Andra I. Duncan

Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting

Objective:Our objective was to quantify incremental risk associated with transfusion of packed red blood cells and other blood components on morbidity after coronary artery bypass grafting. Design:The study design was an observational cohort study. Setting:This investigation took place at a large tertiary care referral center. Patients:A total of 11,963 patients who underwent isolated coronary artery bypass from January 1, 1995, through July 1, 2002. Interventions:None. Measurements and Main Results:Among the 11,963 patients who underwent isolated coronary artery bypass grafting, 5,814 (48.6%) were transfused. Risk-adjusted probability of developing in-hospital mortality and morbidity as a function of red blood cell and blood-component transfusion was modeled using logistic regression. Transfusion of red blood cells was associated with a risk-adjusted increased risk for every postoperative morbid event: mortality (odds ratio [OR], 1.77; 95% confidence interval [CI], 1.67–1.87; p < .0001), renal failure (OR, 2.06; 95% CI, 1.87–2.27; p < .0001), prolonged ventilatory support (OR, 1.79; 95% CI, 1.72–1.86; p < .0001), serious infection (OR, 1.76; 95% CI, 1.68–1.84; p < .0001), cardiac complications (OR, 1.55; 95% CI, 1.47–1.63; p < .0001), and neurologic events (OR, 1.37; 95% CI, 1.30–1.44; p < .0001). Conclusions:Perioperative red blood cell transfusion is the single factor most reliably associated with increased risk of postoperative morbid events after isolated coronary artery bypass grafting. Each unit of red cells transfused is associated with incrementally increased risk for adverse outcome.

Influence of Concentric Left Ventricular Remodeling on Early Mortality After Aortic Valve Replacement

BACKGROUND Severe left ventricular (LV) hypertrophy increases risk for adverse outcome after aortic valve replacement. Whether LV geometry influences mortality risk after aortic valve replacement is unclear. And, whether LV mass or relative wall thickness (RWT) better predicts risk for adverse postoperative outcomes is unknown. The purpose of this investigation was to examine the influence of LV geometry and LV hypertrophy on morbidity and in-hospital mortality after aortic valve replacement, and to determine whether LV mass or RWT had better prognostic ability. METHODS Between January 1996 and June 2004, 5,083 patients underwent aortic valve replacement. Preoperative echocardiographic data was used to calculate LV mass and RWT. Left ventricular geometry was classified into one of four categories on the basis of LV mass indexed to body height and RWT: (1) concentric hypertrophy, (2) eccentric hypertrophy, (3) concentric remodeling, and (4) normal. Postoperative mortality and multisystem morbidities of patients with concentric geometries were compared to patients with nonconcentric geometries by propensity and logistic regression modeling. Also, prognostic ability of RWT and LV mass was compared. RESULTS Nine hundred sixty-four patients with concentric geometry were propensity-matched to 964 patients with nonconcentric geometry. In-hospital mortality (38 [3.9%] versus 18 [1.9%]; p = 0.007), cardiac morbidity (33 [3.4%] versus 17 [1.8%]; p = 0.022), and prolonged intubation (85 [8.8%] versus 58 [6.0%]; p = 0.019) were higher in patients with concentric versus nonconcentric geometry. Increasing RWT, not LV mass, was associated with adverse outcomes. CONCLUSIONS Concentric geometries are associated with increased risk for in-hospital mortality after aortic valve replacement. Increased RWT is associated with adverse outcomes. Preoperative risk stratification should include assessments of LV hypertrophy and LV geometry.

Recent metformin ingestion does not increase in-hospital morbidity or mortality after cardiac surgery.

