Tidings from the Citadel: Warfarin in Dialysis

Warfarin Use and the Risk for Stroke and Bleeding in Patients With Atrial Fibrillation Undergoing Dialysis – Mitesh Shah et al. Circulation. 2014;129:1196-1203.

A retrospective cohort study

Population based cohort of patients > 65 years who were admitted in Canada – Quebec and Ontario with a primary or secondary diagnosis of atrial fibrillation. Used ICD-9/10 codes to determine diagnosis and complications such as bleeding and stroke complications. Drug prescriptions were identified by database in Canada where patients > 65 have prescription benefit. Warfarin use was identified by a filled prescription within 30-days of AF diagnosis

626 dialysis patients and 204,210 nondialysis patients. Did not separate by stages of CKD, but rather HD or non-HD. Dialysis patients were younger, male, CHF, HTN, DM, CAD, and bleeding history.

Dialysis vs Non-Dialysis: CHADS2 >/= 2: 72% versus 55% (indication for anti-coagulation); HAS-BLED >/=3: 85% versus 25%. Similar rates of warfarin prescription (46% vs 51%)

image (3)

In non-dialysis patients, warfarin users had a lower incidence of stroke (2.19 vs 2.51/100 person-years)

In dialysis patients, stroke incidence was similar for warfarin and non-warfarin users: 3.37 versus 2.91/100 person-years.

After adjusting for confounders – warfarin use had a HR of 1.14 in diaylsis patients. While in non-dialysis patients had a HR 0.87 with warfarin use.

After adjusting for confounders, warfarin use, was associated with a 44% higher risk for bleeding event in dialysis patients and 19% higher risk in nondialysis patients.


image (4)

Maester’s Collection: Tetralogy of Fallot

Tetralogy of Fallot (ToF) is one of the first congenital cardiac diseases described with reports dating back to the 17th and 18th centuries. It was clinically defined by Fallot during the late 19th century and with the advent of the Blalock-Taussig procedure (GoreTex graft from the subclavian artery to the pulmonary artery), clinicians finally had a procedure to treat ToF.

It has become obvious with further studies that ToF is a broad spectrum of phenotypes due to varied combination congenital cardiac abnormalities. The phenotype of ToF depends on the dominant anatomic feature of the tetrad. Right ventricular hypertrophy is essentially a secondary phenomenon and likely doesn’t contribute to the heterogeneity of presentations. Further, the physiologic significance of the override aorta is questionable and thus not a dominant factor in the presenting physiology in ToF. The infundibular stenosis and the ventricular septal defect thus remain as the essential elements determining the clinical and physiologic patterns in patients with the tetralogy of Fallot.


The first 3 phenotypes are forms of tetralogy where the 2 components are of equal but varying severity. The final 2 phenotypes, not discussed here, are ones where one lesion is severe and the other is very mild taking on the phenotype of the dominant lesion.

Extreme Tetralogy

Severe pulmonary stenosis or atresia with large VSD

Also called pseudotruncus arteriosus particularly when there is pulmonary atresia. This pattern appears similar to transposition of the great arteries with a single ventricle and pulmonic stenosis. This phenotype produces a large right-to-left shunt with severely reduced or absent pulmonary blood flow from the right ventricle. As a result there is equalization of left and right ventricular pressures. In this pattern there is early evidence of cyanosis and collateralization to the pulmonary circulation.

These patients tend of have lower peripheral arterial saturation and can be astonishingly low on exertion. Previously, these patients died from episodes of extreme hypoxia, but if significant collaterals are formed, they may live into adulthood. Softer systolic murmur at the left sternal border with only the aortic component of S2 is heard. Sometimes a continuous murmur is heard as a result of the flow through collateral arteries to the lungs.

The right heart is enlarged on EKG. Collateral vessels can cause a nodular hilar pattern on x-ray and the heart size is usually enlarged. On barium swallow these vessels may indent the esophagus.

The hypoplasia of the main and at times of the right and left pulmonary arteries renders the establishment of a systemic anastomosis very difficult. Surgery to relieve pulmonic stenosis are not as helpful because the PAs are small or atretic.

Classic Tetralogy

Severe pulmonary stenosis with large VSD

The classic tetralogy phenotype produces a large right-to-left shunting and a small amount of left-to-right shunting. The pulmonic stenosis reduces pulmonary blood flow and thus lowers the pulmonary artery pressures.

Classic triad of early cyanosis, squatting to alleviate symptoms, and episodes of paroxysmal dypsnea. The physical exam reveals cyanosis along with clubbing. Ascultation usually reveals a systolic murmur at the left sternal border, but its location and intensity may vary. Second heart sound generally has a pulmonic component, but A2 may dominate.

EKG reveals RVH. X-rays demonstrate decreased pulmonary vascularity in the lung fields and small pulmonary arteries. Surprisingly the overall heart size is normal, but is often boot shaped by RVH. Right sided aortic is present in 1/4 of patients with classic tetralogy.

