Prospective risk analysis in patients submitted to myocardial revascularization surgery

Abstracts

OBJETIVO: Análise estratificada de risco em Cirurgia de Revascularização Miocárdica (CRVM). MÉTODOS: Estudou-se, de forma prospectiva, 814 pacientes, aplicando-se dois índices prognósticos (IP): Parsonnet e Higgins Modificado. O IP Higgins foi Modificado por substituição da variável "valor do índice cardíaco" por "síndrome de baixo débito cardíaco", na admissão à Unidade de Terapia Intensiva (UTI). A capacidade discriminatória para morbimortalidade de ambos foi analisada através de curva ROC (receiver operating characteristic). Identificou-se, através de regressão logística, os fatores associados, de forma independente aos eventos. RESULTADOS: A taxa de mortalidade foi de 5,9% e a de morbidade, 35,5%. O IP Higgins Modificado, que analisa variáveis pré, intra-operatórias e variáveis fisiológicas na admissão à UTI, demonstrou áreas sob a curva ROC de 77% para mortalidade e de 67%, para morbidade. Por sua vez, o IP Parsonnet, que analisa somente variáveis pré-operatórias, demonstrou áreas de 62,2% e 62,4%, respectivamente. Doze variáveis caracterizaram-se como fatores prognósticos independentes: idade, diabete melito, baixa superfície corpórea, creatinina (>1,5 mg/dl), hipoalbuminemia, cirurgia não-eletiva, tempo prolongado de circulação extracorpórea (CEC), necessidade de balão intra-aórtico pós-CEC, síndrome de baixo débito cardíaco na admissão do paciente à UTI, freqüência cardíaca elevada, queda do bicarbonato sérico e alargamento do gradiente alvéolo-arterial de oxigênio nesse período. CONCLUSÃO: O IP Higgins Modificado mostrou-se superior ao IP Parsonnet na estratificação de risco cirúrgico, salientando a importância da análise de eventos intra-operatórios e variáveis fisiológicas na admissão do paciente à UTI, quando da definição prognóstica.

cirurgia de revascularização miocárdica; índices prognósticos; morbidade e mortalidade

ORIGINAL ARTICLE

Prospective risk analysis in patients submitted to myocardial revascularization surgery

Antonio Carlos Mugayar Bianco; Ari Timerman; Ângela Tavares Paes; Carlos Gun; Rui Fernando Ramos; Ronald Brewer Pereira Freire; César Nicolas Conde Vela; Antonio Aurélio de Paiva Fagundes Junior; Luís Cláudio Behrmann Martins; Leopoldo Soares Piegas

Instituto Dante Pazzanese de Cardiologia - São Paulo, SP - Brazil Conclusion

^{Mailing address}

ABSTRACT

OBJECTIVE: To perform a stratified risk analysis in Myocardial Revascularization Surgery (MRS).

METHODS: 814 patients were prospectively studied by applying two prognostic indexes (PI): Parsonnet and Modified Higgins. The Higgins PI was modified by substituting the variable "cardiac index value" by "low cardiac output syndrome" at the Intensive Care Unit (ICU) admission. The discriminatory capacity for morbimortality of both indexes was analyzed by ROC (receiver operating characteristic) curve. Logistic reaction identified the associated factors, independently from the events.

RESULTS: Mortality and morbidity rates were 5.9% and 35.5%, respectively. The Modified Higgins PI, which analyzes pre- and intra-operative and physiological variables at the ICU admission showed areas under the ROC curve of 77% for mortality and 67% for morbidity. The Parsonnet PI, which only analyzes pre-operative variables, showed areas of 62.2% and 62.4%, respectively. Twelve variables were characterized as independent prognostic factors: age, diabetes mellitus, low body surface, creatinine levels (>1.5 mg/dL), hypoalbuminemia, non-elective surgery, prolonged time of extracorporeal circulation (ECC), necessity of post-ECC intra-aortic balloon, low cardiac output syndrome at the ICU admission, elevated cardiac frequency, decrease in serum bicarbonate concentrations and increase of the alveolar-arterial oxygen gradient within this period.

CONCLUSION:The Modified Higgins PI showed to be superior to the Parsonnet PI at the surgical risk stratification, showing the importance of the analysis of intraoperative events and physiological variables at the patient's ICU admission, when prognostic definition is achieved.

