Impact of occult renal impairment on early and late outcomes following coronary artery bypass grafting

Abstract

OBJECTIVES

High serum creatinine is considered an independent risk factor for poor outcomes following coronary artery bypass grafting (CABG). However, the impact of occult renal impairment (ORI), defined as an impaired glomerular filtration rate (GFR) with a normal serum creatinine (SCr) level, remains unclear. Thus, we sought to investigate the impact of ORI on outcomes after CABG.

METHODS

Among patients undergoing their first percutaneous coronary intervention (PCI) or CABG enrolled in the CREDO-Kyoto Registry (a registry of first-time PCI and CABG patients in Japan), 1842 patients with normal SCr levels undergoing CABG were enrolled in the study. Patients were divided into two groups based on preoperative estimated GFR calculated by the Cockcroft–Gault equation: 1339 patients with estimated GFR of ≥60 ml/min/1.73 m² (normal group) and 503 with estimated GFR of <60 ml/min/1.73 m² (ORI group).

RESULTS

Preoperative estimated GFR differed between the groups (51.3 ± 6.6 vs 85.8 ± 23.0 ml/min/1.73 m², P < 0.01). ORI was associated with high in-hospital mortality (3.2 vs 1.0%, P < 0.01) and need for dialysis (2.0 vs 0.2%, P < 0.01). In terms of long-term outcomes, ORI was associated with high mortality compared with the normal (hazard ratio [95% confidence interval]: 1.72 [1.16–2.54], P < 0.01) and high incidence of composite cardiovascular events (death, stroke or myocardial infarction: 1.53 [1.16–2.02], P < 0.01).

CONCLUSIONS

ORI was an independent risk factor for early and late death as well as cardiovascular events in patients undergoing CABG with normal SCr levels. A more accurate evaluation of renal function through a combination of SCr and estimated GFR is needed in patients with normal SCr levels.

INTRODUCTION

Preoperative renal dysfunction is considered an independent risk factor for mortality and morbidity after coronary artery bypass grafting (CABG) [1–4]. SCr is the most popular test for renal failure. However, SCr does not usually rise until the glomerular filtration rate (GFR) has fallen by 75% or more, and it can remain within the normal range even when renal function is significantly impaired [5–7]. SCr level depends on muscle mass, age, gender and metabolism, and it thus has a limited ability to identify patients with preoperative renal failure [7, 8].

Some populations with normal SCr but impaired renal function have been reported. Occult renal impairment (ORI) is defined as an impaired GFR with a normal SCr level [9]. Because GFR has been reported to estimate poor outcomes more accurately than SCr after cardiac surgery [10, 11], patients with ORI may be more strongly associated with poor outcomes compared with those with normal SCr levels and GFRs. However, the impact of ORI on outcomes after CABG, particularly on long-term outcomes, remains poorly understood. Thus, in the present study, we sought to investigate the impact of ORI on early and late outcomes after CABG using a large Japanese multicentre registry of myocardial revascularization.

PATIENTS AND METHODS

Study population

The Coronary REvascularization Demonstrating Outcome Study in Kyoto (CREDO-Kyoto) is a multicentre registry in Japan that includes 9877 consecutive patients undergoing their first percutaneous coronary intervention (PCI) or CABG and excludes those with acute myocardial infarction within a week before the index procedure [12]. This study was approved by the institutional review boards or ethics committees of all participating institutions (see Supplemental Table). Because the subjects were retrospectively enrolled, written informed consent was not obtained, in accordance with the guidelines for epidemiological studies issued by the Ministry of Health, Labor and Welfare of Japan. Seventy-three patients were excluded because of their refusal to participate in the study when contacted for follow-up [12].

Between January 2000 and December 2002, 9877 patients were identified as having undergone either PCI (6878 patients) or CABG (2999 patients) without a prior history of coronary revascularization. Among the 2999 patients undergoing CABG, 484 undergoing concomitant valvular, left ventricular or major vascular operations were excluded from the current analysis. Among the 2515 patients undergoing isolated CABG, those with normal SCr levels (0.6–1.1 mg/dl for males; 0.4–0.8 mg/dl for females) were included. Therefore, the study group comprised 1842 patients with normal SCr who were undergoing their first isolated CABG.

