Renal Dysfunction As a Predictor of Long-Term Mortality in Middle-Aged Women Following an Acute Coronary Syndrome

Abstract

Objectives:

To evaluate the association between renal dysfunction and long-term prognosis among middle-aged women surviving an acute coronary syndrome (ACS).

Methods:

The Stockholm Female Coronary Risk Study included 291 women <66 years of age, hospitalized for an ACS during the years 1991–1994 in the greater Stockholm area. All patients underwent clinical screening for cardiovascular prognostic factors, including blood samples 3 to 6 months after the event. Creatinine clearance (CCr) was calculated using the Cockcroft-Gaults formula and related to adverse outcome. Hazard ratios (HR) for all-cause and cardiovascular mortality, respectively, in each tertile of CCr, were estimated using Cox proportional hazards regression.

Results:

There were 40 deaths, including 23 from cardiovascular causes, during a median follow-up of 9 years. The unadjusted HRs for all-cause mortality for the 1st (CCr < 78 mL/min) and 2nd (CCr 78–96 mL/min) tertiles of CCr were 3.17 (95% confidence interval [CI] 1.34-7.45) and 1.76 (95% CI 0.69-4.48), respectively, compared with the 3rd (CCr > 96 mL/min) tertile. The corresponding HRs for cardiovascular mortality were 3.66 (95% CI 1.20-11.1) and 1.28 (95% CI 0.34-4.76). After multivariable adjustment for potential confounders, the association between all-cause mortality and CCr in the 1st compared with the 3rd tertile remained statistically significant (HR 4.37, 95% CI 1.39-13.7).

Conclusions:

Renal dysfunction is related to long-term mortality in middle-aged women hospitalized for ACS.

Introduction

Patients with chronic kidney disease (CKD) have an increased risk of cardiovascular events independent of other risk factors.¹ In patients undergoing coronary artery bypass grafting (CABG), CKD is associated with an increased risk of postoperative complications, early death, and an increased long-term incidence of myocardial infarction (MI) and all-cause mortality.^2,3 This is also true for patients undergoing percutaneous coronary interventions or surviving acute coronary syndromes (ACS).^4
–6

An important aspect not covered by earlier long-term follow-up studies on patients with ACS is what impact CKD has on long-term prognosis, specifically in women. As the majority of patients in studies on coronary heart disease (CHD) are male, the results in previous studies may not be applicable to women. Data are lacking especially for middle-aged or younger women. Also, as women compared with men, have lower creatinine clearance (CCr) at the same age, weight, and serum creatinine (Cr) levels, the impact of a decreased glomerular filtration rate on long-term prognosis may be different among women compared with men.

The objective of this study was to investigate to what extent CKD assessed by calculated CCr is associated with long-term all-cause and cardiovascular mortality after an ACS among middle-aged women.

Materials and Methods

Study population

The Stockholm Female Coronary Risk Study included women ≤ 65 years of age residing in the greater Stockholm area, who were hospitalized for an ACS in one of the ten coronary care units in Stockholm between February 1991 and February 1994. This study has been described in detail previously.⁷

All female patients with a discharge diagnosis of MI or unstable angina were invited to participate. A total of 291 women were included when a clinical evaluation, including exercise testing and coronary angiography, was performed 3–6 months after the index event. Blood samples were drawn after fasting overnight, and serum triglycerides, total cholesterol, high-density lipoprotein cholesterol (HDL-C), and Cr levels were determined. Low-density lipoprotein cholesterol (LDL-C) was calculated using Friedewald's formula.⁸ Information on current medication, menopausal status, diabetes mellitus, and hypertension was obtained. Blood pressure was measured after 5 minutes at rest in the supine position. Height, weight, and the waist-hip ratio were measured, and body mass index (BMI) was calculated.

Elective coronary angiography was performed according to Judkins' technique.⁹ Ventriculography was performed at the end of each coronary angiography, visually assessed by two independent observers, and categorized as either normal or having any dysfunction.⁶

Definitions

Diagnosis of MI was based on the World Health Organization (WHO) criteria, including typical chest pain, enzyme patterns, and diagnostic electrocardiogram changes classified according to the Minnesota code. Patients were categorized as having unstable angina if they were admitted to the hospital because of angina at rest, new onset or accelerated angina within 4 weeks after undergoing CABG, or angina within 2 weeks of an MI.

