A decade's experience with quality improvement in cardiac surgery using the Veterans Affairs and Society of Thoracic Surgeons national databases.

OBJECTIVE: To review the Department of Veteran Affairs (VA) and the Society of Thoracic Surgeons (STS) national databases over the past 10 years to evaluate their relative similarities and differences, to appraise their use as quality improvement tools, and to assess their potential to facilitate improvements in quality of cardiac surgical care. SUMMARY BACKGROUND DATA: The VA developed a mandatory risk-adjusted database in 1987 to monitor outcomes of cardiac surgery at all VA medical centers. In 1989 the STS developed a voluntary risk-adjusted database to help members assess quality and outcomes in their individual programs and to facilitate improvements in quality of care. METHODS: A short data form on every veteran operated on at each VA medical center is completed and transmitted electronically for analysis of unadjusted and risk-adjusted death and complications, as well as length of stay. Masked, confidential semiannual reports are then distributed to each program's clinical team and the associated administrator. These reports are also reviewed by a national quality oversight committee. Thus, VA data are used both locally for quality improvement and at the national level with quality surveillance. The STS dataset (217 core fields and 255 extended fields) is transmitted for each patient semiannually to the Duke Clinical Research Institute (DCRI) for warehousing, analysis, and distribution. Site-specific reports are produced with regional and national aggregate comparisons for unadjusted and adjusted surgical deaths and complications, as well as length of stay for coronary artery bypass grafting (CABG), valvular procedures, and valvular/CABG procedures. Both databases use the logistic regression modeling approach. Data for key processes of care are also captured in both databases. Research projects are frequently carried out using each database. RESULTS: More than 74,000 and 1.6 million cardiac surgical patients have been entered into the VA and STS databases, respectively. Risk factors that predict surgical death for CABG are very similar in the two databases, as are the odds ratios for most of the risk factors. One major difference is that the VA is 99% male, the STS 71% male. Both databases have shown a significant reduction in the risk-adjusted surgical death rate during the past decade despite the fact that patients have presented with an increased risk factor profile. The ratio of observed to expected deaths decreased from 1.05 to 0.9 for the VA and from 1.5 to 0.9 for the STS. CONCLUSION: It appears that the routine feedback of risk-adjusted data on local performance provided by these programs heightens awareness and leads to self-examination and self-assessment, which in turn improves quality and outcomes. This general quality improvement template should be considered for application in other settings beyond cardiac surgery.

The STS National Database: current changes and challenges for the new millennium. Committee to Establish a National Database in Cardiothoracic Surgery, The Society of Thoracic Surgeons.

BACKGROUND: The Society of Thoracic Surgeons (STS) established the National Database (NDB) for Cardiac Surgery in 1989. Since then it has grown to be the largest database of its kind in medicine. The NDB has been one of the pioneers in the analysis and reporting of risk-adjusted outcomes in cardiothoracic surgery. METHODS AND RESULTS: This report explains the numerous changes in the NDB and its structure that have occurred over the past 2 years. It highlights the benefits of these changes, both to the individual member participants and to the STS overall. Additionally, the vision changes to the NDB and reporting structure are identified. The individuals who have participated in this effort since 1989 are acknowledged, and the STS owes an enormous debt of gratitude to each of them. CONCLUSIONS: Because of their collective efforts, the goal to establish the STS NDB as a "gold standard" worldwide for process and outcomes analysis related to cardiothoracic surgery is becoming a reality.

Using database information in your clinical practice.

BACKGROUND: Clinical database information is in wide demand, but it is not always used to its full potential. Clinicians must learn to be the experts and to assert leadership in the use of their own data. METHODS: Clinical databases provide unique perspectives on the full process of care for a heterogeneous population of patients. They focus beyond individual providers, to their interaction within a complex system of care. Exploring questions that arise from this data can identify system issues, which are invisible to individual practitioners or specialties using other traditional review methods. RESULTS: Clinical database information helped our hospital staff identify problems in their approach to a small subset of high risk coronary bypass patients. Multiple system changes resulted in a dramatic reduction in mortality. Collateral impact on all care reduced overall mortality from 4.5% to below 2%. CONCLUSIONS: The greatest opportunities for improvement in patient care often lie in the areas where specialties or teams interface, eg, in overlapping or transferring care. The whole system of care for each patient must be optimized, not just the individual specialty components. Clinical database information provides a way to evaluate and improve the overall process and results of the broader system of patient care.

