Effect of red blood cell storage duration on major postoperative complications in cardiac surgery: A randomized trial.

BACKGROUND: Although observational studies suggest an association between transfusion of older red blood cell (RBC) units and increased postoperative risk, randomized trials have not supported this. The objective of this randomized trial was to test the effect of RBC storage age on outcomes after cardiac surgery. METHODS: From July 2007 to May 2016, 3835 adults undergoing coronary artery bypass grafting, cardiac valve procedures, or ascending aorta repair, either alone or in combination, were randomized to transfusion of RBCs stored for ≤14 days (younger units) or for ≥20 days (older units) intraoperatively and throughout the postoperative hospitalization. According to protocol, 2448 patients were excluded because they did not receive RBC transfusions. Among the remaining 1387 modified intent-to-treat patients, 701 were randomized to receive younger RBC units (median age, 11 days) and the remaining 686 to receive older units (median age, 25 days). The primary endpoint was composite morbidity and mortality, analyzed using a generalized estimating equation (GEE) model. The trial was discontinued midway owing to enrollment constraints. RESULTS: A total of 5470 RBC units were transfused, including 2783 in the younger RBC storage group and 2687 in the older RBC storage group. The GEE average relative-effect odds ratio was 0.77 (95% confidence interval [CI], 0.50-1.19; P = .083) for the composite morbidity and mortality endpoint. In-hospital mortality was lower for the younger RBC storage group (2.1% [n = 15] vs 3.4% [n = 23]), as was occurrence of other adverse events except for atrial fibrillation, although all CIs crossed 1.0. CONCLUSIONS: This clinical trial, which was stopped at its midpoint owing to enrollment constraints, supports neither the efficacy nor the futility of transfusing either younger or older RBC units. The effects of transfusing RBCs after even more prolonged storage (35-42 days) remains untested.

Prognostic Value of Preoperative Red Cell Distribution Width: Fine-Tuning by Mean Corpuscular Volume.

BACKGROUND: Abnormal red cell distribution width (RDW), reflecting heterogeneity of red blood cell (RBC) size, is associated with cardiovascular disease outcomes. However, whether RBC size itself, expressed as mean corpuscular volume (MCV), provides additional prognostic value is unclear. We therefore investigated the relationship between outcomes after cardiac surgery and both RDW and MCV simultaneously. METHODS: From January 2010 to January 2014, 16,097 patients underwent cardiac surgery at Cleveland Clinic and had complete blood count findings available for analysis. Outcomes included RBC transfusion, postoperative complications, and intensive care unit (ICU) and postoperative hospital lengths of stay. Risk-adjusted associations of RDW and MCV with outcomes and their relative importance in predicting outcome were identified by random forest machine learning. RESULTS: High RDW was associated with more RBC transfusions. Except for postoperative atrial fibrillation, risks of complications and ICU and postoperative lengths of stay were at their minimum when RDW was normal, 13% to 14%. The relationship of MCV to complications was U-shaped: high (macrocytosis) and low (microcytosis) values were associated with higher risk. RDW was an important risk factor for most postoperative outcomes and lengths of stay; MCV was less so, but provided prognostic value in addition to RDW alone, particularly when there was macrocytosis. CONCLUSIONS: Abnormal RDW and MCV are associated with higher risk of transfusion and postoperative outcomes after cardiac surgery. RDW is one of the most important variables in predicting outcomes, but MCV provides additional prognostic value. Both should be taken into consideration when estimating the perioperative risk of patients undergoing cardiac surgery.

Real Age: Red Blood Cell Aging During Storage.