Keywords: myocardial revascularization surgery, prognostic indexes, morbidity and mortality

The results of myocardial revascularization surgery (MRS) are influenced by the clinical characteristics of patients, as well as by the aspects that are inherent to the surgical procedure and extracorporeal circulation^1,2.

The prognostic definition, through pre-operative clinical parameters, is important for the surgical indication; however, it does not always reflect the prognosis, when the patient is admitted to the Intensive Care Unit (ICU). Such fact has called our attention to the possibility of applying the ICU Admission Score for Predicting Morbidity and Mortality Risk of Coronary Artery Bypass Grafting, developed in 1997 by DR Thomas Higgins at the Cleveland Clinic, which analyzes pre-operative and intra-operative factors and physiological variables when the patient is admitted at the ICU, and which has, in theory, a broader prognostic range³.

There was, however, a significant problem regarding its use for all patients submitted to MRS, represented by the variable "cardiac index value at the ICU admission", whose measurement required bedside hemodynamic monitoring.

In an attempt to simplify it and amplify its use, we substituted the variable represented by the cardiac index < 2.1 1.min^-1/m^-2, at the patient's admission to the ICU³, by "low cardiac output syndrome", characterized by arterial pressure < 60 mmHg; need for vasoconstrictor (noradrenaline) when trying to keep the mean arterial pressure > 60 mmHg or the need for dopamine at a dose > 6 µg/kg/min.

We then named this score "Modified Higgins prognostic index".

METHODS

The present study was developed at Instituto Dante Pazzanese de Cardiologia, between March 5, 2002 and June 3, 2003.

Two prognostic indexes (PI) were prospectively applied: the Parsonnet PI (Tables I and ^II)⁴ , which analyzes only pre-operative variables and the Modified Higgins³ PI (Tables III and ^IV).

Data collection was carried out by the main investigator during the pre-operative period through the perfusion record analysis and surgical description, and collection of the patient's physiological parameters in the first 20 minutes of his/her ICU admission.

Death due to all causes as well as all severe complications³ were considered dependent variables, in addition to those with a lower prognostic significance such as atrial fibrillation (AF), bronchospasm, bypass surgery scar infection and others. The outcomes, mortality and morbidity were analyzed within a period that went from the surgical ward admission up to 30 days of post-operative evolution.

A sample of 814 patients was calculated based on the clinical experience and retrospective data, with an estimated incidence of deaths and complications of 15%, considering an absolute precision of 2.5% and a level of significance of 0.05%.

Sensitivity and specificity assessment of Parsonnet and Modified Higgins PI was carried out by ROC (receiver operating characteristic) curve. The discriminatory capacity of both indexes for mortality and morbidity was obtained by calculating the area under the curve.

All risk factors included in the prognostic indexes as well as others considered pertinent by the authors, were analyzed. Initially, the association of each variable with mortality was evaluated, and the statistically significant ones were identified (p< 0.100).

The qualitative variables were evaluated by Chi-square test and the quantitative ones by Student's T-test.

Subsequently, the selected variables (p< 0.100) were analyzed together through a logistic regression model (odds ratio), in order to determine their prognostic weight^5-9.

RESULTS

The studied patients' age ranged from 30 and 85 years, with a mean age (mean ± SD) of 61.8 ± 10.2 yrs. Of the 814 individuals studied, 557 (68.4%) were male and 34.6% were diabetic patients.

Regarding ejection fraction, 71.5% of them presented 50% or higher, 22.5% had between 30% and 49% and 6% had lower than 30%; the mean was 62%.

Of the procedures that were carried out, 677 (83.2%) were considered elective, 89 (10.9%) were characterized as urgent procedures and 48 (5.9%) were emergency procedures. Only two (0.2%) were characterized as post-accident emergencies at the hemodynamics laboratory.

Most of the surgical procedures consisted of revascularization of 2 to 4 coronary arteries. The data regarding the morbid events that took place in the ICU are shown in Table V.

The occurrence of only one morbid event meant an increase of 50% in hospital stay duration. Morbidity after hospital release was basically restricted to healing problems of the bypass surgery scar, such as localized edema, local inflammatory reaction or minor infectious processes.