Mild Tetralogy

Mild-moderate pulmonary stenosis with small-moderate VSD

In this phenotype the congenital anatomic abnormalities are at balance with each other. The left-to-right or right-to-left shunting is generally limited by the VSD size. As a result, RV pressures may equal LV pressures, but are often less. PA pressures are generally normal or slightly elevated.

These patients tend to have milder symptoms and are acyanotic. The small size of the VSD produces a loud systolic murmur at the left sternal border in nearly all patients and often with a thrill. S2 is variable.

EKG findings tend to be variable as well depending on the degree and direction of shunting. Sometimes RVH maybe present.

Radiographs are also dependent on degree and direction of shunting. Lung field vascularity maybe normal to slightly slightly increased as the pulmonary blood is less restricted and the VSD may allow left-to-right shunting.



Palliative Shunts

Palliative shunts can be used to delay surgical repair, but they are rarely or no longer used today. The Blalock-Taussig shunt is described above. Potts shunt connects the descending  aorta to the left pulmonary artery. Potts shunts have been abandoned because of difficulty in removing during corrective surgery. The Waterston shunt connects the ascending aorta to the right pulmonary artery. The Potts and Waterston shunts are sometimes still used to relieve pulmonary artery stenosis or atresia. These are not used in cases of mild tetralogy as they increase blood flow.

Surgical Correction

With advances in surgical technique, corrective surgery currently is the standard treatment. It is often performed with the goals of relieving of the RVOT obstruction, completely separating the pulmonic and systemic circulations by closing the VSD, and minimizing valve incompetence. When technically feasible, the RVOT is opened by resecting the subinfundibular muscle bundles and patching the area open. Other times, a conduit with an artificial pulmonic valve is inserted between the RV and MPA.

Flea Bytes: BNP Test Characteristic in ED Patients with Dyspnea


The BNP test alone was more accurate than any history or physical exam finding in predicting CHF as the cause of dyspnea. One must remember that BNP has a range of NPV and PPV depending on the value and you shouldn’t use a single cut point in making clinical decisions.

Breathing Not Properly Study: N Engl J Med 2002; 347:161-167 http://www.nejm.org/doi/full/10.1056/NEJMoa020233#t=abstract


Maester’s Collection: Stable Coronary Artery Disease


Managing stable ischemic heart disease is challenging in our current age of readily available PCI and improving stent technology.  To paraphrase Shakespeare,


“To cath or not to cath, that is the question”

First, we should define the severity of cardiac symptoms, so that everyone is on the same page. A common method used in guidelines and studies is the Canadian Cardiovascular Society Angina Severity Class.

Class I: Ordinary activity such as walking and climbing stairs does not cause angina.
Class II: Slight/modest limitation of ordinary activity such as brisk walking or walking greater than 2 blocks or climbing more than a flight of stairs.
Class III: Marked limitation of ordinary activity such was walking 1-2 blocks or climbing a set of stairs at normal pace.
Class IV: Any physical activity causes discomfort and symptoms maybe present at rest.



Incidence of coronary artery disease is around 1% annually, but increases to nearly 4% in the elderly.


In women, the prevalence of CAD is 5-7% in 45-64 year olds and 10-12% in 65-84 year olds. The prevalence is surprisingly a bit lower in men age 45-64 years at 4-7% compared to women of the same age group. However, the prevalence rises to 12-14% in 65-84 year old men.


PCI versus OMT for Stable CAD

COURAGE is a North American multi-center RCT. It was composed of 2 parallel groups. The intervention arm was optimal medical therapy (OMT) plus PCI, while the control arm was OMT alone. The study enrolled 2297 patients from 1999 to 2004. Patients were followed for a mean of 4.6 years. Inclusion criteria included Stable CAD, CCS class I, II, III, or stabilized IV, at least one vessel with 70% stenosis, and objective myocardial ischemia. Patients were excluded if they had unstable CCS class IV, marked ischemia during Bruce I – early part of the exercise protocol, LVEF <30%, refractory HF, shock, >50% occlusion of the left main, had revascularization in past 6 months, and if they were deemed unsuitable for PCI. As an aside, I’m uncertain as to what “unsuitable for PCI” means exactly and how much freedom the investigators had to include/exclude based on this criteria, however we must acknowledge it may contribute to bias.

Courage-BaselineDemographicsThe primray endpoint was a composite of death from any cause and nonfatal MI. 


PCI + OMT did not improve the risk of death or MI, but did improve rates of revascularization. Further, rates of angina or “symptoms” were consistently lower in the PCI group. However, medical therapy alone also improved symptoms by 1-year. Interestingly, in sub group analyses women trended towards benefit as it related to the primary outcome, but in the extended follow up this trend was lost.

Critiques/Criticisms: The predominance of white men limits the generalizability of the COURAGE trial. The rate of medical adherence was impressive and likely could not be achieved in a “real clinical setting”. The majority of patients received older bare metal stents because drug-eluting were not approved until the final 6 months of the study.  Furthermore, 14.5% of lesions were treated only with angioplasty without stenting. Lastly, the results were not stratified by ischemic burden, as it would reason that those with greater ischemic burden would preferentially benefit from PCI.