BACKGROUND: During cold storage, some red blood cell (RBC) units age more rapidly than others. Yet, the Food and Drug Administration has set a uniform storage limit of 42 days. Objectives of this review are to present evidence for an RBC storage lesion and suggest that functional measures of stored RBC quality-which we call real age-may be more appropriate than calendar age. METHODS: During RBC storage, biochemical substances and byproducts accumulate and RBC shape alters. Factors that influence the rate of degradation include donor characteristics, bio-preservation conditions, and vesiculation. Better understanding of markers of RBC quality may lead to standardized, quantifiable, and operationally practical measures to improve donor selection, assess quality of an RBC unit, improve storage conditions, and test efficacy of the transfused product. RESULTS: The conundrum is that clinical trials of younger versus older RBC units have not aligned with in vitro aging data; that is, the units transfused were not old enough. In vitro changes are considerable beyond 28 to 35 days, and average storage age for older transfused units was 14 to 21 days. CONCLUSIONS: RBC product real age varies by donor characteristics, storage conditions, and biological changes during storage. Metrics to measure temporal changes in quality of the stored RBC product may be more appropriate than the 42-day expiration date. Randomized trials and observational studies are focused on average effect, but, in the evolving age of precision medicine, we must acknowledge that vulnerable populations and individuals may be harmed by aging blood.

Wrong-Patient Blood Transfusion Error: Leveraging Technology to Overcome Human Error in Intraoperative Blood Component Administration.

BACKGROUND: Confirmation of match between patient and blood product remains a manual process in most operating rooms (ORs), and documentation of dual-signature verification remains paper based in most medical institutions. A sentinel event at Johns Hopkins Hospital in which a seriously ill patient undergoing an emergent surgical procedure was transfused with a unit of incompatible red blood cells that had been intended for another patient in an adjacent OR led the hospital to conduct a quality improvement project to improve the safety of intraoperative blood component transfusions. METHODS: A multidisciplinary quality improvement project team led a four-phase implementation of bedside bar code transfusion verification (BBTV) for intraoperative blood product administration. Manual random sample audits of blood component transfusions were used to examine accuracy of documentation from July 2014 through June 2016. After the transition to the Epic anesthesia information management system (AIMS) in July 2016, automated Epic reports were generated to provide population-level audits. RESULTS: After initiation of BBTV and the addition of Epic AIMS, compliance with obtaining three metrics on documentation of patient identification (two electronic signatures, start and stop times of transfusion, and blood volume transfused) was improved during a one-year period to > 96%. Pre-Epic audits had shown a mean compliance of only 86%, mainly reflecting a lack of paper blood component requisitions. CONCLUSION: By implementing BBTV and using a novel intraoperative documentation process within the Epic AIMS, a safer process of blood transfusion in the ORs was initiated and documentation improved.

Preoperative Anemia in Cardiac Operation: Does Hemoglobin Tell the Whole Story?

BACKGROUND: Preoperative anemia, defined by hemoglobin level, is associated with elevated risk after cardiac operation. Better understanding of anemia requires characterization beyond this. This investigation focuses on red cell size and its association with patient characteristics and outcomes after cardiac operation. METHODS: From January 2010 to January 2014, 10,589 patients underwent elective cardiac operations at Cleveland Clinic. Anemia was characterized as normocytic, microcytic, or macrocytic based on mean corpuscular volume (MCV). Models for hospital complications were developed using multivariable logistic regression. Other outcomes were postoperative transfusion and intensive care unit (ICU) and postoperative hospital lengths of stay. RESULTS: A total of 2,715 patients (26%) were anemic. Of these, 2,365 (87%) had normocytic, 219 (8.1%) microcytic, and 131 (4.8%) macrocytic anemia. Non-anemic patients (n = 2,041, 26%) received transfusions compared with 1,553 (66%) normocytic, 148 (68%) microcytic, and 97 (74%) macrocytic anemia patients. Patients with normocytic or macrocytic anemia had more renal failure (normocytic: odds ratio (OR) 1.9, macrocytic: OR 3.5), other complications (normocytic: OR 1.3, macrocytic: OR 2.2) and death (normocytic: OR 2.0, macrocytic: OR 6.2) than non-anemic patients; patients with microcytic anemia had fewer reoperations (OR 0.35) and less postoperative atrial fibrillation (OR 0.50). Anemic patients experienced longer ICU (27 versus 48 hours, p < 0.001) and postoperative hospital (6.1 versus 7.4 days, p < 0.001) length of stay than non-anemic patients. CONCLUSIONS: Cardiac surgical patients are often anemic. Demographic characteristics, comorbidities, and outcomes are dissimilar according to red cell size. Patients with microcytic anemia had the lowest hemoglobin levels, yet the best clinical outcomes among anemic patients. MCV from the standard complete blood count adds additional information beyond hemoglobin for targeted intervention.