ICU stay duration varied broadly, from one hour (early death) to 1,730 hours (chronic patients). In the absence of complications, this time is represented by a median of 44 hours. Hospital stay duration, when measured in days starting from the day of the surgical procedure, was in average 8 days, ranging from 1 day (early death) to a maximum of 173 days (chronic patients).

There were 48 deaths during the first 30 days of evolution, representing a mortality rate of 5.9%.

Low cardiac output syndrome was the main factor responsible for death in 26 (54.2%) patients. In 75% of the times, death occurred prematurely, within the first 48 hours of post-operative evolution. Acute transoperative myocardial infarction was the triggering factor in 30% of the cases.

The deaths that occurred during the late hospital phase were related to a wider range of factors. Its main representative was severe stroke or cerebral vascular accident. The extra-hospital deaths were represented exclusively by sudden death. The causes of death are shown in Table VI.

Table VII

Table VIII

At a second phase, the discriminatory capacity of Parsonnet PI and Modified Higgins PI was evaluated when predicting mortality and morbid events, through the ROC (receiver operating characteristic) curve.

Regarding mortality, the area under the ROC curve of the Modified Higgins PI presents a value of 77% (CI 95%, 68.2-86.0), whereas this value for the Parsonnet PI is 62.2% (CI 95%, 51.8-72.6).

The areas under the ROC curve have, respectively, the values of 67% and 62.4%, with confidence intervals ranging from 62.9 to 71.0 for the Higgins PI and 58.2 to 66.6 for the Parsonnet PI.

The x² test was used in search for an association between the qualitative variables and death, and Student's T test was used to compare the means of each quantitative variable value between the patients who died and those who survived.

With the objective of simplifying the analysis, the quantitative variables were categorized in two classes, according to a cutoff chosen based on clinical criteria, literature data and statistical analysis (ROC curve and x² test). This cutoff remained unaltered for serum albumin values in the pre-operative period and body surface area, and showed altered values regarding the original PI for age (> 70 yrs to 65 yrs), creatinine (1.9 mg/dL to 1.5 mg/dL), time of extra-corporeal circulation (> 160 min to > 100 min), cardiac frequency (> 100 bpm to > 110 bpm) and alveolar-arterial pulmonary gradient (> 250 mmHg to 320 mmHg). Therefore, they were re-categorized.

Among the variables with statistical significance, 12 were selected for analysis through logistic regression (Tables IX and ^X). The exclusion criterion of a variable was based on the variable's low incidence, its characterization as a post-operative complication or for being redundant in relation to the others. The following were excluded: catastrophic states, types of grafting used during the surgical procedure, need for sympathomimetic amines for a prolonged period in the post-operative period, acute perioperative myocardial infarction, and reoperation in the early phases of the post-operative period, major neurological complications, and persistent coma.

Variables such as serum albumin measurement, time of extra-corporeal circulation, low cardiac output syndrome at ICU admission and sodium bicarbonate measurement at the initial post-operative gasometry present a high prognostic value for mortality. When analyzing morbidity, these same variables are identified as having high statistical significance, adding to others such as the patient's age and cardiac frequency at the ICU admission.

DISCUSSION

A higher mortality and morbidity was observed in the urgency/emergency surgeries with a relative risk of 1.17 (relative risk of 1.17 (CI 95%, 0.55-2.46; p = 0.844) for mortality and 0.99 (IC 95%, 0.62-1.57; p = 0.955) for morbidity.

The surgeries that were carried out exclusively with venous graft (15.1%) presented a higher morbimortality than those carried out with at least one arterial graft (left internal thoracic artery¹⁰, right left internal thoracic artery and/or radial arteries). It has been shown, right in the early phases of the post-operative period, a beneficial effect of using arterial grafts, which is not restricted to the use of the left internal thoracic artery, but also includes the use of the right left internal thoracic artery and radial arteries.

Comparing the ROC curves for mortality and morbidity, we observed that the Modified Higgins PI has a higher discriminatory capacity when compared to Parsonnet PI, which is accentuated when mortality is analyzed.

Patients older than 65 yrs^1,11-13 and diabetic ones^14-17 have a poorer evolution prognosis. Another important risk factor is the low body surface (<1.72m²)^3,18,19. It generates a relative risk of 1.35 (CI 95%, 0.66-2.77; p=0.405) for mortality and 1.17 (CI 95%, 0.83-1.65; p = 0.360), for morbidity.

