was based on the Curaçao criteria: (1) recurrent, spontaneous episodes of epistaxis, (2) mucocutaneous telangiectases, (3) visceral AVMs (pulmonary, gastrointestinal, cerebral, and hepatic), and (4) a family history of HHT. When 3 criteria are met, it is considered as a definite diagnosis, 2 as a probable diagnosis, and 0 to 1 as an unlikely diagnosis.
The patient population had the following characteristics: mean age of 46 years (range, 3 to 85), 34 men and 64 women, and 86 adults and 12 children <18 years of age. Eighty-four patients had been treated with ≥1 PAVM embolization procedure. The control population consisted of 14 patients with documented PAVMs too small to warrant embolization (feeding arteries <2 mm in diameter), so no embolization was performed. Patients were interviewed by telephone by a single investigator, and verbal consent was obtained for an approximate 5-minute survey of questions regarding their PAVMs and cardiac histories ( Table 1 ). Ninety-eight patients were contacted and both consented and completed the survey; no patient declined. Statistical calculations (including the t test, chi-square, and Fisher’s exact test) were performed to compare the patient characteristics and coronary artery disease (CAD) risk factors.
|
|
|
Results
Survey data are listed in Table 2 . In all, 6 patients (6%) reported typical angina pectoris–like pain, and all of these are in the treated PAVM group and are all before PAVM treatment. Three of these patients had experienced a myocardial infarction (MI). Five patients had had a cardiac catheterization; of these, 4 had normal coronary arteries, and 1 had a single vessel occlusion with otherwise normal coronary arteries. One patient had a normal electrocardiogram at the time of angina-like chest pain; and did not experience an MI or undergo cardiac catheterization. One other patient with atypical symptoms had a negative cardiac catheterization finding. None of these patients had a medical history of any cardiac disease. Additionally, 11 patients self-reported a different type of chest tightness, which seemed to reflect dyspnea on exertion, a common symptom of PAVMs, typically corrected by an embolization procedure. Viewed another way, 33 (39%) of the 84 patients with treatable-sized PAVMs were symptomatic before undergoing an embolization procedure. Of these, 27 (82% of symptomatic patients) experienced dyspnea on exertion, related chest tightness, or both, and the remaining 6 (18% of symptomatic patients) experienced typical angina pectoris–like pain. In the control group of untreated patients with small PAVMs, none experienced any symptoms suggesting a coronary event. Two patients (14%) did experience dyspnea on exertion, and only 1 other patient had a previous cardiac medical history (mitral valve prolapse). The family cardiac history of both the treated and untreated control groups was not significantly different (p = 0.21).
| Patient Characteristic | PAVM Embolization Procedure | p | |
|---|---|---|---|
| Yes (n = 84) | No (n = 14) | ||
| Age (yrs) | |||
| ≥18 | 78 | 8 | <0.01 |
| <18 | 6 | 6 | |
| Mean age | 48 | 29 | <0.01 |
| Men/women | 28/56 | 6/8 | 0.55 |
| HHT diagnosis | |||
| Definite | 68 | 12 | |
| Probable | 7 | 1 | |
| No | 5 | 0 | 0.92 |
| Unknown | 4 | 1 | |
| Genetic testing | 55 (65%) | 8 (57%) | 0.61 |
| HHT mutation | |||
| Endoglin | 22 | 4 | |
| Activin-like receptor kinase type 1 | 3 | 1 | 0.18 |
| No mutation | 30 | 3 | |
| Mean, range of PAVMs treated | 3.7, 1–24 | 0 | |
| Dyspnea on exertion | 27 (32%) | 2 (14%) | 0.22 |
| Typical angina pectoris | 6 (7%) | 0 (0%) | 0.59 |
| Postembolization dyspnea-on-exertion improvement | 25 (66%) | Not applicable | |
| Body mass index (kg/m 2 ) | 77 | 8 | 0.33 |
| <25.0 (normal) | 41 (53%) | 5 (62%) | 0.29 |
| 25.0–30.0 (overweight) | 17 (22%) | 3 (38%) | |
| >30.0 (obese) | 19 (25%) | 0 | |
| Hypertension | 22 (26%) | 0 | 0.03 |
| Diabetes mellitus | 5 (6%) | 0 | 1 |
| Smoker | 48 (57%) | 3 (21%) | 0.23 |
| Past | 37 (77%) | 3 (100%) | 0.55 |
| Current | 11 (23%) | 0 | |
| Mean pack-year exposure | 17.9 | 10.3 | 0.2 |
The statistical results comparing the treatment group (patients who underwent a PAVM embolization procedure) and the control group, show that the 2 groups are significantly different (p <0.05) when comparing current age, hypertension, and hyperlipidemia. However, the statistical results of the treatment group, comparing those 6 patients who had an ACS event to the 72 patients who did not, show that the groups are not significantly different in terms of CAD risk factors before a PAVM embolization procedure, as listed in Table 3 .
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