Patients were defined as having diabetes if they were taking insulin or oral hypoglycemic agents or had a previous diagnosis of diabetes mellitus. Hypertension was defined as systolic blood pressure > 140 mm or diastolic blood pressure > 90 mm Hg or both or a history of hypertension before the index event. Smoking status was categorized as smoker, never smoker, or former smoker, defined as smoking cessation > 1 year before examination. Patients were defined as postmenopausal if menstrual bleeding was absent for at least 6 months before the study examination, a history of bilateral oophorectomy, or age > 50 years if currently taking hormone replacement therapy (HRT) started before menopause.

Renal Function

During the entire study period, serum Cr was analyzed by a nonkinetic alkaline picrate method (Jaffé) at the same laboratory.¹⁰ CCr was calculated using the equation of Cockcroft and Gault¹¹: \documentclass{aastex} \usepackage{amsbsy} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{bm} \usepackage{mathrsfs} \usepackage{pifont} \usepackage{stmaryrd} \usepackage{textcomp} \usepackage{portland, xspace} \usepackage{amsmath, amsxtra} \pagestyle{empty} \DeclareMathSizes {10} {9} {7} {6} \begin{document} \begin{align*}& (140 - age) \times (weight \ in \ kg) / (72 \times (serum \ creatinine \ \\ & \quad in \ \mu mol / L / 88.4) ) \ for \ men. \ In \ women \ the \ value \ is \ \\ & \quad multiplied \ by \ 0.85 \end{align*} \end{document}

Renal function was classified according to tertiles of CCr: > 96 mL/min, 78–96 mL/min, and < 78 mL/min.

Outcome data

Every person living in Sweden has a unique identification number that was used for record linkage to registers of death and hospital discharges, respectively. In the study population, mortality was ascertained during follow-up by means of record linkage to the Swedish National Cause of Death register, where all the deaths in Sweden are registered, with essentially complete coverage. Causes of death during follow-up were identified by record linkage to the Swedish National Patient Register, where all hospital discharges are registered, as well as from the National Cause of Death register. When the primary cause of death was coded 410–424 in the International Classification of Diseases version nine, death was defined as cardiovascular. Information on emigration was obtained from Statistics Sweden.

The study complies with the Declaration of Helsinki, and it was approved by the ethics committee at the Karolinska Institutet. All patients gave their oral approval and a written informed consent before participation in the study.

Statistical analyses

Patient characteristics for subjects with different levels of CCr are presented using means with 1 standard deviation (SD) or counts with their proportions for continuous and nominal variables, respectively. Comparisons of continuous variables among different groups were made using one-way analysis of variance (ANOVA). Chi-square test was used to compare categorical data. The hazard ratio (HR) of death was estimated crude and in multivariable analysis using Cox's proportional hazards regression. Adjustment for age, left ventricular dysfunction, diabetes mellitus, index event (MI vs. unstable angina pectoris), smoking, LDL-C levels, hypertension, previous MI, triglycerides, and use of β-blockers was made in multivariable analysis. Estimates of HR were accompanied by asymptotic 95% confidence intervals (CI). In the multivariable analysis, age, LDL-C, and triglycerides were included in the model as a continuous variable and other variables in general as dichotomous representing presence or absence of the characteristic. All analyses were performed using SAS 9.1 (SAS Institute Inc, Cary, NC).

Results

Patient characteristics

There were 291 women, with a mean age of 57 years (range 30–65 years), included in the study population. Sixty-one (21%) patients had a history of prior MI. The index event was MI in 110 (37%) and unstable angina in 181 (63%) patients. Thrombolytic therapy was given to 52 patients (46% of all MIs). During the period between the index event and clinical screening, percutaneous coronary intervention (PCI) and CABG were performed in 28 and 15 patients, respectively. The mean CCr was 87 mL/min, and 17 (5.8 %) subjects had CKD, defined as CCr < 60 mL/min. No patient had severe renal insufficiency, defined as CCr < 30 mL/min.

With decreasing CCr, the patients were older, had lower BMI and waist-hip ratio, higher LDL-C and HDL-C levels, and were less likely to have diabetes mellitus but more likely to have unstable angina, a history of MI, and reduced left ventricular function (Table 1). Smoking and hypertension did not differ among categories of renal function.

Table 1.