The importance of the presence of renal dysfunction prior to the surgical procedure as an independent prognostic factor was confirmed (serum creatinine levels > 1.5 mg/dL³). Similarly, pre-operative albumin levels are associated to a higher morbimortality, justified by its association to complications such as cardiac failure, acute respiratory failure, acute renal failure, longer periods of mechanical ventilation support, higher incidence of gastrointestinal bleeding and nosocomial infections^3,19-22. The hypoalbuminemic patients (17.7%) presented a relative risk of 1.37 (CI 95%, 1.07-5.35; p = 0.0360) for mortality and 2.67 (CI 95%, 1.74-4.10; p = 0.179), for morbidity.

The prolonged extra-corporeal circulation predisposes to post-operative complications, and the need for intra-aortic balloon reflects the presence of accentuated hemodynamic instability. It was observed that an extra-corporeal circulation time > 100 min characterized as an independent factor for morbimortality. The need for intra-aortic balloon (IAB), however, occurred in a small number of patients (2.1%). Mortality in this group was 47.8%, against 5% who did not need it (p< 0.001). This mortality rate above the one predicted in the literature (30 to 40%) is justified by the resistance to its early indication, making its use restricted to extremely severe situations.

The low cardiac output syndrome at the ICU admission can be transitory and respond to measures such as volemic replacement and short periods of inotropic support. Its duration beyond 24 hours rules out the possibility of being connected to the effects of stunned myocardium and systemic inflammatory response syndrome, which, in theory, have been reverted and is correlated with a high mortality (39%)²⁴.

The low cardiac output syndrome, identified at the ICU admission through the previously mentioned clinical criteria, has a high capacity to predict death, even when the syndrome is transitory. This group of patients presented a mortality of 27.3%, compared to 2.7% in the rest of the group. Therefore, low output manifestations, even when at its initial post-operative phases, generated a high relative risk of 5.78 (CI 95%, 2.58-12.96; p < 0.001) for mortality and 11.53 (CI 95%, 5.42-24.54; p < 0.001), for morbidity.

Higgins and cols³ considered as prognostic factors five physisological variables measured at ICU admission: arterial-alveolar pulmonary gradient, cardiac frequency, cardiac index, central venous pressure, and arterial blood sodium bicarbonate measurement. Of these, the cardiac index was substituted by the variable "low cardiac output syndrome" and the central venous pressure did not show statistical significance when analyzed.

The arterial-alveolar pulmonary gradient is increased during the initial phases of the post-operative period due to alterations such as increase of the pulmonary vascular permeability, accumulation of interstitial liquid and increase of intrapulmonary shunt, secondary to the appearance of atelectasias. The gradient was considered increased when its values were > 320 mmHg. It was demonstrated that patients with an increased arterial-alveolar pulmonary gradient presented a relative risk of 1.60 (CI 95%, 0.70-3.63; p = 0.266) for mortality and 0.93 (CI 95%, 0.66-1.30; p = 0.652), for morbidity.

Cardiac frequency was considered a risk factor when it was > 110 bpm and it confers a relative risk of 1.09 (CI 95%, 0.50-2.36; p = 0.833) for mortality and 1.40 (CI 95%, 0.96-2.03; p = 0.787), for morbidity²³. Tachycardia in the early phases of the pre-operative period generates an increment in energy consumption and oxygen consumption by the myocardium.

Serum bicarbonate levels < 19 mmol/L generated a relative risk of 3.09 (CI 95%, 1.46-6.56; p = 0.003) for mortality and 1.54 (CI 95%, 0.98-2.42; p = 0.059), for morbidity²⁵. Its decrease, ultimately, means the presence of tissue hypoxia.

Although it was not our primary objective, some post-operative complications showed to be important development interferences. Among them we cite persistent left ventricular dysfunction, characterized by the need for sympathomimetic amines for a period of time > 40 of post-operative phase; acute transoperative myocardial infarction^26-28, and the type 1 or major neurological complications, plegias, stupor and coma^29-32; respiratory failure³³ and the need for prolonged mechanical ventilation (time > 72 hs); acute renal failure^34-38; severe infections of the surgical wound, extensive pneumonias and sepsis^39-42. The patients who presented one of these complications has an elevation in their mortality rate (p< 0.001).

Other complications, such as bypass grafting site infections, urinary tract infections and atrial fibrillation⁴³ did not show interference in mortality; however, they are noteworthy due to fact that they increased hospital stay duration.