Using information from databases to improve clinical practice: lessons learned under fire.

BACKGROUND: Information derived from a clinical database can be used to produce reports with valid, applicable data, to assemble outcomes data for purposes of marketing or practice survival, and to improve and optimize patient care. METHODS: In response to lay-press figures suggesting a high risk-adjusted mortality at St. Peter's Hospital (Albany, NY), a multidisciplinary group at the hospital undertook a collaborative review of information from our clinical database to identify any patterns that could explain this disturbing summary statistic. RESULTS: This review showed that for the vast majority (> 95%) of cases mortality was on a par with or lower than the statewide average. The elevated mortality was confined to a small and specific subset of high-risk patients. Once identified, attention was focused on this group of patients, and both cardiologists and surgeons discussed practice changes based on information from the clinical database. Since program changes were instituted in 1993, mortality has decreased in both high-risk and all other patients. Overall mortality for coronary artery bypass graft patients has decreased from 4.5% to about 1.5%. CONCLUSIONS: Use of a database to extract a single number reflecting patient-care performance, eg, short-term mortality, can conceal more information than it conveys. Only in-depth analysis of the information in the database can identify areas for improvement in clinical practice.

Deleterious effects of buthionine sulfoximine on cardiac function during continuous endotoxemia.

Sepsis has been associated with reversible cardiac injury. To determine whether this injury is mediated by generation of reactive oxidants, tissue glutathione (GSH)--the major intracellular antioxidant--was depleted before endotoxemia. Basal values of cardiac contractile function, perfusion, and cardiac output were measured 5-7 days postsurgery. Salmonella enteritidis endotoxin was continuously infused at 3 micrograms/kg/hr iv via an osmotic pump (Alzet Corp). Endotoxemia significantly reduced myocardial glutathione content (394 +/- 46) to 206 +/- 9 micrograms/g), indicating oxidant stress during endotoxemia. Buthionine sulfoximine (BSO) pretreatment significantly reduced cardiac glutathione in sham pigs from 394 +/- 46 to 199 +/- 26 micrograms/g; and in endotoxemic pigs, BSO pretreatment significantly reduced cardiac glutathione to 106 +/- 18 micrograms/g. Vehicle- and BSO-treated endotoxemic groups demonstrated similar cardiovascular responses to endotoxin challenge. Heart rate increases (122 +/- 15 to 140 +/- 17 bpm) and cardiac outputs decreases (1.50 +/- 0.24 to 1.11 +/- 0.35 l/min) were similar, indicating similar cardiovascular insults induced by endotoxemia. Percent short axis shortening and end-systolic pressure-diameter relation (ESPDR) were significantly reduced in BSO pretreated compared with vehicle-treated endotoxemic pigs. Results support a conclusion that endotoxemia-induced cardiac injury is mediated, in part, by free radical injury. This conclusion is based upon the finding that endogenous myocardial glutathione was depleted by continuous endotoxin infusion and that prior depletion of myocardial glutathione by buthionine sulfoximine exacerbated cardiac injury.

Pericardial drainage procedures.

Pericardial drainage is generally required for diagnosis and treatment of patients with pericardial effusion, and often for associated cardiac tamponade. Various effective techniques of drainage are available, each with different advantages and disadvantages; no single technique is optimal for all patients and circumstances. Pericardiocentesis or percutaneous tube drainage may be useful for patients with acute tamponade, although these carry significant risks and are safest when done under controlled conditions in the catheterization laboratory. Our preferred method of pericardial drainage in the majority of patients is the subxiphoid approach. This technique allows rapid access to the pericardium and is associated with low morbidity and excellent long-term results. Patients with infectious pericarditis should be monitored closely following pericardial drainage, and early pericardiectomy may be required, especially in children with H. influenzae. Extensive pericardiectomy by either sternotomy or thoracotomy should be reserved for patients with pericardial constriction or failures of pericardial window.