Data collection and definitions

Demographic, angiographic and procedural data were collected from hospital charts or databases at the various centres by independent clinical research coordinators according to pre-specified definitions. Baseline clinical characteristics, such as myocardial infarction, heart failure, diabetes, hypertension, current smoker status, atrial fibrillation, chronic obstructive lung disease and malignancy, were regarded as present when these diagnoses were recorded in the hospital charts. Left ventricular ejection fraction (LVEF) was measured either by contrast left ventriculography or echocardiography. Anaemia was defined as a blood haemoglobin level of <12 g/dl, as previously described [12].

Estimated GFR (eGFR, ml/min/1.73 m²) was estimated by the Cockcroft–Gault formula [13, 14]. Renal impairment was regarded as present when GFR estimated by the Cockcroft–Gault formula was less than 60 ml/min/1.73 m², as per the National Kidney Foundation's definition [15]. Thus, ORI was defined as an impaired eGFR (<60 ml/min/1.73 m²) with a normal SCr level.

End-points

An independent clinical events committee adjudicated events. Death was regarded as cardiovascular in origin unless obvious non-cardiovascular causes could be identified. Any death during the index hospitalization was regarded as cardiovascular death. Myocardial infarction was adjudicated according to the definition in the Arterial Revascularization Therapy Study [16]. Within 1 week of the index procedure, only Q-wave myocardial infarction was adjudicated as myocardial infarction. Stroke was defined as any new permanent global or focal neurological deficit that could not be attributed to other neurological or medical processes. In the majority of patients, strokes were diagnosed by neurologists and confirmed by computed tomography or magnetic resonance imaging head scans. Stroke at follow-up was defined as symptomatic stroke.

The primary end-point measure was death from any cause. Secondary end-points were stroke, myocardial infarction, the need for any revascularization procedures and major adverse cardiovascular events (MACEs; which include death, stroke or myocardial infarction) during the follow-up period. As in-hospital end-points, the need for postoperative dialysis and the ratio of peak postoperative to preoperative Cr ≥2 were also evaluated.

Statistical analyses

All continuous variables were expressed as the mean ± standard deviation. Differences in baseline characteristics between the two groups were examined by an unpaired t-test and a Fisher's exact test. Categorical variables were presented as number and percentage and were compared with a χ² test. Cumulative incidence was estimated by the Kaplan–Meier method, and differences were assessed using a log-rank test. Outcomes after CABG in the presence or absence of ORI were compared by logistic regression or Cox proportional hazards models. Cox proportional hazards models were adjusted for the following clinically relevant confounders: age, gender, body mass index, emergency procedure, prior myocardial infarction, congestive heart failure, stroke, peripheral arterial disease, atrial fibrillation, chronic obstructive pulmonary disease, malignancy, hypertension, diabetes mellitus, hyperlipidaemia, anaemia, current smoker status, on- or off-pump surgery, triple-vessel disease, left main disease, total occlusion, proximal left anterior descending artery disease and use of left internal thoracic artery, right internal thoracic artery, gastroepiploic artery, radial artery or saphenous vein. They were consistent with previous reports from the current registry. Continuous variables were dichotomized using clinically meaningful reference values or median values.

All reported P-values were two-sided. All analyses were conducted by a statistician using SAS software version 9.2 (SAS Institute, Inc., Cary, NC, USA) and S-Plus version 7.0 (Insightful, Corp., Seattle, WA, USA). The authors had full access to the data and take responsibility for its integrity.

