Effect of Angiotensin-Converting Enzyme Inhibition on Survival in 3773 Chinese Type 2 Diabetic Patients
We assessed the effects of angiotensin-converting enzyme (ACE) inhibition on survival and cardiorenal outcomes in a consecutive cohort of Chinese type 2 diabetic patients with varying degree of albuminuria, ranging from normoalbuminuria to macroalbuminuria. A total of 3773 consecutive Chinese type 2 diabetic patients were followed prospectively for a mean period of 35.8 months. Clinical end points included all-cause mortality, with cardiovascular end point defined as first hospitalization because of ischemic heart disease, congestive heart failure, revascularization procedures, or cerebrovascular accident as well as renal end point defined as dialysis, doubling of baseline plasma creatinine, or plasma creatinine ≥500 μmol/L. The use of ACE inhibitor was 26.3% in normoalbuminuric (NA), 70.1% in microalbuminuric (MI), and 82.6% in macroalbuminuric (MA) groups. Albuminuria was a major predictor for all-cause mortality with 4-fold difference between NA and MA patients. The 7-year cumulative mortality rate was 7.1%, 10.8%, and 21.7% in the NA, MI, and MA groups, respectively. The use of ACE inhibition was associated with significant reduction of mortality (hazard ratio 0.41 and 95% confidence interval, 0.29, 0.58) in the entire group and was most evident in high-risk patients who had cardiorenal complications or retinopathy at baseline for all albuminuric groups (NA 0.76 [0.31,1.87]; MI 0.32 [0.16, 0.65]; and MA 0.20 [0.13, 0.33]). The prognostic value of albuminuria for death in type 2 diabetes and the beneficial effects of ACE inhibitors in Chinese type 2 diabetic patients with micro- or macroalbuminuria has been confirmed. The effects of ACE inhibitors in type 2 diabetic patients with normoalbuminuria require further evaluation.
Type 2 diabetes mellitus is the most common cause of end-stage renal disease, accounting for 40% of all new cases of end-stage renal disease in most developed countries.1,2 Albuminuria is a powerful and independent predictor for death and cardiorenal outcomes in type 2 diabetic patients.3–5 Several landmark studies have confirmed the beneficial effects of angiotensin-converting enzyme (ACE) inhibitors on cardiovascular mortality and morbidity in high-risk patients6 as well as that of angiotensin II (Ang II) antagonists on renal end points in type 2 diabetic patients with left ventricular hypertrophy, clinical proteinuria, and renal insufficiency.7–9 However, the beneficial effect of renin-angiotensin-aldosterone blockage in type 2 diabetic patients with normoalbuminuria is unknown.6,10–12 In this large-scale observational study of Chinese type 2 diabetic patients, we examined the effects of renin-angiotensin-aldosterone system (RAAS) inhibition on survival and cardiorenal outcomes in a consecutive cohort referred to our clinic since 1995 to address these 2 important therapeutic issues.
Patients and Methods
Between 1995 and 2000, a consecutive cohort of 5004 Chinese type 2 diabetic patients from the Prince of Wales Hospital underwent detailed assessment using the European DiabCare protocol.13 Of these, 3773 who had been observed at least 6 months were included in the analysis. Patients with type 1 diabetes, defined as those presenting with diabetic ketoacidosis, acute presentation with heavy ketonuria (>3+), or continuous requirement of insulin within 1 year of diagnosis,14 were excluded. Among these 3773 patients, 1419 (37.6%) were considered to be high risk, defined as those with a known history of cardiovascular complications including ischemic heart disease, heart failure, stroke and/or peripheral vascular disease, or renal impairment with plasma creatinine ≥150 μmol/L, or presence of retinopathy at baseline assessment. Low-risk subjects were defined as those without the above complications. The type of ACE inhibitors used included captopril, enalapril, lisinopril, ramipril, and perindopril, whereas losartan, candesartan, and valsartan were the Ang II antagonists used.