A Randomized Clinical Trial of Red Blood Cell Transfusion Triggers in Cardiac Surgery.

BACKGROUND: Class I evidence supporting a threshold for transfusion in the cardiac surgical setting is scarce. We randomly allocated patients to a transfusion hematocrit trigger of 24% versus 28% to compare morbidity, mortality, and resource use. METHODS: From March 2007 to August 2014, two centers randomly assigned 722 adults undergoing coronary artery bypass graft surgery or valve procedures to a 24% hematocrit trigger (n = 363, low group) or 28% trigger (n = 354, high group). One unit of red blood cells was transfused if the hematocrit fell below the designated threshold. The primary endpoint was a composite of postoperative morbidities and mortality. Treatment effect was primarily assessed using an average relative effect generalized estimating equation model. RESULTS: At the second planned interim analysis, the a priori futility boundary was crossed, and the study was stopped. There was no detected treatment effect on the composite outcome (average relative effect odds ratio, low versus high, 0.86, 95% confidence interval: 0.29 to 2.54, p = 0.71). However, the low group received fewer red blood cell transfusions than the high group (54% versus 75%, p < 0.001), mostly administered in the operating room (low group, 112 [31%]; high group, 208 [59%]), followed by intensive care unit (low, 105 [31%]; high, 115 [34%]) and floor (low, 41 [12%]; high, 42 [13%]). The low group was exposed to lower hematocrits: median before transfusion, 22% (Q1 = 21%, Q3 = 23%) versus 24% (Q1 = 22%, Q3 = 25%). CONCLUSIONS: Negative exposures differed between treatment groups, with lower hematocrit in the 24% trigger group and more red blood cells used in the 28% group, but adverse outcomes did not differ. Because red blood cell use was less with a 24% trigger without adverse effects, our randomized trial results support aggressive blood conservation efforts in cardiac surgery.

Does a similar procedure result in similar survival for women and men undergoing isolated coronary artery bypass grafting?

OBJECTIVES: To (1) identify sex-related differences in risk factors and revascularization strategies for patients undergoing coronary artery bypass grafting (CABG), (2) assess whether these differences influenced early and late survival, and (3) determine whether clinical effectiveness of the same revascularization strategy was influenced by sex. METHODS: From January 1972 to January 2011, 57,943 adults-11,009 (19%) women-underwent primary isolated CABG. Separate models for long-term mortality were developed for men and women, followed by assessing sex-related differences in strength of risk factors (interaction terms). RESULTS: Incomplete revascularization was more common in men than women (26% vs 22%, P < .0001), but women received fewer bilateral internal thoracic artery (ITA) grafts (4.8% vs 12%; P < .0001) and fewer arterial grafts (68% vs 70%; P < .0001). Overall, women had lower survival than men after CABG (65% and 31% at 10 and 20 years, respectively, vs 74% and 41%; P ≤ .0001), even after risk adjustment. Incomplete revascularization was associated equally (P > .9) with lower survival in both sexes. Single ITA grafting was associated with equally (P = .3) better survival in women and men. Although bilateral ITA grafting was associated with better survival than single ITA grafting, it was less effective in women-11% lower late mortality (hazard ratio, 0.89 [0.77-1.022]) versus 27% lower in men (hazard ratio, 0.73 [0.69-0.77]; P = .01). CONCLUSIONS: Women on average have longer life expectancies than men but not after CABG. Every attempt should be made to use arterial grafting and complete revascularization, but for unexplained reasons, sex-related differences in effectiveness of bilateral arterial grafting were identified.