RESULTS

Baseline characteristics and operative outcomes

Of the 1842 patients undergoing CABG with normal SCr levels, 503 (27.3%) had ORI. The baseline characteristics of the patients in the two groups are presented in Table 1. The ORI group included more elderly patients, but fewer male and lower body mass index patients. In addition, the ORI group generally included patients with heart failure, history of stroke and peripheral arterial disease. Preoperative coronary characteristics were similar between the groups, except for the ratio of triple-vessel disease. The ratio of off-pump surgery and the average number of anastomotic sites per patient were similar between the groups (Table 2). Arterial grafts in such vessels as the left internal thoracic artery, the gastroepiploic artery and the radial artery were used more commonly in the Normal group.

Table 1:

Baseline characteristics

	Occult renal impairment (n = 503)		Normal (n = 1339)		P-value
Age (years)	73.6 ± 6.4		64.1 ± 9.1		<0.01
≥75	240	48%	149	11%	<0.01
Male gender	327	65%	1057	79%	<0.01
Estimated glomerular filtration rate (ml/min/1.73 m2)	51.3 ± 6.6		85.8 ± 23.0		<0.01
Body mass index	21.6 ± 2.8		24.5 ± 3.0		<0.01
Emergency procedure	38	8%	62	5%	0.01
Old myocardial infarction	210	42%	388	29%	<0.01
Heart failure	124	25%	240	18%	<0.01
Ejection fraction (%)	59.6 ± 14.8		61.6 ± 13.5		<0.01
Stroke history	119	24%	236	18%	<0.01
Peripheral arterial disease	118	23%	213	16%	<0.01
Atrial fibrillation	24	5%	62	5%	0.89
Chronic obstructive pulmonary disease	16	3%	21	2%	0.03
Hypertension	346	69%	967	72%	0.67
Diabetes mellitus	215	43%	581	43%	0.83
Hyperlipidaemia	235	47%	761	57%	<0.01
Anaemia (haemoglobin <12 mg/dl)	203	40%	255	19%	<0.01
Current smoker	88	17%	380	28%	<0.01
Malignancy	42	8%	74	6%	0.03
Coronary characteristics
Triple-vessel disease	358	71%	853	64%	<0.01
Left main disease	141	28%	401	30%	0.42
Chronic total occlusion	212	42%	603	45%	0.27
Proximal left anterior descending artery disease	300	60%	793	59%	0.87

Table 2:

CABG data

	Occult renal impairment (n = 503)	Normal (n = 1339)	P-value
Off-pump	54.3%	56.3%	0.43
No. of anastomotic sites	3.3 ± 1.2	3.3 ± 1.1	0.57
Type of bypass grafts
Left internal thoracic artery	90.7%	94.4%	<0.01
Right internal thoracic artery	29.0%	31.4%	0.33
Right gastroepiploic artery	21.5%	29.1%	<0.01
Radial artery	28.0%	39.1%	<0.01
Saphenous vein	67.6%	53.6%	<0.01

In-hospital outcomes

Observed in-hospital mortality was significantly higher in the ORI group (3.2 vs 1.0%, P < 0.01; Table 3). In-hospital incidence of stroke in the ORI group tended to be higher than the Normal group (P = 0.06). The need for dialysis after CABG was more common in the ORI group (2.0 vs 0.2%, P < 0.01). In addition, patients with a ratio of post- to preoperative SCr level of ≥2 were more common in the ORI group (9.3 vs 6.1%, P = 0.02). A similar tendency was seen in the adjusted outcomes (Table 4). Adjusted in-hospital mortality was also significantly higher in the ORI group (odds ratio [95% confidence interval]: 2.41 [1.04–5.57], P = 0.04). The need for dialysis after CABG was more common in the ORI group (5.93 [1.46–24.12], P = 0.01).

Table 3:

Observed in-hospital outcomes

	Occult renal impairment (n = 503)		Normal (n = 1339)		P-value
Death	16	3.2%	13	1.0%	<0.01
Stroke	12	2.4%	16	1.2%	0.06
Myocardial infarction	11	2.2%	16	1.2%	0.11
Need for dialysis	10	2.0%	3	0.2%	<0.01
Post/pre Cre ≥2a	47	9.3%	82	6.1%	0.02

^aRatio of peak postoperative to preoperative SCr level.

