INTRODUCTION

Kidney failure is one of the most important risk factors for complications after both acute ischemic cardiac events [1-4] and in cardiac surgery, causing longer hospital stays and higher in-hospital mortality [5,6]. In coronary artery bypass grafting (CABG), the previous presence of renal dysfunction is an independent predictor of operative mortality [5.7 to 12].

Plasma creatinine (PCr) alone is not sufficiently precise to identify renal dysfunction - in large population samples, up to 17% of patients with "normal" PCr show significant impairment of renal function [6,13-15].

The measurement of glomerular filtration rate (GFR) by 24-hour endogenous creatinine clearance (eCrCl) provides accurate and useful results in the evaluation of renal function in clinical practice. However, this method requires an extended period of time for obtaining the urine and is also subject to errors such as those arising from incomplete collection of urine in the period. Thus, some equations were developed to estimate creatinine clearance using only the values †of PCr, body weight, age and gender [5,13,16-20]. Because of its simplicity, one of the most widely used equations is that proposed by Cockcroft & Gault (CG) [21]. Corrections for obesity or overweight were developed for this formula, in order to minimize the overestimation of CrCl [22], but these calculations are often not normally used.

The main objective of this study was to compare patients with PCr d" 1.5 mg/dL (133 micromol/L) who had undergone CABG, distributing them into two groups according to estimated creatinine clearance, normal or reduced, and finding differences rates of in-hospital mortality, total hospital length of stay (THLS) and postoperative hospital stay (POHS). Secondly, we also compared these findings with those of other patients (reduced CrCl and PCr > 1.5 mg / dL).

METHODS

This is a historical cohort study. All patients undergoing CABG between January 1996 and June 2004, at the Institute of Cardiology of Rio Grande do Sul were assessed for inclusion. Surgery without cardiopulmonary bypass and other surgical procedure simultaneous with CABG (such as valve surgery, aneurism repair, ventricular repair, carotid endarterectomy) were considered exclusion criteria. Pre-, intra- and postoperative data were collected from medical records of each patient using a standardized protocol. Quality control of collected data was performed on random samples of records for each month of all nine years included in the study.

MÉTODOS

Data related to several variables were recorded, such as gender, age, weight, height, PCr, dates of hospitalization, surgery and hospital discharge (allowing to establish the total length of hospital stay [TLHS] - from admission until hospital discharge; time and postoperative hospital stay [POHS] - surgery until hospital discharge). Mortality (all cause) was considered as the one during the hospitalization. PCr was measured by the same method and same laboratory. In case of more than one measure of PCr to the same patient, it was considered the last value before surgery. As the CrCl estimated by CG equation (CrCl_GG) does not consider any correction for body surface area (BSA) standard of 1.73 m², the following calculation was included in our formula: [CrCl/m² = 1.73 (mL/min/1.73 m²): ((140 - age (years)) × weight (kg))/(72 × CPR (mg/dL)) x (1.73 (m²)/SC (m²)) (0.85 if female)]. And as CrCl_GG overestimates GFR in patients with overweight or obese, the correction formula proposed by Saracino et al. [22] was applied to those with body mass index (BMI) greater than 25 kg/m2 - CrCl x (1.25 - {0.012 x BMI}). The BSA was calculated using the formula of Dubois - [BS (m²) = 0.007184 × height (cm) 0725 × weight (kg) 0425] [23]. The end result of all these calculations was called CrCle / 1.73 m².

Patients with PCr < 1.5 mg/dL were divided into two groups: eCrCl/1.73 m² > 60 mL/min - normal renal function Group A - or <60 mL/min - reduced kidney function, Group B ( according to the classification of chronic kidney disease stage 3 of the National Kidney Foundation [24]). All other patients had PCr> 1.5 mg/dL and eCrCl/ 1.73 m2 <60 mL / min (Group C).

A ROC curve for each gender was designed to establish the cutoff value of PCr required to detect an eCrCl/1.73 m²<60 mL/min. Another ROC curve was established to compare PCr and eCrCl/1.73 m² in order to predict mortality. A dispersion graph was created with all values † eCrCl/1.73 m² with their respective values †of PCr. Statistical analysis was performed with SPSS for Windows, version 16.0 (SPSS Inc., Chicago, IL, USA). Tukey test was used for paired comparisons of data between the three different groups.

The study was approved by the Research Ethics Committee of the Institute of Cardiology of Rio Grande do Sul and by the Research Ethics Committee of Lutheran University of Brazil.

RESULTS

In the 102 months of analysis, 4,765 patients had undergone CABG with cardiopulmonary bypass, of whom 4688 (98.4%) had all the necessary data for calculating the eCrCl/1.73 m².