Fasting blood samples were taken for measurement of plasma glucose, glycohemoglobin (HbA1c), lipid profile (total cholesterol, high-density lipoprotein cholesterol [HDL-C], triglycerides, and calculated low-density lipoprotein cholesterol), and renal and liver functions. A sterile, random spot urine sample was used to measure albumin creatinine ratio (ACR) followed by a timed collection (4-hour or 24-hour) for albumin excretion rate. Using the ACR from these 2 samples, normoalbuminuria was defined as a mean ACR ≤3.5 mg/mmoL, microalbuminuria ACR between 3.5 and 25 mg/mmoL, and macroalbuminuria ≥25 mg/mmoL.15
In this analysis, mortality data were obtained from the Hong Kong Death Registry ascertained by review of case notes. Details of all medical admissions with primary and secondary diagnosis as well as medication history and last available plasma creatinine results were retrieved from the Central Computerized System at the Hospital Authority Head Office, which is the governing body of all the public hospitals in Hong Kong and captures more than 90% of these data. Cardiovascular end point was defined as hospitalizations because of ischemic heart disease, congestive heart failure, stroke, and revascularization procedures. Renal end point was defined as dialysis or doubling of baseline plasma creatinine or absolute value ≥500 μmol/L.
Plasma glucose was measured by a hexokinase method (Hitachi 911 automated analyzer, Boerhringer Mannheim). HbA1c was measured by an automated ion-exchange chromatographic method (Bio-Rad Laboratory; with reference range 5.1 to 6.4%). Interassay and intraassay coefficients of variation for HbA1c were ≤3.1% at values <6.5%. Total cholesterol, triglycerides, and HDL-C were measured by enzymatic methods on a Hitachi 911 automated analyzer (Boehringer Mannheim) using reagent kits supplied by the manufacturer of the analyzer. Low-density lipoprotein cholesterol was calculated by the Friedewald equation for TG <4.5 mmol/L.16 The precision performance of these assays was within the manufacturer’s specifications. Urinary creatinine (Jaffe kinetic method) and albumin (immunoturbidimetry method) were also measured by the Hitachi 911 analyzer using reagent kits supplied by the manufacturer. The interassay precision CV was 12.0% and 2.3% for urinary albumin concentrations of 8.0 mg/L and 68.8 mg/L, respectively. The lowest detection limit was 3.0 mg/L. Plasma creatinine (Jaffe kinetic method) was measured on a Dimension AR system (Dade Behring).
The analysis was performed using the Statistical Package for Social Sciences (version 9.0) statistical package. Plasma triglyceride, plasma creatinine, and albuminuria were logarithmically transformed because of skewed distributions. All data are expressed as mean±SD or median (interquartile range), as appropriate. The Student t test or ANOVA was used for between-group comparisons for continuous variables and the χ2 test for categorical variables. The Cox regression model was used to estimate the hazard ratio (HR) with 95% confidence interval (CI) for mortality and clinical end points, with the assumption that the effects of the different variables on survival are constant. Independent variables were subjected to univariate analysis, and then variables showing statistically significant result were entered as covariates in the multivariate analysis. Kaplan-Meier analysis was used to estimate the cumulative incidence of death and cardiorenal outcomes, and the log-rank test was used to demonstrate trend for survival. P<0.05 (2-tailed) was considered to be significant.
A total of 3773 patients with at least 6 months of observational follow-up after assessment between January 1995 and May 2001, with a mean follow-up period of 35.8±18.3 months, were enrolled for survival analysis. At baseline, 2048 (54.9%) patients had normoalbuminuria (NA), 1047 (28.1%) had microalbuminuria (MI), and 634 (17.0%) had macroalbuminuria (MA). The use of RAAS inhibitors was 26.3%, 70.1%, and 82.6% in NA, MI, and MA patients, respectively.
The majority of these patients (84.3%) were treated with an ACE inhibitor (Table 1). A small proportion (15.7%) of patients had been treated with an ACE inhibitor and then switched to an Ang II antagonist or treated with an Ang II antagonist alone. For patients in whom the RAAS inhibition was discontinued (n=144), the reasons for discontinuation as recorded in the case notes were as follow: 12.0% because of cough, 6.3% because of hyperkalaemia, and 18.4% because of increased plasma creatinine. In 38.6% of patients, the reasons for treatment discontinuation were unclear.