Characteristics of Study Population in Relation to Tertiles of Creatinine Clearance Calculated by Cockcroft-Gaults Formula

		Calculated CCr (mL/min)
Characteristic ^a	All	> 96	78–96	< 78	p value
Number of patients	291	97	97	97
Age (years)	57 (7)	52 (7)	57 (6)	59 (6)	< 0.001
Calculated CCr (mL/min)	87 (22)	111 (16)	86 (5)	66 (10)	< 0.001
Diabetes mellitus (%)^b	11	13	11	9	0.59
Hypertension (%)	50	50	48	53	0.81
Systolic blood pressure (mm Hg)	123 (21)	122 (18)	119 (17)	122 (17)	0.35
Left ventricular dysfunction (%)	14	5	21	15	0.017
Prior myocardial infarction (%)	21	21	22	20	0.94
Body mass index (kg/m²)	26.9 (5.0)	29.7 (4.5)	26.6 (3.8)	24.5 (3.0)	< 0.001
Waist-hip ratio	0.84 (0.08)	0.86 (0.08)	0.84 (0.08)	0.81 (0.08)	< 0.001
Diagnosis at index event
Myocardial infarction (%)	37	39	33	41
Unstable angina pectoris (%)	63	61	67	59	0.47
Laboratory values^c
Total cholesterol (mmol/L)	6.49 (1.27)	6.28 (1.28)	6.59 (1.21)	6.59 (1.20)	0.12
LDL-C (mmol/L)	4.07 (1.29)	3.82 (1.31)	4.18 (1.35)	4.19 (1.18)	0.07
HDL-C (mmol/L)	1.48 (0.42)	1.37 (0.35)	1.50 (0.43)	1.57 (0.45)	0.02
Triglycerides (mmol/L)	1.77 (1.83)	1.95 (2.62)	1.72 (1.52)	1.64 (0.92)	0.47
Menopausal status
Premenopausal (%)	25	40	25	11	< 0.001
Cigarette smoking
Smokers	20	23	20	18
Previous smokers	48	47	46	51
Nonsmokers	32	30	34	32	0.92

Age, CCr, laboratory values, blood pressure, body mass index, and waist-hip ratio are given as means with standard deviations.

Includes individuals with a prior dignosis of diabetes or on antidiabetic medication.

To convert total cholesterol, LDL-C, and HDL-C to mg/dL, divide by 0.02586; and to convert triglycerides to mg/dL, divide by 0.01129.

CCr, creatinine clearance; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.

Mortality

During 9 years of follow-up, there were 40 deaths, including 23 from cardiovascular causes. Cancer was the predominant noncardiovascular cause of death. We found substantially lower survival rates during follow-up in patients with the lowest levels of CCr compared with those at the highest level of CCr. In patients with CCr < 78 mL/min and 78–96 mL/min, 22% and 12%, respectively, died during follow-up compared with 7% among those with CCr > 96 mL/min. The unadjusted point estimates for all-cause and cardiovascular mortality, respectively, among those with decreased CCr suggested a graded relationship between adverse outcome and renal dysfunction (Table 2). After adjustment for confounders, only the difference in all-cause mortality between the highest and lowest tertiles of CCr remained statistically significant (Table 2). For comparison, Table 3 presents age-adjusted HRs for cardiovascular mortality in relation to various patient characteristics.

Table 2.

Hazard Ratios for Mortality in Relation to Tertiles of Calculated Creatinine Clearance Among 291 Women Surviving Acute Coronary Syndrome

	Calculated CCr (mL/min)
	> 96	78–96	< 78
	HR	HR (95% CI)	HR (95% CI)	p for trend
All-cause mortality
Number of deaths	7 deaths (7%)	12 deaths (12%)	21 deaths (22%)
Crude	1.0	1.76 (0.69-4.48)	3.17 (1.34-7.45)	0.006
Multivariable adjustment^a	1.0	0.97 (0.27-3.42)	4.37 (1.39-13.7)	0.002
Cardiovascular mortality
Number of deaths	4 (4%)	5 (5%)	14 (14%)
Crude	1.0	1.28 (0.34-4.76)	3.66 (1.20-11.1)	0.011
Multivariable adjustment^a	1.0	0.47 (0.09-2.51)	4.04 (0.97-16.9)	0.007

Multivariable adjustment was made for age, diabetes mellitus, index event (myocardial infarction vs. unstable angina pectoris), left ventricular dysfunction, low-density lipoprotein cholesterol, use of β-blockers, smoking, triglycerides, previous myocardial infarction, and hypertension.

CI, confidence interval; HR, hazard ratio.

Table 3.

