Developing performance indicators for cardiac surgery: a demonstration project in Victoria.

Six Victorian cardiac surgical units pooled data in order to undertake a demonstration project aimed at developing performance indicators to assess outcomes following cardiac surgery. The outcome of the project was an indicative report for the purpose of monitoring surgical performance indicators in a format suitable for: (i) the general public; (ii) the Victorian State Government; and (iii) the participating units and surgeons. Each participating cardiac surgical unit had an existing database used for recording information from each procedure. A request was made to each unit to extract a subset of data from all cases entered over the past 5 years. The proposed list of performance indicators included surgical mortality (within the period of admission for surgery), complication rates (including sternal infection, postoperative myocardial infarction, postoperative stroke, haemorrhage requiring return to theatre), and length of hospital stay. A model was developed from the data and used to provide risk-adjusted measures of hospital performance. Cases from five cardiac surgical units (n = 10 715) were included in the final analysis. A risk-adjusted model (including age, sex, diabetes, hypertension, smoking, procedure type, urgency of procedure) was developed for surgical mortality. Performance indicators for coronary artery bypass graft surgery, including mortality, sternal infection rate and length of hospital stay are presented. From the available data, performance indicators for cardiac surgery in Victorian hospitals compared favourably with international benchmarks. This project has demonstrated that prospective data collection using a standardised system could readily produce local risk-adjustment models for cardiac surgery to aid in developing appropriate performance indicators.

Exercise hemodynamic performance of the pulmonary autograft following the Ross procedure.

BACKGROUND AND AIMS OF THE STUDY: The Ross procedure, in which the aortic valve is replaced with the patient's own pulmonary valve (pulmonary autograft), is considered an excellent alternative for younger patients requiring elective aortic valve replacement. Although resting pulmonary autograft hemodynamics are excellent, exercise hemodynamic data are lacking. The study aim was to measure the hemodynamic performance of the pulmonary autograft with exercise Doppler echocardiography (DE). METHODS: Twenty-four Ross procedure patients (20 males, four females; mean age 46 +/- 11 years) were studied at 25 +/- 14 months after aortic valve replacement with a pulmonary autograft. Patients had baseline supine DE to measure the maximum velocity (Vmax), and the peak and mean pressure gradient across the pulmonary autograft. Effective orifice area was calculated from the continuity equation and indexed to body surface area (EOAi). Patients then underwent symptom-limited upright bicycle exercise with supine DE repeated immediately on stopping exercise. For comparison, 10 normal controls (age 41 +/-10 years) and five mechanical aortic valve patients (mean age 55 +/- 10 years) were studied. RESULTS: At rest: Ross procedure patients had similar Vmax (1.2 +/- 0.2 m/s), peak gradient (6 +/- 2 mmHg), mean gradient (4 +/- 1 mmHg) and EOAi (1.7 +/- 0.4 cm2/m2) to those of normal controls. Mechanical-valve patients had significantly higher Vmax (2.5 +/- 0.2 m/s, p <0.001), peak gradient (25 +/- 4 mmHg, p <0.001) and mean gradient (14 +/- 3 mmHg, p <0.001) than Ross patients and normal controls. At exercise: Ross procedure patients had similar Vmax (1.8 +/- 0.4 m/s versus 2.1 +/- 0.2, p = NS), peak gradient (14 +/- 6 mmHg versus 17 +/- 4, p = NS) and mean gradient (8 +/- 4 mmHg versus 10 +/- 2, p = NS) to normal controls, with no significant change in EOAi. Mechanical-valve patients had significantly higher Vmax (3.4 +/- 0.3, p <0.001), peak gradient (48 +/- 7 mmHg, p <0.001) and mean gradient (30 +/- 5 mmHg, p <0.001) than Ross patients and normal controls. CONCLUSIONS: Aortic valve replacement using the Ross procedure provides excellent hemodynamic results at rest and on exercise, with DE parameters indistinguishable from those of normal controls. This study provides further support for the use of the Ross procedure as a preferred method of aortic valve replacement in younger patients.