COURAGE Extended Follow Up

The initial data from the COURAGE study demonstrates that PCI improves angina, but does NOT improve survival or risk of MI. The notion that PCI for stable CAD does not prevent MIs makes physiological sense. MIs are due to plaque rupture, which does not correlate with degree of luminal stenosis. A recently published study follows up on the initial COURAGE study. These patients were followed for 15 years and 381 additional deaths occurred during the extended period. There were a total of 284 deaths (25%) in PCI group and 277 (24%) in OMT. So at even 15 years out, PCI did not demonstrate a survival benefit.

FFR-guided PCI versus OMT

FAME-2 was a multi-center international, non-blinded RCT occurring primarily in United States and Europe. This trial looked to address whether using fractional flow reserve (FFR) guidance to target hemodynamically significant stenosis in stable ischemic disease was of benefit. The primary end point was a composite of all-cause mortality, non-fatal MI, and unplanned hospitalization with urgent revascularization. The follow up for this study was much shorter than for COURAGE at a mere 7-months. 881 patients with stable ischemic heart disease were randomized to either OMT + FFR-guided PCI (n=447) or OMT alone (n=441). Patients who had tortuous or calcified arteries non-amenable to FFR were excluded. Optimal medical therapy was defined as the use of aspirin, beta-1 selective beta blocker, ACEI or ARB, statin +/- ezetemibe titrated to LDL <70, and PCI group also received clopidogrel 75mg for at least a year. Patients undergoing PCI received aspirin and clopidogrel load prior to the intervention.

Patients were primarily male and about 1/4th were diabetic. More than half only had single vessel disease, but more than half had at least one significant lesion in the pLAD or mLAD.

At 7-months, 12.7% of patients suffered the primary end point in the OMT alone group, while only 4.3% in the FFR + OMT group had the primary outcome occur. However, this difference was largely driven by the need for urgent revascularization, while there was no difference in the death or MI. In post-hoc analysis, the authors found a decrease in the rate of death and MI the period between 8 days post intervention and the end of the follow up period. Also there was significant cross-over the study where 44% of patients in the OMT group received a stent and 9% received an additional stent in the FFR group.


  • Optimal medical therapy is equivalent to revascularization in regards to mortality and CV events, particularly MI
  • Both medical therapy and revascularization improve angina
  • Revascularization offers better relief of angina then medical therapy
  • Initial revascularization had decreased need for future revascularization
  • FFR guided therapy may improve death and MI after the immediate periprocedural period, i.e. greater than 7 days.


Shared decision making tools from the Mayo Clinic


  1. National Institutes of Health NH, Lung, and Blood Institute. Morbidity & Mortality: 2012 Chart Book on Cardiovascular, Lung, and Blood Diseases. Bethesda, MD: National Heart, Lung, and Blood Institute; 2012
  2. N Engl J Med 2007; 356:1503-1516
  3. N Engl J Med 2015; 373:1937-1946
  4. N Engl J Med 2012; 367:991-1001
  5. Mayo Clinic Shared Decision Making National Resource Center. Link.

Flea Bytes: Valsalva Maneuver for SVT


Valsalva_maneuver (1)

Mechanism of Action

When the Valsalva is initiated it increases intrathroacic pressure which is relayed to the aorta and baroreceptors located there. Stimulation of the baroreceptor reflex causes slowing of the heart by decreasing sympathetic outflow and increased parasympathetic vagal firing. Then there is an decrease in venous return due to sustained increase in intrathoracic pressure which stimulates sympathetic outflow. This also causes increased peripheral venous vasocontriction, which will cause an increase in venous return when the Valsalva is released. This rush of blood back to the heart causes increased cardiac output that results in increased stretch of the baroreceptors causing increased vagal stimulation to slow down conduction through the AV node. The standard Valsalva breaks SVT about 20% of the time.

Modified Valsalva Maneuver

The modified Valasalva maneuver attempts to maximize the efficiency of this vagal maneuver by increasing venous return. The modified Valsalva is a typical Valsalva maneuver performed in the partially recumbent position followed by a passive leg raise. Appelboam in the REVERT trial examined the efficacy of the modified Valsalva for cardioversion in 413 emergency department patients with SVT in 10 centers in the England. Patients were randomized to either standard or modified Valsalva. Exclusion criteria includes need for immediate cardioversion, SBP < 90, AF or Aflutter, contraindication to Valsalva (such as aortic stenosis, recent MI, glaucoma, or retinopathy), inability to lie flat, and 3rd trimester of pregnancy. Standard maneuver had a conversion rate of 17% at 1 minute, while the modified maneuver had conversion rate of 43%. It reduced the frequency with which adenosine had to be given from 69% to 50% and the need for any anti-arrhythmic treatment from 80% to 57%. It however failed to reduce length of stay or discharge from the ED.


  1. Appelboam, A et al.”Postural modification to the standard Valsalva manoeuvre for emergency treatment of supraventricular tachycardias (REVERT): a randomised controlled trial”. The Lancet. 2015. epub:. Link.
  2. Porth, CJ et al. “The Valsalva maneuver: mechanisms and clinical implications.”. Heart and Lung. 1984. Sept. 13 (5):507-518. Link.