Subxiphoid pericardial drainage for pericardial tamponade.

As a result of recent reports and enthusiasm for video-assisted thorascopic pericardiectomy, we reviewed our experience with subxiphoid pericardial drainage. From August 15, 1988, to June 7, 1993, 155 patients underwent subxiphoid pericardial drainage for pericardial effusion associated with pericardial tamponade. The group comprised 85 female (55%) and 70 male patients whose ages ranged from 5 weeks to 88 years. The procedure was carried out with general anesthesia in 113 patients (72%) and with local anesthesia and sedation in 42 patients. Underlying cancer was present in 82 patients; 73 patients had benign disease. Follow-up is complete in all patients. The overall 30-day mortality was 20%; in patients with cancer it was 32.9% (27/82) versus 5.4% (4/73) for patients with benign disease. No postoperative death was attributed to the surgical procedure. Recurrent pericardial tamponade necessitating further surgical intervention occurred in four patients (2.5%), two with cancer (2.4%) and two with benign disease (2.7%). Median survival after subxiphoid pericardial drainage in patients with benign disease was more than 800 days versus 83 days in patients with cancer (p < 0.01). Median survival after pericardial drainage in patients with cancer who had malignant pericardial effusion was 56 days compared with 105 days for patients with cancer who did not have tumor in the pericardium (p < 0.05). We believe that subxiphoid drainage is the procedure of choice for patients with pericardial tamponade. It is accomplished quickly, is associated with minimal morbidity, and prevents recurrent tamponade in 97.4% (151/155) of patients.

How a New York cardiac surgery program uses outcomes data.

In New York State, a risk-adjusted outcomes system has been used by the Department of Health to monitor all cardiac operations since January 1989. Hospital-specific and physician-specific results are published annually. In this report we describe the experience of one hospital in New York State whose results showed a higher than expected surgical mortality. Staff reactions were initially skeptical, and case reviews found no quality-of-care problems. However, a different approach using statistical analysis of the detailed case-specific outcomes data was more revealing. The excess mortality was localized to patients having high-acuity, emergency coronary artery bypass grafting, particularly those who had suffered a preoperative acute myocardial infarction less than 6 hours before, those who were in shock, or those who were in a hemodynamically unstable condition. The staff responded with a focused effort to optimize the management of these patients, resulting in zero mortality for emergency coronary artery bypass grafting during the following year. In the process, staff from all departments joined together in a more collaborative approach to the cardiac surgery program. Outcomes data can be useful for effecting program improvement if comparable norms and open access for flexible analysis are available.

Nuclear scan-guided rib biopsy.

The bone scan is a sensitive screening device that is frequently used to stage the condition of patients with known or suspected malignant disease. Abnormal findings on bone scan are associated with corresponding normal findings on radiographs in approximately 50% of cases. Definitive tissue diagnosis of the bone lesion is often needed to determine optimal therapy, but localization of the lesion is imprecise unless it is palpable. Use of the nuclear scan to localize and mark the rib enhances the precision of the biopsy procedure. Thirty-three consecutive patients with cancer who had bone scans suggestive of rib abnormalities underwent nuclear scan-guided biopsy. Each patient had a repeat localizing scan with a maximum permissible dose of technetium 99m radionuclide on the day of the planned biopsy. The site of abnormality was marked with methylene blue injected into the skin overlying the lesion and down to the periosteum at the specific site. The patient was then taken to the operating room and the marked area was excised through a small incision. Pathologic abnormality was identified in all but one of the resected specimens, an accuracy rate of 97%. Despite a presumed or proved diagnosis of cancer in 33 patients, 16 specimens (48%) were benign. There were no complications associated with this technique, which reduces the morbidity and increases the precision of rib biopsy.