Ross procedure. Inserting the autograft using a fully supported root replacement method; techniques and results.

BACKGROUND AND AIM OF THE STUDY: The root replacement (RR) method for insertion of the pulmonary autograft (PA) has resulted in improved immediate aortic valve competence. However, the unsupported pulmonary artery wall is thinner, more elastic, and thus more prone to dilatation than the normal aortic root. This might predispose to late aortic regurgitation (AR) due to splaying of the aortic commissures, similar to the mechanism of AR in Marfan's syndrome. METHODS: A fully supported root replacement (FSRR) method was designed and implemented in 78 patients, with preservation of the aortic root and proximal ascending aortic remnant fully to surround and support the PA root. Additional aortic annulus reduction was performed in 29 patients, and adjustment of the sinotubular diameter in 27. RESULTS: Seventy-eight patients were analyzed with sequential Doppler echocardiography. The maximal neoaortic sinus diameter remained constant for up to three years after surgery (mean 34.3 +/- 4.0 mm) compared with before surgery (35.2 +/- 4.0 mm). There was one early death, no late deaths or reoperations, and at last follow up AR was nil/trivial in 72% of patients, mild in 27% and moderate in 1%. There was no progression of AR over four years' follow up. By comparison, four patients previously underwent unsupported RR for insertion of the PA; in these patients, mean neoaortic sinus diameter increased significantly from 31 +/- 6 mm to 41 +/- 3 mm at three years after surgery (p = 0.005). CONCLUSIONS: Insertion of the PA using a FSRR method prevents dilatation of the neoaortic sinuses and sinotubular junction without need for prosthetic material, and provides similar results to conventional RR with regard to aortic valve competence. Retaining the advantages of RR in this manner and maintaining aortic root size may prove valuable in the longer term.

Minimally invasive coronary artery bypass surgery without cardiopulmonary bypass.

OBJECTIVE: To report on the initial results of minimally invasive direct coronary artery bypass surgery (MIDCAB) without cardiopulmonary bypass. This is a potential alternative to conventional coronary artery bypass graft surgery, recently introduced to Australia. DESIGN: Prospective survey of patient data. SETTING: Royal Melbourne Hospital campus, Melbourne, Victoria, July 1996 to June 1997. PATIENTS: The first 23 consecutive patients to have a MIDCAB procedure without cardiopulmonary bypass via a small left thoracotomy. The left anterior descending coronary artery was revascularised with the left internal mammary artery. All patients had either recurrent stenosis after previous angioplasty or anatomy unsuitable for angioplasty. OUTCOME MEASURES: Operative morbidity and mortality; graft patency; and patient symptom relief and reoperation rates. RESULTS: Mean age of patients was 57.9 years (range, 29-81), and mean follow-up was 4.0 months (range, 1-10). There was no operative mortality, cardiac infarction or stroke. Mean postoperative stay in the Intensive Care Unit was 30.7 hours and in hospital, 5.3 days. Only one patient needed a blood transfusion (packed red cells). Initial patency of the grafts was confirmed by either angiography (five) or continuous pulse-wave Doppler (23). One patient underwent angioplasty for a stenosis distal to the anastomosis, and two patients (9%) required reoperation for recurrent angina. CONCLUSIONS: MIDCAB can be performed safely, and patient recovery is faster than after conventional coronary artery surgery.

Insertion of the pulmonary autograft as an inclusion cylinder--modifications to existing techniques.