Process Improvement in Thoracic Donor Organ Procurement: Implementation of a Donor Assessment Checklist.

BACKGROUND: Donor organs are often procured by junior staff in stressful, unfamiliar environments where a single adverse event can be catastrophic. A formalized checklist focused on preprocedural processes related to thoracic donor organ procurement could improve detection and prevention of near miss events. METHODS: A checklist was developed centered on patient identifiers, organ compatibility and quality, and team readiness. It went through five cycles of feedback and revision using a panel of expert procurement surgeons. Educational in-service sessions were held on the use of the checklist as well as best organ assessment practices. Near miss events before the survey were tallied by retrospective review of 20 procurements, and near misses after checklist implementation were prospectively recorded. We implemented the checklist for 40 donor lung and heart procurements: 20 from Cleveland Clinic and 20 from the University of Minnesota. A final survey assessment was used to determine ease of use. RESULTS: Nine near miss events were reported in 20 procurements before use of the checklist. Thirty-one near miss events of 40 organ procurements were identified and potentially prevented by the checklist. Eighty-seven percent of fellows found the checklist to be unobtrusive to work flow, and 100% believed its use should be mandatory. Mortality was the same before and after implementation of the checklist despite increased patient volumes. CONCLUSIONS: Implementation of a simple checklist for use during thoracic organ procurement uncovered a substantial number of near miss events. A preprocedural checklist for all thoracic organ transplants in the United States and abroad is feasible and would likely reduce adverse events.

Compliance with Surgical Care Improvement Project for Body Temperature Management (SCIP Inf-10) Is Associated with Improved Clinical Outcomes.

BACKGROUND: In an effort to measure and improve the quality of perioperative care, the Surgical Care Improvement Project (SCIP) was introduced in 2003. The SCIP guidelines are evidence-based process measures designed to reduce preventable morbidity, but it remains to be determined whether SCIP-measure compliance is associated with improved outcomes. METHODS: The authors retrospectively analyzed the electronic medical record data from 45,304 inpatients at a single institution to assess whether compliance with SCIP Inf-10 (body temperature management) was associated with a reduced incidence of morbidity and mortality. The primary outcomes were hospital-acquired infection and ischemic cardiovascular events. Secondary outcomes were mortality and hospital length of stay. RESULTS: Body temperature on admission to the postoperative care unit was higher in the SCIP-compliant group (36.6° ± 0.5°C; n = 44,064) compared with the SCIP-noncompliant group (35.5° ± 0.5°C; n = 1,240) (P < 0.0001). SCIP compliance was associated with improved outcomes in both nonadjusted and risk-adjusted analyses. SCIP compliance was associated with a reduced incidence of hospital-acquired infection (3,312 [7.5%] vs.160 [12.9%] events; risk-adjusted odds ratio [OR], 0.68; 95% CI, 0.54 to 0.85), ischemic cardiovascular events (602 [1.4%] vs. 38 [3.1%] events; risk-adjusted OR, 0.60; 95% CI, 0.41 to 0.92), and mortality (617 [1.4%] vs. 60 [4.8%] events; risk-adjusted OR, 0.41; 95% CI, 0.29 to 0.58). Median (interquartile range) hospital length of stay was also decreased: 4 (2 to 8) versus 5 (2 to 14) days; P < 0.0001. CONCLUSION: Compliance with SCIP Inf-10 body temperature management guidelines during surgery is associated with improved clinical outcomes and can be used as a quality measure.

Increasing Disadvantage of "Watchful Waiting" for Repairing Degenerative Mitral Valve Disease.