Age (mean �± SD) of the study population was 61.2 �± 10.1 years, with a prevalence of 71.5% males. Table 1 shows the main characteristics of all patients in each group.

Among patients with PCr < 1.5 mg/dL, the prevalence of eCrCl/1.73 m2 < 60mL/min was 27.8%.

Maximum PCr of 1.5 mg/dL as the cutoff point had a sensitivity of 18.9% and a specificity of 100% to detect a eCrCl/1.73 m2 <60 mL/min, with a negative predictive value of 72.2 % and a positive predictive value of 100%. Table 2 and Figures 1 and 2 show the coordinates of sensitivity and specificity of the ROC curve for PCr values †to detect eCrCl/1.73 m2 <60 mL/min for both genders.

A graph of dispersion (Figure 3) was established in order to visually demonstrate the possible values †of eCrCl/1.73 m2 for the same value of PCr.

Total length of hospital stay (TLHS) and length of postoperative hospital stay (POHS)

Patients in groups B and C had TLHS and POHS significantly higher than group A (P <0.05, Tukey test) - Figure 4.

In-hospital mortality

The rates of in-hospital mortality (from any cause) in groups A and B were 2.8% and 5.8% respectively, while in group C was 11.6%.

Group B patients had a relative risk (RR) of in-hospital mortality of 2.09 (CI95% :1,54-2, 84) compared to group A, while those in group C had an RR of 4, 18 compared to group A (CI95%: 2.86 to 6.12) - Figure 5.

DISCUSSION

The study included all patients undergoing CABG with cardiopulmonary bypass regardless of the criterion for elective, urgent or emergency, and the rate of in-hospital mortality was 4.1%. This finding is similar to those obtained by Roques et al. [25] and Vogt et al. [26]. Roques et al. [25], whose database was also used to develop the EuroSCORE and included 19,030 patients from 128 surgical centers in Europe, found overall in-hospital mortality of 4.8%. And Vogt et al. [26], when studied data from 10,000 consecutive patients from 81 cardiac surgery centers in Germany, found a mortality rate of 3.91% in 30 days postoperatively.

Our data showed a significant increase in TLHS and POHS related to the finding of reduced renal function (Figure 4), as previously observed by Witczak et al. [8], when studying patients with chronic renal failure undergoing major cardiovascular surgical procedures. Our patients with PCr < 1.5 mg/dL, but with eCrCl/1.73 m² <60 mL/min (Group B) had significantly higher TLHS (2.9 days longer) and POHS (1.8 more days) than those in Group A. Comparing the group C (those with arguably compromised kidney function) with group A, we find even greater increase of TLHS and POHS (6.7 and 4.1 days longer, respectively).

Duncan et al. [14] used the CG equation to estimate the CrCl in the general population of 2781 patients who had PCr determined for clinical purposes and found that 14% of patients could be defined as belonging to what we consider group B. In our study, we found similar results using the equation CG alone. However, after adjusting for body surface area of †1.73 m2 and applying the Saracino formula for overweight and obesity, the prevalence of patients with PCr < 1.5 mg/dl and with impaired renal function increased to 27.8%.

In a previous study of 1,495 patients in our city, Burmeister et al. [15] tested the sensitivity and specificity of a PCr of 1.2 mg/dL to detect an eCrCl <60 mL/min through the CG equation and obtained 49.1% and 95.2%, respectively. With a cutoff point of PCr 1.5 mg/dL, sensitivity of 18.9% and a specificity of 100%. Therefore, a patient with PCr above 1.5 mg/dL certainly has kidney failure, although we can not rule out some degree of renal dysfunction when the creatinine is equal to or less than 1.5 mg / dL due to its low sensitivity. And while values †greater than 1.2 mg/dL are suggestive of renal dysfunction, such values are unable to fully confirm or rule out this hypothesis. In our study, evaluating the coordinates of the ROC curve constructed to detect an e CrCl <60mL/min by PCr, we found that a properly balanced cutoff would be 0.91 mg/dL (sensitivity = 85.2% and specificity = 84.4%) for women and 1.11 mg/dL (sensitivity = 86.0% and specificity 85.9%) for men.

Wijeysundera et al. [13] presented a graph in which the dispersion of the CrCl for the values †of PCr was higher than in our chart (Figure 3), which is certainly due to the fact that they considered the CG equation alone, without adjustment for surface body or for overweight or obesity. We can also observe patients already presenting a CrCl less than 60 mL/min with a PCr of 0.79 mg/dL, a value that, without the use of a formula for estimating the CrCl would be considered normal. Moreover, it is remarkable that in the case of any cutoff point for a lower PCr detecting a compromised CrCl, more patients with normal renal function would be considered as having renal dysfunction.