BACKGROUND:Perioperative treatment of type 2 diabetes with metformin, an oral hypoglycemic drug, is thought to increase the risk of life-threatening postoperative lactic acidosis. In contrast, metformin improves serum glucose control and has beneficial cardiovascular effects, which may decrease the risk of adverse outcomes. In this investigation we sought to determine the influence of metformin treatment on mortality and morbidity compared with treatment with other oral hypoglycemic drugs in diabetic patients undergoing cardiac surgery. METHODS:In this retrospective investigation, 1284 diabetic patients, with recent oral hypoglycemic ingestion (presumed to be 8–24 h preoperatively), underwent cardiac surgery from 1994–2004. Propensity scores were calculated from a logistic model which included baseline characteristics and perioperative variables. Four-hundred-forty-three (85%) of the metformin-treated patients were matched on nearest propensity score using greedy matching techniques with 443 nonmetformin-treated patients. Postoperative outcomes were compared between matched metformin- and nonmetformin-treated patients. RESULTS:In-hospital mortality, cardiac, renal, and neurologic morbidities were similar between groups. Metformin-treated patients had less postoperative prolonged tracheal intubation [OR (95% CI), 0.3 (0.1, 0.7), P = 0.003], infection [0.2 (0.1, 0.7), P = 0.007] and overall morbidities [0.4 (0.2, 0.8), P = 0.005]. CONCLUSIONS:These data suggest that recent metformin ingestion is not associated with increased risk of adverse outcome in cardiac surgical patients. Alternatively, metformin treatment may have beneficial effects.

Levosimendan in Patients with Left Ventricular Dysfunction Undergoing Cardiac Surgery

BACKGROUND Levosimendan is an inotropic agent that has been shown in small studies to prevent or treat the low cardiac output syndrome after cardiac surgery. METHODS In a multicenter, randomized, placebo‐controlled, phase 3 trial, we evaluated the efficacy and safety of levosimendan in patients with a left ventricular ejection fraction of 35% or less who were undergoing cardiac surgery with the use of cardiopulmonary bypass. Patients were randomly assigned to receive either intravenous levosimendan (at a dose of 0.2 μg per kilogram of body weight per minute for 1 hour, followed by a dose of 0.1 μg per kilogram per minute for 23 hours) or placebo, with the infusion started before surgery. The two primary end points were a four‐component composite of death through day 30, renal‐replacement therapy through day 30, perioperative myocardial infarction through day 5, or use of a mechanical cardiac assist device through day 5; and a two‐component composite of death through day 30 or use of a mechanical cardiac assist device through day 5. RESULTS A total of 882 patients underwent randomization, 849 of whom received levosimendan or placebo and were included in the modified intention‐to‐treat population. The four‐component primary end point occurred in 105 of 428 patients (24.5%) assigned to receive levosimendan and in 103 of 421 (24.5%) assigned to receive placebo (adjusted odds ratio, 1.00; 99% confidence interval [CI], 0.66 to 1.54; P=0.98). The two‐component primary end point occurred in 56 patients (13.1%) assigned to receive levosimendan and in 48 (11.4%) assigned to receive placebo (adjusted odds ratio, 1.18; 96% CI, 0.76 to 1.82; P=0.45). The rate of adverse events did not differ significantly between the two groups. CONCLUSIONS Prophylactic levosimendan did not result in a rate of the short‐term composite end point of death, renal‐replacement therapy, perioperative myocardial infarction, or use of a mechanical cardiac assist device that was lower than the rate with placebo among patients with a reduced left ventricular ejection fraction who were undergoing cardiac surgery with the use of cardiopulmonary bypass. (Funded by Tenax Therapeutics; LEVO‐CTS ClinicalTrials.gov number, NCT02025621.)

The impact of gender on in-hospital mortality and morbidity after isolated aortic valve replacement.

The objective of our retrospective investigation was to examine the influence of gender on in-hospital mortality and morbidity after isolated aortic valve replacement (AVR). Between January 1993 and June 2002, 2212 patients (782 females, 1430 males) underwent AVR. Propensity matching was used to adjust for numerous differences in baseline characteristics and perioperative variables between groups. Unadjusted in-hospital mortality was higher in females (27 [3.5%] females versus 23 [1.6%] males; P = 0.005). An analysis using 1:1 matching by propensity score did not find a significant difference in in-hospital mortality [OR (95% confidence intervals), 1.0 [0.4, 2.6]; P = 0.99) or overall morbidity (1.4 [0.7, 2.5]; P = 0.29) between groups. Further analyses, including classification of women and men into quintile groups by propensity scores and logistic regression models with propensity score adjustment, found that females were at increased risk for cardiac morbidity [OR (95% CI), 3.4 [1.1, 10.8]; P = 0.038), but not mortality (0.9 [0.3, 2.5]; P = 0.88) nor other morbidities. These results suggest that there is no greater than a 2.5-fold increase in risk for females compared with males undergoing AVR. Female gender, however, may impart increased risk for cardiac morbidity after AVR.