BACKGROUND: Successful durable repair of severe degenerative mitral regurgitation with low operative mortality encourages intervention in asymptomatic patients rather than "watchful waiting." Our objectives were to assess trends in patient characteristics, timing of intervention, and evolving surgical techniques at a high-volume center, and determine effects of these changes on outcomes after mitral valve (MV) repair over a 25-year period. METHODS: From January 1, 1985, to January 1, 2011, 5,902 patients underwent isolated repair (with or without tricuspid repair for functional regurgitation) for degenerative MV disease at Cleveland Clinic. For illustration, the experience is presented in 3 eras: 1985 to 1997 (era 1, n = 1,184), 1997 to 2005 (era 2, n = 2,400), and 2005 to 2011 (era 3, n = 2,318). RESULTS: In era 3, more patients were asymptomatic on presentation (44% in New York Heart Association [NYHA] class I vs 25% in era 1), with less heart failure (11% vs 29%) and atrial fibrillation (9.9% vs 23%). Full sternotomy decreased from era 1 (n = 1,100/93%) to era 2 (n = 602/25%) (era 3, n = 717/31%), and robotic surgery emerged (n = 577/25%) in era 3. Median length of stay shortened (era 1 = 7 days, era 2 = 5.9 days, era 3 = 5.2 days, p < 0.0001), and in-hospital mortality remained low (era 1 = 5/0.42%, era 2 = 5/0.21%, era 3 = 1/0.043%); 0.73% overall required reoperation on the repaired valve before discharge, and 97% had 0 to 1+ regurgitation at discharge. CONCLUSIONS: Treatment trends over 25 years reveal that rather than watchful waiting, a more aggressive approach to degenerative MV disease, with earlier intervention for severe regurgitation in asymptomatic patients and less invasive operative techniques, is successful, safe, and effective.

Prolonged effect of postoperative infectious complications on survival after cardiac surgery.

BACKGROUND: Whether patients having infections after cardiac surgery are at a survival disadvantage after hospital discharge is unclear. Our objectives were (1) to identify characteristics of such patients and (2) to determine whether this complication is associated with increased mortality beyond hospital discharge. METHODS: In all, 30,414 patients were discharged after isolated coronary artery bypass grafting, valve, ascending aorta repair, or combined procedures from January 2000 to January 2011. Surgical site infection, septicemia, pneumonia, and urinary tract infection occurred in 1,868 patients (6.1%). Propensity matching was used to account for differences in perioperative characteristics and postoperative in-hospital events between these patients and those not having postoperative infections, to give 1,593 propensity-matched pairs. Time-related mortality and instantaneous risk were compared. RESULTS: Surgical site infection occurred in 122 patients (0.40%), sternal wound infection in 263 (0.86%), septicemia in 656 (2.2%), urinary tract infection in 853 (2.8%), and pneumonia in 513 (1.7%). Infections were associated with older age, female sex, larger body mass index, and multiple comorbidities. Among 1,593 propensity-matched pairs, postdischarge survival at 6 months and at 1, 5, and 10 years, respectively, was 89%, 86%, 67%, and 45% for patients without infections, and 86%, 83%, 63%, and 43% (p = 0.008) for patients with infections. Survival differences resulted from a higher, but gradually declining, early instantaneous risk during the first year after surgery. Elevated risk was of shorter duration for surgical site infections than for other infections. CONCLUSIONS: Postoperative infection is associated with a high-risk patient profile, and risk of death is elevated early after hospital discharge. Reasons for this prolonged effect are unclear.

Contemporary bloodletting in cardiac surgical care.

BACKGROUND: Health care providers are seldom aware of the frequency and volume of phlebotomy for laboratory testing, bloodletting that often leads to hospital-acquired anemia. Our objectives were to examine the frequency of laboratory testing in patients undergoing cardiac surgery, calculate cumulative phlebotomy volume from time of initial surgical consultation to hospital discharge, and propose strategies to reduce phlebotomy volume. METHODS: From January 1, 2012 to June 30, 2012, 1,894 patients underwent cardiac surgery at Cleveland Clinic; 1,867 had 1 hospitalization and 27 had 2. Each laboratory test was associated with a test name and blood volume. Phlebotomy volume was estimated separately for the intensive care unit (ICU), hospital floors, and cumulatively. RESULTS: A total of 221,498 laboratory tests were performed, averaging 115 tests per patient. The most frequently performed tests were 88,068 blood gas analyses, 39,535 coagulation tests, 30,421 complete blood counts, and 29,374 metabolic panels. Phlebotomy volume differed between ICU and hospital floors, with median volumes of 332 mL and 118 mL, respectively. Cumulative median volume for the entire hospital stay was 454 mL. More complex procedures were associated with higher overall phlebotomy volume than isolated procedures; eg, combined coronary artery bypass grafting (CABG) and valve procedure median volume was 653 mL (25th/75th percentiles, 428 of 1,065 mL) versus 448 mL (284 of 658 mL) for isolated CABG and 338 mL (237 of 619) for isolated valve procedures. CONCLUSIONS: We were astonished by the extent of bloodletting, with total phlebotomy volumes approaching amounts equivalent to 1 to 2 red blood cell units. Implementation of process improvement initiatives can potentially reduce phlebotomy volumes and resource utilization.