Global morbidity was 35.5%, which is a higher rate than that found in the literature that ranges from 15% to 30%^44,45. This is related to the fact that we identified post-operative complications as a whole. When we analyzed what was the significance of the occurrence of a single complication, regardless of its severity, we obtained an increase of 50% in hospital stay duration as a result.

Most of the studies in literature describe a hospital mortality of around 1%, which ranges from 1% to 6%. In a recent report, Almeida and cols⁴⁶ described a post-surgical mortality of 11.3%.

One can affirm, based on such data, that the predicted mortality for low-risk patients submitted to elective surgery is 1%. Nevertheless, it can depend on factors such as the presence of co-morbidities and transoperative complications.

The mortality rate identified up to 30 days of post-operative evolution in this study was 5.9%. Although high, it is acceptable, justified by the great incidence of complications and for a long period of post-operative follow-up, which lasted thirty days.

The mortality rates obtained for the different groups of risk stratification deserve a few comments.

Regarding the Parsonnet PI, a higher mortality is observed in patients stratified as being low-risk. In the other groups, it is observed a lower mortality than the predicted one. However, when the obtained stratified mortality is compared to the one predicted by the Higgins PI, we observe that the values are close to the predicted ones, with a slight increase in patients classified as being high-risk.

The higher mortality found in low-risk patients that were stratified by the Parsonnet PI, was due to the elevated incidence of cerebral vascular accident and coma in this group.

The higher mortality observed among the patients stratified as being high-risk by the Modified Higgins PI serve as an alert for the need to conduct a more aggressive initial approach in this group.

Finally, we observed that 0.9% of the deaths occurred in the surgical ward and were linked to severe left ventricular dysfunction. The early hospital deaths (39.6%) occurred all at the ICU and their main cause was the low cardiac output syndrome. Late hospital deaths (41.6%) were due to a broader range of factors and their main factors were: cerebral vascular accident, severe infections and uncontrollable arrhythmias. The extra-hospital deaths (0.2%) were exclusively linked to sudden death.

The main cause of deaths, when analyzed as a whole, was low cardiac output syndrome.

We conclude that the Modified Higgins PI was superior to the Parsonnet PI at surgical risk stratification. It is noteworthy, through a comparative analysis of both indexes and the individualized study of prognostic factors, the importance of intra-operative events and the physiological variables at the patient's ICU admission. Hence, the Modified Higgins PI characterizes as a concrete and viable alternative for the stratified risk analysis in patients submitted to MRS, when he or she is admitted to the ICU. It is worth mentioning that the current prognostic indexes used in general intensive care (APACHE, SOFA, MPM), present severe limitations to their use in patients who undergo MRS.

REFERENCES

1. Eagle KA, Guyton RA, Davidoff R et al. ACC/AHA Guidelines for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association. Task Force On Practice Guidelines (Committee To Revise The 1991 Guidelines For Coronary Artery Bypass Graft Surgery). J Am Coll Cardiol 1999; 34: 1262-346.

2. Morris DC, St. Clair D. Management of patients after cardiac surgery. Curr Probl Cardiol 1999: 166-228.

3. Higgins TL, Estefanous GF, Loop FD et al. ICU Admission score for predicting morbidity and mortality risk after coronary artery bypass grafting. Ann Thorac Surg 1997; 64: 1050-8.

4. Parsonnet V, Dean D, Bernstein AD. A method of uniform stratification of risk for evaluating the results of surgery in acquired adult heart disease. Circulation 1989; 79 (S-I): I-3-I-12.

5. Lwanga SK, Lemeshow S. Sample size determination in health studies: a practical manual. Geneva 1991; World Health Organization.

6. Guyatt G, Jaeschke R, Heddle N, Cook D, Shannon H, Walter S. Basic statistics for clinicians: 1. Hypothesis testing. Can Med Assoc J 1995; 152: 28-32.

7. Guyatt G, Jaeschke R, Heddle N, Cook D, Shannon H, Walter S. Basic statistics for clinicians: 2. Interpreting study results: confidence intervals. Can Med Assoc J 1995; 152: 169-73.

8. Guyatt G, Jaeschke R, Heddle N, Cook D, Shannon H, Walter S. Basic statistics for clinicians: 3. Assessing the effects of treatment: measures of association. Can Med Assoc J 1995; 152: 351-6.