Recombinant Factor VII Is Associated With Worse Survival in Complex Cardiac Surgical Patients

BACKGROUND Recombinant activated factor VII (rFVIIa) decreases requirements for allogeneic blood transfusion and chest reexploration in patients undergoing cardiac surgery. Whether rFVIIa increases the risk of postoperative adverse events is unclear. We tested whether rFVIIa administration was associated with increased mortality and neurologic and renal morbidity in patients undergoing cardiac surgery. Risk of thromboembolic complications and the dose-response of rFVIIa on mortality and morbidity were also evaluated. METHODS Of 27,977 patients who had complex cardiac surgery, 164 patients (0.59%) received rFVIIa perioperatively. Using propensity-matching techniques, patients were matched to a maximum of 3 control patients. Patients who received rFVIIa were compared with control patients on risk of mortality, neurologic and renal morbidity, and thromboembolic complications, including a composite of myocardial infarction, pulmonary embolism, and deep venous thrombosis. A corresponding dose-response analysis using multivariable logistic regression was also performed. RESULTS Propensity techniques successfully matched 144 patients (88%) with 359 control patients. Of patients who received rFVIIa, 40% experienced in-hospital mortality compared with 18% of control patients (odds ratio, 2.82; 98.3% confidence interval, 1.64 to 4.87; p<0.001). Furthermore, 31% of patients treated with rFVIIa versus 17% of control patients experienced renal morbidity (odds ratio, 2.07; 98.3% confidence interval, 1.19 to 3.62; p=0.002); however, neurologic morbidity and thromboembolic complications were not different among groups. High-dose rFVIIa (>60 μg/kg) did not increase the risk for mortality compared with treatment with low-dose rFVIIa (<60 μg/kg). CONCLUSIONS Administration of rFVIIa is associated with increased mortality and renal morbidity in patients undergoing cardiac surgery.

Occult Tumor Embolization Detected by Intraoperative Transesophageal Echocardiography