Clostridium difficile infection after cardiac surgery: prevalence, morbidity, mortality, and resource utilization.

OBJECTIVE: Despite increasing efforts to prevent infection, the prevalence of hospital-associated Clostridium difficile infections (CDI) is increasing. Heightened awareness prompted this study of the prevalence and morbidity associated with CDI after cardiac surgery. METHODS: A total of 22,952 patients underwent cardiac surgery at Cleveland Clinic from January 2005 to January 2011. CDI was diagnosed by enzyme immunoassay for toxins and, more recently, polymerase chain reaction (PCR) testing. Hospital outcomes and long-term survival were compared with those of the remaining population in propensity-matched groups. RESULTS: One hundred forty-five patients (0.63%) tested positive for CDI at a median of 9 days postoperatively, 135 by enzyme immunoassay and 11 by PCR. Its prevalence more than doubled over the study period. Seventy-seven patients (48%) were transfers from outside hospitals. Seventy-three patients (50%) were exposed preoperatively to antibiotics and 79 (56%) to proton-pump inhibitors. Patients with CDI had more baseline comorbidities, more reoperations, and received more blood products than patients who did not have CDI. Presenting symptoms included diarrhea (107; 75%), distended abdomen (48; 34%), and abdominal pain (27; 19%). All were treated with metronidazole or vancomycin. Sixteen patients (11%) died in hospital, including 5 of 10 who developed toxic colitis; 3 of 4 undergoing total colectomy survived. Among matched patients, those with CDI had more septicemia (P < .0001), renal failure (P = .0002), reoperations (P < .0001), prolonged postoperative ventilation (P < .0001), longer hospital stay (P < .0001), and lower 3-year survival, 52% versus 64% (P = .03), than patients who did not have CDI. CONCLUSIONS: Although rare, the prevalence of CDI is increasing, contributing importantly to morbidity and mortality after cardiac surgery. If toxic colitis develops, mortality is high, but colectomy may be lifesaving.

Hospitalization before surgery increases risk for postoperative infections.

OBJECTIVES: Exposure to a health care facility before surgery may increase risk for postoperative infections. Our objectives were to (1) determine whether the prevalence of postoperative infections was higher among patients who were hospitalized before cardiac surgery, (2) identify risk factors for infection, and (3) evaluate in-hospital outcomes. METHODS: A total of 32,707 patients underwent cardiac surgery from January 1, 2000, to January 1, 2011. Forty percent (13,107) were hospitalized before their surgery date or were transfers from other health care facilities, and 60% (19,600) were same-day admissions. The primary outcome consisted of a composite infection: pneumonia, sepsis, surgical site infection, and urinary tract infection. The secondary outcome was in-hospital death. The propensity method was used to compare infectious complications and mortality between groups. RESULTS: Overall infectious complications occurred in 2327 patients (7.1%). Overall composite and individual infections decreased over the study period (P for trend <.0001). Among 7814 propensity-matched pairs, 522 infections (6.7%) occurred in the same-day admission group versus 676 (8.7%) in the prior hospitalization group, P<.0001. In-hospital mortality was 1.5% (n=120) for the same-day admission group versus 2.8% (n=221) for the prior hospitalization group (P<.0001). CONCLUSIONS: Although the risk of infection decreased over time, the relationship between exposure to a health care facility before surgical intervention and higher infection risk remained substantial. Further investigation into processes of care surrounding infection control is necessary to reduce postoperative infections and associated morbidity.