Table 4:

Adjusted in-hospital outcomes (occult renal impairment vs normal)

	Hazard ratio	95% CI		P-value
Death	2.41	1.04	5.57	0.04
Stroke	1.76	0.81	3.85	0.15
Myocardial infarction	1.76	0.81	3.84	0.15
Need for dialysis	5.93	1.46	24.12	0.01
Post/pre Cre ≥2a	1.33	0.87	2.02	0.19

^aRatio of peak postoperative to preoperative SCr level.

Long-term outcomes

Clinical follow-ups were completed for 98% of cases at 1 year and for 95% of cases at 2 years. The median follow-up period was 3.7 years. Figure 1 shows the Kaplan–Meier curves for survival and freedom from MACEs after CABG. Cumulative survival in the ORI and the Normal groups were 94.4 and 98.1% at 1 year and 83.2 and 93.1% at 5 years, respectively (P < 0.01). Similarly, freedom from MACEs was higher in the Normal group (P < 0.01).

Figure 1:

Kaplan–Meier curves for all-cause death and major adverse cardiovascular events in the presence or absence of occult renal impairment. (A) All-cause death. (B) Major adverse cardiovascular event (death, stroke or myocardial infarction [MI])

Adjusted all-cause mortality was significantly higher in the ORI group (1.72 [1.16–2.54], P < 0.01, Table 5). Adjusted incidence of stroke, myocardial infarction and need for any revascularization did not differ between the groups. However, the incidence of MACEs was significantly higher in the ORI group (1.53 [1.16–2.02], P < 0.01).

Table 5:

Hazard ratios for late outcomes after CABG in patients with vs without occult renal impairment

	Event (%)				Occult renal failure vs Normal
	Occult renal impairment (n = 503)		Normal (n = 1339)		Hazard ratio	95% CI		P-value
Death	67	13.3%	71	5.3%	1.72	1.16	2.54	<0.01
Stroke	43	8.5%	73	5.5%	1.45	0.96	2.18	0.08
Myocardial infarction	19	3.8%	37	2.8%	1.38	0.79	2.40	0.26
Any revascularization	59	11.7%	171	12.8%	0.96	0.71	1.30	0.81
MACEa	108	21.5%	155	11.6%	1.53	1.16	2.02	<0.01

^aComposite of death, stroke or myocardial infarction.

DISCUSSION

Principal findings

Preoperative renal impairment is considered an independent risk factor for mortality and morbidity after CABG [1–4]. SCr is the most popular test for renal impairment. However, some populations had impaired renal function despite normal SCr levels. In the present study, we sought to investigate the impact of ORI on early and late outcomes after CABG. To the best of our knowledge, this is the first report to investigate the late outcome of ORI patients. In addition, there has been no multicentre registry that has investigated the impact of ORI after CABG.

Observed in-hospital outcomes revealed that ORI was associated with high mortality compared with normal renal function. In addition, patients with ORI were associated with a higher need for postoperative dialysis and a higher peak postoperative SCr level. Adjusted in-hospital outcomes also revealed that patients with ORI were associated with higher mortality and the need for dialysis after CABG. In terms of late outcomes, patients with ORI were associated with poor survival compared with those with normal renal function. In addition, ORI was associated with a high incidence of major cardiovascular events. These outcomes indicate that routine estimation or measurement of GFR should be preferred over SCr as a screening method for the detection of higher-risk patients undergoing CABG. This study is important because we demonstrated that ORI is an independent risk factor of death after CABG not only for the in-hospital period but also for the long term.

The relationship between preoperative renal function and postoperative outcomes has been reported in a number of studies [1–4]. SCr level measurement is easy and is the most popular method used to evaluate preoperative renal function in screening tests. However, its use has been questioned in several studies, because it can be normal in spite of impaired renal function and it may underestimate mild renal dysfunction [5–8]. A more accurate approximation of renal function can be obtained using the Cockroft–Gault equation to calculate an eGFR from SCr. Several studies have reported the greater accuracy of eGFR compared with SCr in estimating the prognosis after cardiac surgery [10, 11, 17–19].