Ten pulmonary autograft procedures have been performed (mean age 27, range 19-40 years) all using root replacement techniques. In six operations the inclusion cylinder method was performed, as first preference. Because of root asymmetry associated with the congenitally deformed aortic valves, modifications to the standard procedure were necessary in four out of the six patients. In one, the aortic annular dimension was reduced by a Dacron collar inserted around the proximal autograft suture line. In another the sinotubular dimension was trimmed by wedge excision of part of the non-coronary sinus and ascending aortic tailoring. In the remaining two patients, pericardial patch enlargement of the lower ascending aorta was employed to increase the sino-tubular diameter. In the remaining four patients, full free-standing root replacement was performed as second best option because of inexperience (early in the series), or coronary artery anomalies, and more severe degrees of root symmetry (later in the series). All 10 patients are alive and have been recently reviewed with Doppler echocardiography (mean follow up 14 months, range 4-29 months). Mean aortic gradient was 5 mmHg, aortic regurgitation minimal (trivial seven, mild three), good pulmonary homograft function and normalization of left ventricular function were documented. In summary, we aim to perform the inclusion cylinder technique in most cases, and it is presumed that this may result in better long term aortic valve competence.

Excellent long-term results of cardiac valve replacement with the St Jude Medical valve prosthesis.

BACKGROUND: A retrospective review of the performance of the St Jude Medical valve prosthesis over a 12-year period was undertaken. METHODS AND RESULTS: Between 1978 and 1990, 696 St Jude Medical valve prostheses (351 isolated aortic, 191 isolated mitral, 64 double aortic/mitral) were implanted into 616 patients (mean age, 55 years). Concomitant coronary artery graft surgery was performed in 18%. During mid-1991, follow-up was conducted, yielding a 97% completion (16 lost), for a total of 3075 cumulative patient-years (mean, 5.0 years). Early (30-day) mortality rates were 3.1%, 5.2%, and 6.4% after aortic, mitral, and double valve replacements, respectively. The 5-year actuarial survival rates were 94.1 +/- 1.3%, 85.8 +/- 2.7%, and 86.3 +/- 4.6% and those for 10 years were 89.6 +/- 1.9%, 72.9 +/- 6.1%, and 83.0 +/- 5.4%, respectively. There were no structural valve failures. Anticoagulant-related hemorrhage was the most common valve-related complication (28 events, 0.9% per patient-year), with three being fatal. Thromboembolism (13 events) occurred at a rate of 0.5% per patient-year (0.6% aortic, 0.3% mitral, 0.3% double). All of the nonstructural deteriorations (10 events, 0.3% per patient-year) were paravalvular leaks, including 3 aortic valve replacements in patients who developed hemolytic anemia. There was 1 valve thrombosis (0.03% per patient-year). Patients undergoing coronary artery graft surgery had lower survival and higher complication rates than patients without coronary artery graft surgery. There was a significant improvement (P < .001) in New York Heart Association functional class for the entire patient population. CONCLUSIONS: The St Jude Medical valve prosthesis has performed well in all positions over a 12-year period, with an acceptably low incidence of valve-related complications.

Late functional results after surgical closure of acquired ventricular septal defect.

To assess the longer term outlook for patients who have undergone surgery for acquired (postinfarction) ventricular septal defect, we interviewed and studied 60 survivors from a single regional cardiac center between 3 and 144 months after the operation. Including the patients who died within 1 month of the operation, the 5-, 10-, and 14-year survivals (with standard errors) were 69% (65% to 74%), 50% (44% to 57%), and 37% (27% to 46%). Eighty-two percent of patients were in New York Heart Association class I or II. Ten patients (17%) had a persisting but not hemodynamically significant ventricular septal defect. Mean left ventricular ejection fraction was reduced at 0.39 (standard deviation 0.15), but this did not correlate with either New York Heart Association class or exercise tolerance. Twenty-eight patients (47%) had asymptomatic arrhythmias (17 with ventricular premature beats). Angina and other medical problems were not prevalent.

Pulmonary artery balloon counterpulsation for intraoperative right ventricular failure.

Two cases of severe low cardiac output and right ventricular failure after coronary artery bypass grafting necessitated pulmonary artery balloon counterpulsation after intraaortic balloon pumping and maximal inotropic/pressor support were unsuccessful in maintaining a satisfactory cardiac output. Hemodynamic improvement was sufficient to allow removal of the device 2 and 3 days postoperatively, with survival in 1 patient. Pulmonary artery counterpulsation is less morbid in comparison with other mechanical methods of right ventricular support and is applicable in right ventricular failure of intermediate severity.