End-systolic elastance as an evaluation of myocardial function in shock.

The response of the heart during sepsis has been studied in human and animal models with disparate results. Because sepsis induces marked peripheral vascular changes, to accurately determine the cardiac response, one must use indices of cardiac performance that are independent of loading conditions and heart rate. The slope of the end-systolic pressure-diameter relationship (ESPDR) has been proposed to have these properties. Pigs were equipped with transducers to measure left ventricular pressure, internal short axis diameter (D), and pulmonary and coronary artery blood flows. After 7-10 days of basal observations, an endotoxin-loaded osmotic pump delivering endotoxin at 10 micrograms/kg/hr was implanted into each pig. Fourteen pigs were so treated, and 4 expired before 24 hr of endotoxin challenge. In the surviving pigs, cardiac output, heart rate, dP/dtmax, and peak systolic pressures were elevated. However, both ESPDR and % D shortening were both significantly depressed. These data suggest that the cardiac response to chronic endotoxin challenge includes a loss of inotropic state as indicated by the load-insensitive indicator, ESPDR, and confirmed by depressed % D shortening. One possible mechanism for reduced inotropic state during endotoxin challenge could be the loss of coronary perfusion. The surviving endotoxin-challenged pigs demonstrated a significant increase in coronary perfusion while stroke work remained unchanged, suggesting that depressed cardiac inotropic state during endotoxin challenge was not caused by reduced coronary blood flow. Rather, the myocardium was relatively overperfused. Another possible mechanism for the loss of cardiac inotropism during endotoxin challenge may be endotoxin-induced generation of reactive oxygen free radicals. This possibility was tested by measuring total cardiac gluthathione, a cellular component depleted by oxidant stress. Endotoxemia reduced these levels 50%. These results suggest that cardiac injury may be mediated by the generation of reactive oxygen free radicals. Further study will determine if this mechanism participates in the loss of cardiac inotropism during endotoxin challenge.

Variation in end-systolic pressure-diameter relationship using dP/dtmin or P/Dmax as a definition of end-systole in chronic endotoxemic pigs.

We have reported that cardiac inotropism is reduced in various shock states, most recently during chronic endotoxemia (Lee et al.: American Journal of Physiology 254:H324-H330, 1988) [1]. We based this conclusion upon the alterations observed in the slope of the end-systolic pressure-diameter relationship (ESPDR). Recently, Dietrick and Raymond (Dietrick and Raymond: Surgical Infection Society, 7th Annual Meeting, May, 1987, p 83) [2] have reported that the slope of the end-systolic pressure-wall thickness relationship was augmented in the early stages of sepsis and depressed immediately prior to expiration. One major difference between our studies is the definition of end-systole; we used the time when the ratio of pressure-to-diameter (P/D) in the left ventricle is maximal (P/Dmax), whereas they used the time when the first derivative of pressure is minimal (dP/dtmin). In order to determine if the discrepancy between our conclusions could be explained by the differing definitions of end-systole, data from previous studies were reanalyzed, and the slope of the pressure-diameter relationship at P/Dmax and at dP/dtmin was calculated. Pigs were equipped with instruments to measure left ventricular pressure, short axis diameter, and ECG. Observations during the basal state were obtained 3-7 days after surgery. Chronic endotoxemia was induced by intravenous infusion of S. enteriditis endotoxin via an osmotic minipump at 10 micrograms/kg/hr. During the basal state, the value for the slope of ESPDR at dP/dtmin was lower than the value for the slope of ESPDR at P/Dmax, and there was a good correlation between the two values. During chronic endotoxemia, the slope of ESPDR at dP/dtmin did not change. However, the slope of ESPDR at P/Dmax decreased significantly suggesting that chronic endotoxemia reduced cardiac inotropism. This conclusion is supported by the findings that chronic endotoxemia reduced steady-state values of percentage diameter-shortening (an estimate of ejection fraction) and stable stroke work at significantly higher end-diastolic diameter. These data indicate that it is possible to calculate differing slopes of ESPDR from the same observations dependent upon the time during the cardiac cycle chosen as end-systole. More importantly, these data suggest that during chronic endotoxemia, ventricular relaxation dynamics may change so that postsystolic shortening becomes more prominent and therefore higher values for the slope of ESPDR using pressure and diameter values at dP/dtmin can be calculated.