Cooper et al. [9] in a recent study by the Society of Thoracic Surgeons, and Holzmann et al. [18], with data from a Swedish hospital, showed an increased odds ratio for mortality among patients with mild renal insufficiency (30-59 mL/min) of 1.55 and 1.3, respectively, and in patients with severe renal insufficiency (<30 mL/min) of 2.87 and 2.8 in each study, respectively. Among our patients, group B, with an average of CrCl 51.4 mL/min, and group C with 34.8 mL/min, showed a RR of 2.1 (95% CI: 1.5 to 2.8 ) and 4.18 (95% CI: 2.9 to 6.1) compared to group A (mean CrCl = 80.2 mL/min). Pimenta et al. [5] in a retrospective analysis of in-hospital outcome of renal function in 274 patients, also demonstrated a higher rate of mortality among those with lower values †of CrCl.

For our patients, the area under the ROC curve for CrCl and PCr for predicting mortality was 0.664 and 0.595, respectively, comparable to what was found by Holzmann et al. [18] (0.71 and 0.62). These data show that CrCl is better than PCr in predicting hospital mortality.

Our study may have had some limitations, as it has been performed with patients from a single hospital, which restricts the applicability of its results to other similar centers. Group A also tended to lower average age, lower levels of criteria for emergency care and a lower prevalence of cerebrovascular disease, which could represent biases. Moreover, our calculations of RR were not adjusted for potential confounding factors. Also we did not discriminate against those who needed dialysis in the postoperative period, in addition, some factors may have affected the measurement of PCr as the use of diuretics, nephrotoxic drugs such as antibiotics, intravenous radiocontrast, or heart failure.

On the other hand, our study assumes importance due to the significantly increased size of the sample and the fact that our results were comparable to other similar studies. In addition, the Institute of Cardiology of Rio Grande do Sul is one of the most important reference in cardiovascular surgery in southern Brazil.

In conclusion, this cohort of patients undergoing CABG alone, over one quarter of those with PCr less than 1.5 mg/dL had significant loss of kidney function - CrCl below 60 mL/min. This substantial group of patients, whose renal dysfunction would have not be detected if we had take into account only the parameter of the "normal" PCr, presented twice as much risk of mortality as the TLHS and POHS in relation to other patients with PCr < 1,5mg/dL.

Therefore, we strongly recommend that patients with indication for CABG have their CrCl estimated even when the result of the PCr is within normal values, in order to identify hidden renal disease from any cause and alert to this powerful risk factor.

REFERENCES

1. Felicio ML, Andrade RR, Castiglia YMM, Silva MAM, Vianna PTG, Martins AS Cistatina C e taxa de filtração glomerular em cirurgia cardíaca com circulação extracorpórea. Rev Bras Cir Cardiovasc. 2009;24(3):305-11. [MedLine] View article

2. Cadore MP, Guaragna JCVC, Anacker JFA, Albuquerque LC, Bodanese LC, Piccoli JCE Proposição de um escore de risco cirúrgico em pacientes submetidos à cirurgia de revascularização miocárdica. Rev Bras Cir Cardiovasc. 2010;25(4):447-56. [MedLine] View article

3. De Bacco MW, Sartori AP, Sant'Anna JRM, Santos MF, Prates PR, Kalil RAK Fatores de risco para mortalidade hospitalar no implante de prótese valvar mecânica. Rev Bras Cir Cardiovasc. 2009;24(3):334-40. [MedLine] View article

4. Rodrigues FB, Bruetto RG, Torres US, Otaviano AP, Zanetta DM, Burdmann EA Effect of kidney disease on acute coronary syndrome. Clin J Am Soc Nephrol. 2010;5(8):1530-6. [MedLine]

5. Pimenta E, Ramos RF, Gun C, Santos ES, Timerman A, Piegas LS Evolução da função renal na fase aguda do infarto do miocárdio como fator prognóstico de eventos na fase intrahospitalar e em um ano de seguimento. Arq Bras Cardiol. 2006;86(3):170-4. [MedLine]

6. Santopinto JJ, Fox KA, Goldberg RJ, Budaj A, Piñero G, Avezum A Creatinine clearance and adverse hospital outcomes in patients with acute coronary syndromes : findings from the global registry of acute coronary events (GRACE). Heart. 2003;89(9):1003-8. [MedLine]

7. Asimakopoulos G, Karagounis AP, Valencia O, Alexander N, Howlader M, Sarsam MA Renal function after cardiac surgery off- versus on-pump coronary artery bypass : analysis using the Cockroft-Gault formula for estimating creatinine clearance. Ann Thorac Surg. 2005;79(6):2024-31. [MedLine]

8. Witczak B, Hartmann A, Svennevig JL Multiple risk assessment of cardiovascular surgery in chronic renal failure patients. Ann Thorac Surg. 2005;79(4):1297-302. [MedLine]