Surgical treatment for infarct-related ventricular septal defects. Improved early results combined with analysis of late functional status.

A total of 101 patients (mean age 64.9 years) underwent surgical correction of postinfarction ventricular septal defect at this institution over a 15-year period (1973 to 1988). The overall early mortality rate was 20.8%, although the most recent experience with 36 patients (January 1987 to October 1988) has seen this decline to 11.1%. Factors found to influence early death significantly, when analyzed univariately, were as follows: (1) site of infarction (anterior 12.1%, inferior 32.6%, p = 0.02); (2) time interval between infarction and operation (less than 1 week 34.1%, greater than 1 week 10.5%, p = 0.008); (3) cardiogenic shock (present 38.1%, absent 8.5%, p = 0.001). Nonsignificant variables included preoperative renal function, age, and concomitant coronary artery bypass, although older age (greater than 65 years) became significant when examined in a multivariate fashion. Of the 80 hospital survivors, eight were subsequently found to have a recurrent or residual defect necessitating reoperation, with survival in seven. Late follow-up is 99% complete and reveals an actuarial survival rate for 100 patients of 71.1% at 5 years (95% confidence interval 60.6 to 80.0), and 40.0% at 10 years (95% confidence interval 21.7 to 58.4). A significant recent change in policy of not using coronary angiography in patients with a ventricular septal defect caused by anterior wall infarction has not resulted in any increase in either the early mortality or in the late prevalence of angina. The functional status of 38 surviving patients has been analyzed by a graded treadmill exercise protocol, whereas left ventricular functional assessment was by nuclear scan with additional information on mitral valve function by echocardiogram. Color Doppler flow mapping has been used to determine the presence of a residual defect. Most late survivors have limited exercise tolerance related to both cardiac and noncardiac factors. Left ventricular function is moderately impaired (mean ejection fraction = 0.39). However, many patients are elderly and have adapted to their residual symptoms without significant changes in life-style.

Intramural haematoma of the oesophagus complicated by perforation.

A case is presented where spontaneous bleeding into the wall of the oesophagus was exacerbated by anticoagulant therapy. Subsequently, iatrogenic, full-thickness perforation of the oesophagus occurred during endoscopy and, ultimately, oesophagectomy was required. If this condition is suspected on clinical grounds, the most appropriate sequence of investigations would appear to be contrast radiography in the first instance with cautious use of oesophagoscopy.

Brucellosis complicated by aortic valve endocarditis.

A 30 year old veterinary surgeon developed a febrile illness with serological evidence of Brucellosis. He was known to have aortic valve disease and during the course of the illness, the clinical features of endocarditis became evident, with a vegetation visible echocardiographically on the aortic valve. Because of persisting fever despite appropriate antibiotic therapy, aortic valve replacement with a viable cryopreserved allograft aortic valve was undertaken. Organisms consistent with Brucella species were demonstrated in the excised vegetation. The patient received a six week course of antibiotics and his post-operative course was uneventful.

Surgical excision of primary cardiac tumours in infancy.

Six cases of primary cardiac tumour have been operated upon in a 7 year period from 1 June 1979 until 1 June 1986. All patients were under 6 months of age at the time of operation and two of the patients were in their first week of life. The principal indication for surgery was obstruction mainly at the right or left ventricular outflow tract level. More recently echo evaluation alone has been adequate to define the problem prior to surgery. Surgical excision has been performed without mortality or significant complications. In most cases resection has been complete, although in one case residual tumour has been left because of attachment of the tumour to vital structures. Follow-up of this case has not resulted in further surgery being required because of regrowth of the tumour. In one case, with co-existent congenital heart disease, the tumour was brought to notice after palliative systemic to pulmonary artery shunt had been performed. From the cardiac view point gratifying results have been obtained both in the short and long term following surgical resection. However, for patients with rhabdomyoma, later development of symptomatic tuberosclerosis should be anticipated in 50% of cases.