Outcomes of patients with human immunodeficiency virus infection undergoing cardiovascular surgery in the United States.

INTRODUCTION: Advances in highly active antiretroviral therapy have dramatically improved the lifespan of patients infected with human immunodeficiency virus (HIV). We sought to examine the impact of HIV status on outcomes in patients undergoing cardiovascular surgery. METHODS: We identified 5,621,817 patients who underwent coronary artery bypass graft (CABG), valve, aortic, or other cardiovascular surgery between 1998 and 2009 from the Nationwide Inpatient Sample. Of these, 9771 (0.17%) patients were seropositive for HIV. Using multivariable logistic regression modeling and 1:1 propensity-score matching, we determined the influence of HIV infection on outcomes. RESULTS: The percentage of HIV+ patients undergoing cardiovascular surgery increased significantly from 0.09% to 0.23%. HIV+ patients were more often male, black, younger than 55 years of age, and on Medicaid, and they were more likely to undergo valve and other cardiovascular surgeries, but less likely to have CABG. Among propensity-matched pairs, patients with HIV were at no increased risk for in-patient mortality. HIV+ patients were more likely to receive a blood transfusion and have any postoperative complication. Patients with HIV were less likely to have a postoperative stroke. Rates of pneumonia, renal complications, and wound infection were similar between the groups. The median length of stay and mean total cost were not different between the groups. Factors that predicted in-hospital death in HIV+ patients included metastatic cancer, coagulopathy, renal failure, and aortic, other, or combined surgical procedure. CONCLUSIONS: Cardiovascular surgery can be performed safely on patients with HIV with no increased hospital mortality and only minimal increased need for blood transfusion.

Analysis of Clostridium difficile infections after cardiac surgery: epidemiologic and economic implications from national data.

OBJECTIVES: Clostridium difficile infections (CDIs) have increased during the past 2 decades, especially among cardiac surgical patients, who share many of the comorbidity risk factors for CDI. Our objectives were to use a large national database to identify the regional-, hospital-, patient-, and procedure-level risk factors for CDI; and determine mortality, resource usage, and cost of CDIs in cardiac surgery. METHODS: Using the Nationwide Inpatient Sample database, we identified 349,122 patients who had undergone coronary artery bypass, valve, or thoracic-aortic surgery from 2004 to 2008. Of these, 2581 (0.75%) had been diagnosed with CDI. Multivariable regression analysis and the propensity method were used for risk adjustment. RESULTS: Compared with the West, CDIs were more likely to occur in the Northeast (odds ratio [OR], 1.28; 95% confidence interval [CI], 1.12-1.47) and Midwest (OR, 1.27, 95% CI, 1.11-1.46) and less likely in the South (OR, 0.80; 95% CI, 0.70-0.90). Medium-size hospitals (OR, 0.88; 95% CI, 0.78-0.99) had a lower risk of CDI than did large hospitals. Older age (>75 years; OR, 2.59; 95% CI, 1.93-3.49), longer preoperative length of stay (OR, 1.51; 95% CI, 1.43-1.60), Medicare (OR, 1.21; 95% CI, 1.05-1.39) and Medicaid (OR, 1.60; 95% CI, 1.31-1.96) coverage, and more comorbidities were associated with CDI. Among the matched pairs, patients with CDIs had greater mortality (302 [12%] vs 187 [7.2%], P<.001), a longer median length of stay (21 vs 11 days, P<.001), and greater median hospital charges ($193,330 vs $112,245, P<.001). The cumulative incremental cost of CDIs was an estimated $212 million annually. CONCLUSIONS: Our results have shown that CDI is associated with increased morbidity and resource usage. Additional work is needed to better understand the complex interplay among regional-, hospital-, and patient-level factors.