9. Guyatt G, Jaeschke R, Heddle N, Cook D, Shannon H, Walter S. Basic statistics for clinicians: 4. Correlation and regression. Can Med Assoc J 1995; 152: 497-504.

10. Leavitt BJ, O'connor GT, Olmstead EM et al. Use of the internal mammary artery graft and in-hospital mortality and other adverse outcomes associated with coronary artery bypass surgery. Circulation 2001; 103: 507-12.

11. Poveda JJ, Calvo M, Llorca J, Bernal JM. Factores pre y perioperatorios determinantes de la mortalidad precoz em pacientes mayores de 75 años sometidos a circulación extracorpórea. Rev Esp Cardiol 2000; 3: 1365-72.

12. Ivanov H, Weisel RD, David TE, Naylor CD. Fifteen-year trends in risk severity and operative mortality in elderly patients undergoing coronary artery bypass graft surgery. Circulation 1998; 97: 673-80.

13. He GW, Acuff TE, Ryan WH, Bowman RT, Douthit MB, Mack JM. Determinants of operative mortality in elderly patients undergoing coronary artery bypass grafting. Emphasis on the influence of internal mammary artery grafting on mortality and morbidity. J Thorac Cardiovasc Surg 1994; 1089: 73-81.

14. Gardner TJ, Greene PS, Rykiel MF et al. Routine use of the left internal mammary artery graft in the elderly. Ann Thorac Surg 1990; 49: 188-93.

15. Bucerius J, Gummert JF, Walther F et al. Impact of diabetes mellitus on cardiac surgery outcome. J Thorac Cardiovasc Surg 2003; 51: 11-6.

16. Cole JH, Craver JM, Guyton RA et al. Outcomes of repeat revascularization in diabetic patients with prior coronary surgery. J Am Coll Cardiol 2002; 40: 1968-75.

17. Thourani VH, Weintraub WS, Stein B et al. Influence of diabetes mellitus on early and late outcome after coronary artery bypass grafting. Ann Thorac Surg 1999; 67: 1045-52.

18. Suma H, Takeushi A, Kondo K et al. Internal mammary artery grafting in patients with smaller body structure. J Thorac Cardiovasc Surg 1988; 96: 393-9.

19. Engelman DT, Adams DH, Byrne JG et al. Impact of body mass and albumin on morbidity and mortality after cardiac surgery. J Thorac Cardiovasc Surg 1999; 118: 866-73.

20. Rich MW, Keller AJ, Schechtman KB, Marshall Jr WG, Koucoukos NT. Increased complications and prolonged hospital stay in elderly surgical patients with low serum albumin. Am J Cardiol 1989; 63: 174-8.

21. Gibbs J, Cull W, Henderson W, Daley J, Hur K, Khuri SF. Preoperative serum albumin level as a predictor of operative mortality and morbidity: results from the National VA Surgical Risk Study. Arch Surg 1999; 134: 36-42.

22. Rady MY, Ryan T, Starr NJ. Clinical characteristics of preoperative hypoalbuminemia predict outcome of cardiovascular surgery. J Parenter Enter Nutr 1997; 21: 81-90.

23. Bojar MR. Cardiovascular management. In: Bojar MR. Manual of perioperative care in cardiac surgery. Boston. Blackwell Scientific Publications 1999: 135-229.

24. Gun C, Piegas LS. Síndrome de baixo débito no pós-operatório de cirurgia cardíaca. Rev Soc Cardiol Est São Paulo 2001; 21: 1023-32.

25. Bojar MR. Cardiovascular management. In: Bojar MR. Manual of perioperative care in cardiac surgery. Boston. Blackwell Scientific Publications 1999: 233-57.

26. Schaff HV, Gersh BJ, Fisher LD. Detrimental effect of perioperative myocardial infarction on late survival coronary artery bypass report from Coronary Artery Bypass Study (CASS). J Thorac Cardiovasc Surg 1984; 88: 972-81.

27. Chaitman, BR, Alderman EL, Sheffield LT. Use of survival analysis to determine the clinical significance of new Q waves after coronary bypass surgery. Circulation 1983; 67: 302-8.