Renal dysfunction is an adverse prognostic factor in patients with coronary artery disease [20]. In patients undergoing CABG, moderate and end-stage renal dysfunction are recognized risk factors for increased perioperative mortality and are accounted for in the commonly used cardiac risk stratification scoring systems [21, 22]. In addition, severe preoperative renal disease is associated with a higher incidence of morbidity, need for dialysis and longer hospital stay after CABG [23]. However, limited information exists regarding the influence of normal renal function on the immediate postoperative mortality and long-term survival of patients undergoing isolated first-time CABG.

Several reports have investigated the outcomes of normal renal function to mild renal dysfunction after cardiac surgery. Wijeysundera et al. [19] reported that among 10 751 patients with normal SCr levels undergoing cardiac surgery using cardiopulmonary bypass, 13% had ORI. Compared with patients with normal renal function, those with ORI experienced a more-than threefold an increased risk of in-hospital mortality and renal replacement therapy after cardiac surgery. Wang et al. [10] reported that among 4603 patients with normal plasma Cr levels undergoing cardiac surgery, 12.3% had ORI and that ORI was associated with an increased risk of in-hospital mortality, renal failure requiring dialysis and major morbidity, such as cardiovascular, respiratory, neurological, renal dysfunction and infection. In addition, estimated Cr clearance remained a significant predictor of these outcomes, but plasma Cr level was not a predictor of any outcomes.

Specifically for CABG, Noyez et al. [11] compared the risk stratification by level of SCr with estimated C clearance in 627 patients undergoing isolated CABG. They concluded that the association between preoperative renal failure and adverse outcomes after CABG is stronger with the estimated Cr clearance than with the routinely used SCr and that routine estimation or measurement of Cr clearance should be preferred over SCr as a screening method for the detection of higher-risk patients undergoing CABG. A large single-centre registry of 9159 patients with normal SCr undergoing CABG by Miceli et al. [17] reported that ORI was associated with a doubling in the risk of operative mortality, postoperative renal dysfunction and need for dialysis. In addition, ORI increased the risk of stroke, arrhythmia and prolonged hospital stay. However, all of these reports were limited to early outcomes.

There are several important limitations to this study. First and most importantly, observational studies have problems related to biases and unmeasured confounders. Multivariate analyses may not adequately adjust for these biases. Second, we used the Cockcroft–Gault equation to estimate GFR because it has been most generally used in clinical settings. However, we should have used another method to calculate GFR. GFR can be calculated based on SCr (the Cockcroft–Gault, the Modification of Diet in Renal Disease [MDRD] study or Chronic kidney disease Epidemiology Collaboration [CKD-EPI] equations) or on Cystatin C [24, 25]. Using SCr has significant disadvantages, such as the inability to measure renal function correctly when impairment is 50% or less. Creatinine generation is proportional to muscle mass and related to an individual's age, sex, race and weight. On the other hand, Cystatin C is a cysteine protease inhibitor with a molecular mass of 13 kDa. It has been shown that cystatin C is a more sensitive marker of GFR changes than SCr because its levels are not affected by muscle mass, age, inflammation, fever or exogenous agents. Finally, we have adjusted the differences of patients' backgrounds (e.g. age, sex, myocardial infarction, heart failure) using propensity score analysis. However, the analysis may not adequately adjust for these biases. In addition, our study was nonrandomized; potential confounders may influence our outcomes.

CONCLUSIONS

ORI is an independent risk factor of early and late death as well as MACEs in patients undergoing CABG with normal SCr levels. A more accurate evaluation of renal function through a combination of SCr and eGFR may facilitate a better risk stratification that can optimize therapeutic strategies.

Supplementary Data

Supplementary material is available at ICVTS online.

Funding

This work was supported in part by a Grant for Clinical Research for Evidence Based Medicine from the Ministry of Health, Labor and Welfare in Japan to T. Kimura and an educational grant from the Research Institute for Production Development (Kyoto, Japan).

Conflict of interest: none declared.