9. Cooper WA, O'Brien SM, Thourani VH, Guyton RA, Bridges CR, Szczech LA Impact of renal dysfunction on outcomes of coronary artery bypass surgery : results from the Society of Thoracic Surgeons National Adult Cardiac Database. Circulation. 2006;113(8):1063-70. [MedLine]

10. Durmaz I, Büket S, Atay Y, Yaðdi T, Ozbaran M, Boða M Cardiac surgery with cardiopulmonary bypass in patients with chronic renal failure. J Thorac Cardiovasc Surg. 1999;118(2):306-15. [MedLine]

11. Anderson RJ, O'Brien M, MaWhinney S, VillaNueva CB, Moritz TE, Sethi GK Renal failure predisposes patients to adverse outcome after coronary artery bypass surgery : VA Cooperative Study #5. Kidney Int. 1999;55(3):1057-62. [MedLine]

12. Hillege HL, Nitsch D, Pfeffer MA, Swedberg K, McMurray JJ, Yusuf S Candesartan in Heart Failure : Assessment of Reduction in Mortality and Morbidity  Investigators. Renal function as a predictor of outcome in a broad spectrum of patients with heart failure. Circulation. 2006;113(5):671-8. [MedLine]

13. Wijeysundera DN, Karkouti K, Beattie WS, Rao V, Ivanov J Improving the identification of patients at risk of postoperative renal failure after cardiac surgery. Anesthesiology. 2006;104(1):65-72. [MedLine]

14. Duncan L, Heathcote J, Djurdjev O, Levin A Screening for renal disease using serum creatinine : who are we missing?. Nephrol Dial Transplant. 2001;16(5):1042-6. [MedLine]

15. Burmeister JE, Agnolin R, da Costa MG, Miltersteiner DR, Campos BM Creatinina plasmática normal significa função renal normal. Rev AMRIGS. 2007;51:114-20.

16. Noyez L, Plesiewicz I, Verheugt FW Estimated creatinine clearance instead of plasma creatinine level as prognostic test for postoperative renal function in patients undergoing coronary artery bypass surgery. Eur J Cardiothorac Surg. 2006;29(4):461-5. [MedLine]

17. Kuan Y, Hossain M, Surman J, El Nahas AM, Haylor J GFR prediction using the MDRD and Cockcroft and Gault equations in patients with end-stage renal disease. Nephrol Dial Transplant. 2005;20(11):2394-401. [MedLine]

18. Holzmann MJ, Ahnve S, Hammar N, Jörgensen L, Klerdal K, Pehrsson K Creatinine clearance and risk of early mortality in patients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2005;130(3):746-52. [MedLine]

19. Kertai MD, Boersma E, Bax JJ, van den Meiracker AH, van Urk H, Roelandt JR Comparison between serum creatinine and creatinine clearance for the prediction of postoperative mortality in patients undergoing major vascular surgery. Clin Nephrol. 2003;59(1):17-23. [MedLine]

20. Walter J, Mortasawi A, Arnrich B, Albert A, Frerichs I, Rosendahl U Creatinine clearance versus serum creatinine as a risk factor in cardiac surgery. BMC Surg. 2003;3:4. [MedLine]

21. Cockcroft DW, Gault MH Predicition of creatinine clearance from serum creatinine. Nephron. 1976;16(1):31-41. [MedLine]

22. Saracino A, Morrone LF, Suriano V, Niccoli-Asabella A, Ramunni A, Fanelli M A simple method for correcting overestimated glomerular filtration rate in obese subjects evaluated by the Cockcroft and Gault formula : a comparison with 51Cr EDTA clearance. Clin Nephrol. 2004;62(2):97-103. [MedLine]

23. DuBois D, DuBois EF A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med. 1916;17:863-71.

24. Levey AS, Coresh J, Balk E, Kausz AT, Levin A, Steffes MW National Kidney Foundation practice guidelines for chronic kidney disease : evaluation, classification, and stratification. Ann Intern Med. 2003;139(2):137-47. [MedLine]

25. Roques F, Nashef SA, Michel P, Gauducheau E, de Vicentiis C, Baudet E Risk factors and outcome in European cardiac surgery : analysis of the EuroSCORE multinational database of 19030 patients. Eur J Cardiothorac Surg. 1999;15(6):816-22.

26. Vogt A, Grube E, Glunz HG, Hauptmann KE, Sechtem U, Mäurer W Determinants of mortality after cardiac surgery : results of the registry of the Arbeitsgemeinschaft Leitender Kardiologischer Krankenhausärzte (ALKK) on 10 525 patients. Eur Heart J. 2000;21(1):28-32. [MedLine]

Article receive on Tuesday, May 10, 2011