28. Ramos RF, Oliveira GBF. Infarto agudo do miocárdio no pós-operatório imediato. Rev Soc Cardiol Est São Paulo 2001; 5: 956-63.

29. Atra M, D'Alessandro JR. Alterações do sistema nervoso central e periférico no pós-operatório de cirurgia cardíaca. Rev Soc Cardiol Est São Paulo 2001; 5: 964-9.

30. Ridderstolpe L, Ahlgren L, Rutberg H. Risk factor analysis and delayed cerebral complications after cardiac surgery. Cardiothorac Vasc Anesth 2002; 16: 278-85.

31. Bucerius J, Gummert JF, Borger MA et al. Stroke after cardiac surgery: a risk factor analysis of 16,184 consecutive adult patients. Ann Thorac Surg 2003; 75: 472-8.

32. Hogue CW, De Wet CJ, Schechtman KB, Davila-Roman VG. The importance of prior stroke for the adjusted risk of neurologic injury after cardiac surgery for women and men. Anesthesiology 2003; 98: 823-9.

33. Bianco ACM. Insuficiência respiratória no pós-operatório de cirurgia cardíaca. Rev Soc Cardiol Est São Paulo 2001; 5: 927-38.

34. Provenchere S, Plantefeve G, Hufnagel G et al. Renal dysfunction with normothermic cardiopulmonary bypass: incidence, risk factors, and effect on clinical outcome. Anesth Analg 2003; 96: 1258-64.

35. Ryckwaert F, Boccara G, Frappier JM, Colson PH. Incidence, risk factors, and prognosis of a moderate increase in plasma creatinine early after cardiac surgery. Crit Care Med 2002; 30: 1495-8.

36. Penta De Peppo A, Nardi P, De Paulis R et al. Cardiac surgery in moderate end-stage renal failure: analysis of risk factors. Ann Thorac Surg 2002; 74: 378-83.

37. Lucini N, Nasso G, D'alessandro C et al. Heart surgery interventions in chronic dialysis patients: short and long-term results. Ital Heart J 2002; 3: 746-52.

38. Bojar MR. Post-ICU care and other complications. In: Bojar M.R. Manual of perioperative care in cardiac surgery. Boston. Blackwell Scientific Publications 1999: 261-97.

39. Abboud C.S. Infecções em pós-operatório de cirurgia cardíaca. Rev Soc Cardiol Est São Paulo 2001; 15: 915-21.

40. Kohli M, Yuan L, David T, Gillis G, Comm B, Conly J. A risk index for sternal surgical wound infection after cardiovascular surgery. Infect Control Hosp Epidemiol 2003; 24: 17-25.

41. Leal-Noval SR, Jara-Lopez I, Garcia-Garmendia JL et al. Influence of erythrocyte concentrate storage time on post surgical morbidity in cardiac surgery. Anesthesiology 2003; 98: 815-22.

42. Gummert JF, Barten MJ, Hans C et al. Mediastinitis and cardiac surgery an updated risk factor analysis in 10,373 consecutive adults. J Thorac Cardiovasc Surg 2002; 50: 87-91.

43. Moreira EAR. Arritmias no pós-operatório de cirurgia cardíaca. Rev Soc Cardiol Est São Paulo 2001; 15: 941-52.

44. Hall RE, Ash AS, Ghalli WA. Hospital cost of complications associated with coronary artery bypass graft surgery. Am J Cardiol 1997; 79: 1680-5.

45. Becker RB, Zimmerman JE, Knaus WA et al. The use of APACHE III to evaluate ICU length of stay, resource use, and mortality after coronary artery bypass surgery. J Thorac Cardiovasc Surg 1995; 36: 1-11.

46. Almeida FF, Barreto SM, Couto BRGM, Starling CEF. Preditores de mortalidade hospitalar e de complicações per-operatórias graves em cirurgia de revascularização do miocárdio. Arq Bras Cardiol 2003; 80: 41-50.

Received on 12/22/03

Accepted on 03/04/05

Mailing address

Antonio Carlos Mugayar Bianco

Rua Pelotas, 323/113

04012-001 - São Paulo, SP - Brazil

E-mail: [e-mail]

Received on 12/22/03

Accepted on 03/04/05

Publication Dates

• Publication in this collection
  07 Nov 2005
• Date of issue
  Oct 2005

History

• Accepted
  04 Mar 2005
• Received
  22 Dec 2003