Predonated autologous blood use in elective cardiac surgery.

The risks of homologous blood transfusion are well documented and recently increased with the emergence of acquired immunodeficiency syndrome. Preoperative autologous donation has been suggested to reduce these risks. This is a report concerning 104 consecutive adult autologous donors (group 1) who had an elective cardiac operation. A similar group of 111 patients operated on during the same period but without autologous blood donation was used for comparison (group 2). Both groups contained similar numbers of patients with coronary artery disease, valvular disease, and mixed lesions, and both had several patients with atrial septal defects. Group 2 patients (mean age, 67.8 years) were significantly older than group 1 patients (mean age, 58.9 years) (p less than 0.05). The mean donation in group 1 was 4.1 units, but 12 (11.5%) had to discontinue donations. Increasing angina in 10 (12.2%) of the 82 patients with coronary artery disease was the most common complication, and necessitated hospitalization in two instances. In 77 (75.5%) of the 102 group 1 patients who had operation and 23 (21%) of the 110 group 2 patients, no homologous blood products were required. Group 1 patients used significantly less homologous fresh frozen plasma (0.1 unit versus 0.97 unit; p less than 0.005) and packed red blood cells (0.6 unit versus 2.1 units; p less than 0.001) than group 2 patients. Group 1 patients received 3.3 and 3.1 units of autologous packed cells and plasma, respectively. No complications of autologous transfusion were seen. Predonation of autologous blood is an effective, safe method of reducing homologous blood requirements in elective cardiac operations, but it does carry some risk, especially in patients with coronary artery disease.

Patient-ventilator partitioning of the work of breathing during weaning.

The work of breathing and its division between the patient and the mechanical ventilator were studied during weaning of 5 post-operative surgical patients from Synchronized Intermittent Mandatory Ventilation. Work by the patient (WP) was estimated by integrating the product of flow and pressure over time during intervals when waveforms indicated patient effort; ventilator work (WV) was similarly estimated during positive pressure inspirations. The ratio of WP to the rate of work on the lungs (WL) increased progressively during weaning from 0.14 +/- 0.04 to 1.2 +/- 0.15 while WV/WL dropped from 1.31 +/- 0.08 to 0.13 +/- 0.11. Work on the lungs decreased during weaning. This was due in part to significant improvements in lung mechanics: resistance decreased from 9.9 +/- 0.9 to 6.1 +/- 1.6 cmH2O/1/s and compliance increased from 58 +/- 17 to 102 +/- 30 ml/cmH2O. The patient and ventilator work ratios, and the work of breathing quantify factors which may be directly useful to the clinician and to future systems to automate weaning.

Cardiac function and coronary flow in chronic endotoxemic pigs.