Hazard Ratios For Cardiovascular Mortality In Relation to Different Patient Characteristics

Patient characteristic	Hazard ratio (95% CI) ^a
Never vs. current smoking	1.06 (0.20-5.65)
Diabetes mellitus	6.26 (2.33-16.8)
Myocardial infarction as index event	1.68 (0.62-4.54)
History of hypertension	1.40 (0.49-4.01)
Left ventricular dysfunction	6.54 (2.44-17.5)
Serum total cholesterol (mmol/L)	1.49 (0.85-2.64)
Serum triglycerides (mmol/L)	1.58 (1.00-2.49)
Serum LDL-C (mmol/L)	1.23 (0.67-2.27)
Serum HDL-C (mmol/L)	1.09 (0.55-2.16)

Hazard ratios were calculated after adjustment for age, index event (myocardial infarction vs. unstable angina pectoris), diabetes, left ventricular dysfunction, total cholesterol, and triglycerides and reported with 95% confidence intervals. For continuous variables, hazard ratios were calculated for each 1 standard deviation increase.

We also related the combined end point cardiovascular mortality and new nonfatal MI to renal dysfunction. In patients with CCr < 78 mL/min and 78–96 mL/min, 25% and 12%, respectively, reached the combined end point during follow-up compared with 11% among those with CCr > 96 mL/min. The unadjusted HRs were 1.10 (95% CI 0.49-2.50) and 2.31 (95% CI 1.13-4.72) (p for trend = 0.014) among patients with CCr 78–96 mL/min and < 78 mL/min, respectively. After adjustment for the same confounders as mentioned in Table 2, HRs were 0.60 (95% CI 0.21-1.73) and 1.96 (95% CI 0.76-5.07) (p for trend = 0.064) among patients with CCr 78–96 mL/min and < 78 mL/min, respectively.

Discussion

We studied the long-term prognostic importance of renal dysfunction in women surviving an ACS. We found that a relatively small decrease in calculated CCr was associated with an increased risk of all-cause mortality even after adjustment for potential confounders had been made.

Several studies have shown that CKD is an important risk factor for mortality and cardiovascular disease (CVD), not only among patients with established CHD^2

–6 but also in the general population.^12,13 Gender-specific analyses are rare, and there is little information on what impact, if any, CKD has on risk of premature death or CVD among women. In most studies on patients with CHD, the vast majority of the study subjects are male; this means that generalizing findings to both genders may lead to misinterpretation of data. Also, biologically, women have lower glomerular filtration rates (GFR), yet the definitions for CKD are the same in both men and women. Therefore, an increased risk of mortality associated with renal dysfunction among women may have a different association than for men.

We used the Cockcroft-Gaults formula to assess renal function because our population was relatively young and had no previous diagnosis of CKD. The most commonly used formula to assess renal function besides the Cockcroft-Gaults formula is the simplified Modification of Diets in Renal Disease (MDRD) equation.¹⁴ To our knowledge, none of these formulas have been validated in populations consisting mainly of women. Our study population did not undergo measurement of clearance; thus, we were not able to assess how well either of these two formulas performed in our patients compared with the gold standard. Studies are needed to investigate which formula is most appropriate to use for estimating GFR among middle-aged women with CHD.

Our study population had a relatively normal age-adjusted renal function, with a mean CCr of 87 mL/min. Only 17 (5.8 %) had CKD, defined as CCr < 60 mL/min, and no patients had severe renal insufficiency, defined as CCr < 30 mL/min. In previous studies, the relationship between mild renal insufficiency, CCr 60–90 mL/min, and mortality has either been absent or weak.^15,16 We found that CCr < 78 mL/min was independently related to all-cause mortality, indicating that small changes in CCr among women with established CHD are important. Cr values alone should not be used to assess renal function in order to avoid misclassification. Even if the estimation formulas have drawbacks, they will always give a better assessment of renal function than will serum Cr values alone.

When we used the combined end point of cardiovascular death and new nonfatal MI, the association with renal dysfunction was somewhat less pronounced. This suggest that renal dysfunction may play a greater role in fatal as compared with nonfatal cardiovascular events.

Study limitations

One limitation with our study was the small number of events during follow-up. There were only 23 cardiovascular deaths of a total of 40 deaths. Because of the small number of events, the HRs for cardiovascular mortality among patients with CCr < 78 mL/min were not significant. Even if the point estimates suggested a strong and graded relationship between renal dysfunction and cardiovascular death, no firm conclusions can be drawn because of the small number of events.