rior vena cava (IVC) in 4% to 10% of patients, and of these patients, further extension into the right atrium occurs in 2 to 16%. 2 The evaluation of tumor extension requires preoperative investigation with a chest computed tomography scan and, if intracardiac extension of tumor is suspected, transthoracic echocardiography. Because the presence of vena caval extension of the tumor has no impact on survival, complete resection of tumor at the time of surgery is the goal. 3 The Neves and Zincke classification of RCC tumors is based on tumor extension into the IVC, where a level IV tumor is indicated by extension into the right atrium. Transesophageal echocardiography (TEE) may be used as a surveillance tool during resection of RCC in patients with a level IV tumor because thrombus may embolize into the pulmonary circulation during surgery. Sudden pulmonary tumor embolus during surgical excision has been reported as a cause of intraoperative mortality. 4 In addition, TEE may be used to examine intracardiac tumor involvement and confirm complete removal of the tumor after resection. 5 In this report, the authors present a case in which TEE was instrumental in detecting clinically silent migration of a tumor embolus to the pulmonary artery during the resection of RCC with vena caval extension. CASE REPORT A 68-year-old man with a history of coronary artery disease and RCC presented for nephrectomy, tumor thrombectomy, and coronary artery bypass graft surgery. The preoperative workup included a chest computed tomography scan, which showed intracardiac extension of the RCC into the IVC and the right atrium, and a cardiac catheterization, which revealed a 70% stenosis of the left anterior descending coronary artery. Preoperative transthoracic echocardiography also revealed a large mass in the IVC entering the right atrium and passing through the tricuspid valve with moderate tricuspid regurgitation. A nephrectomy was planned via a midline laparotomy followed by median sternotomy, right atriotomy, tumor thrombectomy, and coronary artery bypass graft surgery on cardiopulmonary bypass without aortic cross-clamping. After the uneventful induction of general anesthesia, intraoperative TEE confirmed previous transthoracic echocardiography findings of an IVC tumor that extended into the right atrium and traversed the tricuspid valve into the right ventricle (Fig 1 and Video 1 [supplementary videos accompanying this article are available online]) and moderate tricuspid regurgitation (TR). Normal right and left ventricular size and function were seen, but no other valvular abnormalities, abnormalities of the main pulmonary artery or its branches, or patent foramen ovale were noted. During nephrectomy and IVC mobilization, the position of the IVC/right atrial tumor mass was continuously monitored by TEE, and its position was unchanged. Subsequently, a sternotomy was performed, and the left internal thoracic artery was harvested. At the time of harvesting of the left internal thoracic artery, a repeat TEE examination could no longer visualize the tumor in the right atrium or ventricle. Further inspection localized the tumor embolus to the main pulmonary artery (Fig 2 and Video 2). Hemodynamics were stable during tumor migration. No TEE signs of increased right ventricular (RV) afterload were seen. The TEE discovery required a revision of the surgical plan including initiation of cardiopulmonary bypass with aortic cross-clamping, cardioplegic-induced cardiac arrest, and an incision to the main pulmonary artery. Subsequently, a thrombus measuring 13 3 cm was removed from the pulmonary artery. After the anastomosis of the left internal thoracic artery, the patient was successfully weaned from cardiopulmonary bypass. TEE examination of the IVC, right atrium, and pulmonary artery showed complete removal of the tumor and resolution of the tricuspid valve incompetence. DISCUSSION Intraoperative TEE played an important role in determining surgical strategy in this report by confirming the tumor location and the degree of cardiac involvement, detecting tumor embolization despite the absence of clinical findings, and confirming complete tumor removal and resolution of tricuspid regurgitation. In addition, TEE functioned as an essential monitoring tool because insertion of a pulmonary artery catheter is contraindicated in the presence of a large intra-atrial tumor. In this report, TEE findings significantly impacted surgical management. Routine TEE monitoring of the tumor position during intraoperative IVC mobilization was uneventful. Unexpectedly, the tumor embolized to the main pulmonary artery during a period without IVC manipulation and without subsequent hemodynamic effect. Only the discovery by TEE alerted anesthesia and surgical teams to tumor migration. The decision to convert from a beating-heart atriotomy to cardioplegicinduced cardiac arrest on full cardiopulmonary bypass was made in order to reduce the possibility of the tumor embolus migrating further into the pulmonary circulation. After tumor removal, TEE was instrumental in confirming complete tumor removal and resolution of tricuspid valve incompetence so that surgical repair of this valve was unnecessary. An intracardiac tumor may cause diverse echocardiographic findings depending on its location. Tumors located in the region of the tricuspid valve may result in TR, intermittent RV inflow obstruction, and subsequent pulmonary thromboembolism. TR occurs if the tumor mass alters the geometry of the valve leaflets or interferes with their complete closure. This was shown echocardiographically by protrusion of the mass through the tricuspid valve annulus during diastole while inhibiting usual systolic closure of valve leaflets. The tumor may also cause RV inflow obstruction, which is detected by TEE findings of right atrial dilation, widened tricuspid orifice, failure of tricuspid valve leaflet coaptation, and an underfilled right ventricle. 6 Acute pulmonary embolus may occur and is detected by indirect consequences of a sudden increase in RV afterload. A 25% to 30% obstruction of the pulmonary vascular bed is

27: RESOURCE UTILIZATION AND COST OF LOW CARDIAC OUTPUT SYNDROME AFTER CARDIAC SURGERY.

Learning Objectives: Patients with myocardial dysfunction requiring cardiac surgery with cardiopulmonary bypass are at high risk for postoperative low cardiac output syndrome (LCOS), a serious condition associated with significant morbidity and mortality. This study assessed costs, hospital resource utilization, and all-cause readmissions for LCOS following cardiac surgery. Methods: This retrospective observational study evaluated 59,810 patients from 164 US hospitals in the Premier Healthcare Database undergoing CABG and/or valve surgery with cardiopulmonary bypass from July 1, 2012 through June 30, 2014. Postoperative LCOS was defined as having one or more of the following: mechanical assist device (intra-aortic balloon pump, ventricular assist device, ECMO) within 5 days after surgery or treatment with ≥2 inotropes on or after postoperative day 1. Generalized linear models adjusted for potential confounding patient and hospital variables were identified in the database. Cost models used gamma distribution, and hospital length of stay model used negative binomial distribution, both with a log link function. Logistic regression was used for binary outcomes for readmission rates. Recycled predictions method was used to evaluate marginal effects of LCOS. Results: Postoperative LCOS was identified in 10.1% (6,067/59,810) of patients. Adjusted mean ± STD hospital cost with LCOS was

Transfusion in Coronary Artery Bypass Grafting is Associated with Reduced Long-Term Survival

57,219 ±

Red Cell Transfusion is Associated With an Increased Risk for Postoperative Atrial Fibrillation

17,901 vs