We have reported that myocardial inotropism was depressed in acute and chronic endotoxemia. One possible mechanism for this observation is that endotoxemia reduces myocardial perfusion and indeed, we observed reduced myocardial perfusion in acute endotoxemia. This study tested the hypothesis that reduced inotropism of chronic endotoxemia was accompanied by reduced coronary artery blood flow. Fifteen pigs were equipped with left atrial and ventricular catheters, circumflex coronary and pulmonary artery flow meters, left ventricular pressure transducer, and ultrasonic crystals in the anterior-posterior axis to measure internal short axis diameter by sonomicrometry. The pigs recuperated for 3 days before basal data were collected over the next 3-5 days. After at least 7 postoperative days, an osmotic pump containing Salmonella enteriditis endotoxin was implanted in 12 pigs. Endotoxin was delivered at 10 micrograms/hr/kg for 2 days, at which time the animals were sacrificed. Osmotic pumps containing sterile saline were implanted in 3 pigs. Eight of the 12 endotoxemic pigs survived; 4 died before the morning of the second day. The survivors exhibited elevated heart rate, peak left ventricular systolic pressure, and cardiac output. Inotropism was evaluated by calculating the slope of the end-systolic pressure-diameter relationship (ESPDR) and % diameter-shortening. ESPDR was significantly depressed on the second endotoxemic day, while % diameter-shortening was depressed on both endotoxemic days. Coronary artery blood flow was significantly elevated on both endotoxemic days, while cross-sectional stroke work was unchanged. Therefore, the ratio of coronary blood flow to stroke work increased on both endotoxemic days. Nonsurvivors exhibited reduced heart rate, cardiac output, peak left ventricular systolic pressure, ESPDR, and % diameter-shortening. Neither coronary artery blood flow nor flow-to-work ratios increased in this group. Sham endotoxemic pigs demonstrated no cardiac or hemodynamic changes over 3 days. These results indicate that depressed inotropism during chronic endotoxemia was not caused by reduced coronary blood flow; rather, the myocardium was relatively overperfused.

Changes in myofibrillar content and Mg-ATPase activity in ventricular tissues from patients with heart failure caused by coronary artery disease, cardiomyopathy, or mitral valve insufficiency.

Force development and shortening by cardiac muscle occur as a result of the interaction between actin and myosin within the myofibrillar lattice. This interaction is dependent upon intracellular ionized calcium and is controlled by the troponin-tropomyosin regulatory proteins situated along the actin filament. In this study, we compared the myofibrillar content and myofibrillar Mg-ATPase activity of normal human ventricular muscle with that of ventricular muscle from patients in end-stage failure caused by coronary artery disease or cardiomyopathy and ventricular muscle from patients with heart failure due to mitral valve insufficiency. The results show that the amount of myofibrillar protein (mg/g wet wt ventricle) in hearts in end-stage failure (coronary artery disease and cardiomyopathy) is significantly lower compared with normal hearts and hearts in failure due to mitral valve insufficiency. However, the Mg-ATPase activity of myofibrils from hearts in both end-stage failure and failure due to mitral valve insufficiency is significantly lower compared with myofibrils from normal hearts. The data suggest that the reduction in the amount of myofibrillar protein in ventricular tissue is a pivotal event that may be responsible for the progression of heart disease to the point of end-stage failure.

Left ventricular function during chronic endotoxemia in swine.

Cardiac performance was studied in 15 chronically instrumented awake pigs during chronic endotoxemia (CET) induced by intravenous infusion of low doses of endotoxin. We sought to test the hypothesis that left ventricular inotropic state was depressed during the stage of chronic endotoxemia when cardiac output, heart rate, and left ventricular systolic pressures are elevated, termed "hyperdynamic sepsis". Left ventricular pressure, internal short axis diameter (SAX), pulmonary artery blood flow, and electrocardiogram were recorded. After initial surgical preparation, each pig was observed for 7-10 days to measure representative basal values. Each pig was then reoperated on day 10 to implant an endotoxin-loaded osmotic pump whose output, infused Salmonella enteritidis endotoxin at a rate calculated to be 10 micrograms.kg-1.h-1 for up to 7 days. Cardiac performance was monitored by measuring dP/dt, heart rate, stroke volume, end-diastolic diameter, percent change in diameter, and the slope of the end-systolic pressure diameter relationship (ESPDR). Data from the basal days were pooled and compared with the data obtained each day of CET by two-way analysis of variance. Ten of 15 pigs survived more than 2 days of CET; 5 died before the morning of the second CET day. The surviving pigs demonstrated elevated systolic pressures, left ventricular maximum rate of pressure development (+dP/dtmax and -dP/dtmax), heart rates, and cardiac output. However, both ESPDR and percent SAX shortening were significantly depressed during both CET days. We conclude that cardiac inotropic state is depressed during hyperdynamic sepsis as indicated by the load-independent parameter ESPDR and confirmed by depressed percent SAX shortening.