Forty-three patients underwent coronary angiography and intervention either with PCI or CABG between the index event and clinical screening. It is likely that some of these patients developed acute kidney injury that affected their Cr values at the time of the study investigation. We could not control for this, and it may have affected our results by classifying patients with transient Cr rises as having renal dysfunction. At clinical screening, patient blood samples were taken before they had coronary angiography, thus avoiding the possibility of rise in Cr values as a result of the contrast medium.

We classified renal function according to tertiles of CCr and not according to Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines.¹⁷ This was done because our study sample had relatively normal renal function, and only 17 (5.8%) had CKD, defined as CCr < 60 mL/min. Our reference group had CCr > 96, which is close to normal values suggested by guidelines. The tertile with lowest values, however, had CCr levels within the mild renal insufficiency range of the guidelines. This may lead to problems in using our findings in clinical practice. Nevertheless, our findings strengthen and extend previous findings of renal dysfunction being an important risk factor for death and CVD, not only in cohorts consisting mainly of men but now also in a cohort with only women.

Conclusions

We found that a small decrease in CCr among middle-aged women surviving an ACS is related to an increased risk of long-term mortality. Women with established CHD should have their renal function carefully assessed in order to predict their future risk of adverse events.

Footnotes

Disclosure Statement

The authors have no conflicts of interest to report.

References

Sarnak

, Levey

, Schoolwerth

et al. Kidney disease as a risk factor for development of cardiovascular disease: A statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation, 2003; 108:2154–2169.

Holzmann

, Ahnve

, Hammar

et al. Creatinine clearance and risk of early mortality in patients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg, 2005; 130:746–752.

Holzmann

, Hammar

, Ahnve

, Nordqvist

, Pehrsson

, Ivert

. Renal insufficiency and long-term mortality and incidence of myocardial infarction in patients undergoing coronary artery bypass grafting. Eur Heart J, 2007; 28:865–871.

Anavekar

, McMurray

, Velazquez

et al. Relation between renal dysfunction and cardiovascular outcomes after myocardial infarction. N Engl J Med, 2004; 351:1285–1295.

Gibson

, Pinto

, Murphy

et al. Association of creatinine and creatinine clearance on presentation in acute myocardial infarction with subsequent mortality. J Am Coll Cardiol, 2003; 42:1535–1543.

Shaw

, Andrianopoulos

, Duffy

et al. Melbourne Interventional Group. Renal impairment is an independent predictor of adverse events post coronary intervention in patients with and without drug-eluting stents. Cardiovasc Revasc Med, 2008; 9:218–223.

Al-Khalili

, Svane

, Wamala

, Orth-Gomer

, Ryden

, Schenck-Gustafsson

. Clinical importance of risk factors and exercise testing for prediction of significant coronary artery stenosis in women recovering from unstable coronary artery disease: The Stockholm Female Coronary Risk Study. Am Heart J, 2000; 139:971–978.

Friedewald

, Levy

, Fredrickson

. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 1972; 18:499–502.

Judkins

. Selective coronary arteriography. I. A percutaneous transfemoral technic. Radiology, 1967; 89:815–824.

10.

Bartels

, Bohmer

, Heierli

. [Serum creatinine determination without protein precipitation.] Clin Chim Acta, 1972; 37:193–197.

11.

Cockcroft

, Gault

. Prediction of creatinine clearance from serum creatinine. Nephron, 1976; 16:31–41.

12.

, Chertow

, Fan

, McCulloch

, Hsu

. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med, 2004; 351:1296–1305.

13.

Meisinger

, Doring

, Lowel

. Chronic kidney disease and risk of incident myocardial infarction and all-cause and cardiovascular disease mortality in middle-aged men and women from the general population. Eur Heart J, 2006; 27:1245–1250.

14.

Levey

, Greene

, Kusek

, Beck

. A simplified equation to predict glomerular filtration rate from serum creatinine. J Am Soc Nephrol, 2000; 11:155A.

15.

Van Biesen

, De Bacquer

, Verbeke

, Delanghe

, Lameire

, Vanholder

. The glomerular filtration rate in an apparently healthy population and its relation with cardiovascular mortality during 10 years. Eur Heart J, 2007; 28:478–483.

16.

Levey

, Eckardt

, Tsukamoto

et al.

Definition and classification of chronic kidney disease: A position statement from Kidney Disease: Improving Global Outcomes (KDIGO)

Kidney Int, 2005; 67:2089–2100.

17.

K/DOQI clinical practice guidelines for chronic kidney disease. Evaluation, classification, and stratification. Am J Kidney Dis, 2002; 39:S1–266.