Figures 1 and 2⇓ and Table 2 show the cumulative mortality in patients categorized according to their baseline albuminuric status and the effects of RAAS inhibition on all-cause mortality. Tables 3 and 4⇓ show the major determinants of mortality in the patient group, which included increased age, male sex, albuminuric status, cardiovascular complications at baseline, and nonusage of RAAS inhibition.
Albuminuria was an independent predictor for all-cause mortality with a 4-fold difference between NA and MA patients (HR 4.75 [95% CI, 3.35, 6.72; P<0.001]) and 2-fold difference between NA and MI patients (HR1.86 [95% CI, 1.27, 2.72; P=0.001]). The 7-year cumulative mortality rate was 7.1%, 10.8%, and 21.7% in NA, MI, and MA groups, respectively. Using Cox regression analysis, for every 1 natural log-value increase in albumin excretion rate, the HR for death was increased by 1.39 (95% CI, 1.30, 1.49; P<0.001; Figure 1 and Tables 3 and 4⇑).
The use of RAAS inhibitors significantly reduced mortality even after controlling for other confounding factors (HR 0.41 [95% CI, 0.29, 0.58]; Tables 2, 3, and 4⇑⇑). On subgroup analysis, RAAS inhibition had a neutral effect in the NA group but showed distinct benefits in the MI and, especially, MA groups (Figure 2, Table 2), particularly for those with cardiovascular complications, renal insufficiency, or retinopathy at baseline. Table 2 and Figure 2 summarize the effects of RAAS inhibitor on composite end points of death and cardiorenal end points, with these agents conferring beneficial effects on death and renal end point in both MI and MA groups, but not in the NA group.
The effect of different patterns of RAAS inhibitor usage on survival is shown in Figure 3. Patients who were treated with a RAAS inhibitor at baseline that was subsequently discontinued had the highest annualized all-cause mortality rate of 5.0%. This is compared with 1.4% in patients who had never been put on RAAS inhibitors, 1.7% in those who persisted with a RAAS inhibitor from baseline, and 1.2% for those who commenced subsequently and persisted with treatment after baseline assessment (P<0.001). Patients in whom treatment with RAAS inhibitor was discontinued had lower baseline HDL-C (1.2±0.4 versus 1.3±0.4 mmol/L, P=0.04), higher plasma creatinine (101 [78,171] versus 91 [75,117] μmol/L, P<0.001), and higher prevalence of retinopathy (65.6% versus 56.2%, P=0.04) compared with the group having persistent use. All results remained the same when time weighing was taken into account.
The results of this large-scale cohort study indicate that inhibition of the RAAS with ACE inhibitors or Ang II antagonists was associated with a significant reduction inmortality and renal end points in high risk patients with known diabetic complications or those with clinical proteinuria or renal insufficiency. No significant outcome benefits were shown in those with normoalbuminuria within the modest follow-up period. These results are in accordance with the greater benefits of Ang II inhibition in patients with severe proteinuria or renal insufficiency (eg, plasma creatinine ≥200 μmol/L) in the Reduction of Endpoints in Non–insulin-dependent diabetes mellitus with the Angiotensin II Antagonist Losartan (RENAAL) and Irbesartan Diabetic Nephropathy Trial (IDNT) studies.7,8 We found that patients with RAAS inhibition had more than 50% risk reduction for all-cause mortality in microalbuminuric patients and 80% in the macroalbuminuric group compared with their counterparts. More importantly, this benefit was highlighted by the similar death rate between macroalbuminuric patients treated with a RAAS inhibitor and microalbuminuric patients not treated with the drug (Figure 2). Translating the results of our study into the number needed to treat, only 3 macroalbuminuric patients needed to be treated to prevent 1 case of death or end-stage renal disease when other risk factors were optimized.
Although there is a large body of evidence from clinical trials supporting the cardiorenal protective effects of RAAS inhibition in type 2 diabetic patients,6–8 their effectiveness in clinical practice, especially for low-risk patients, remains to be determined.17 In the United Kingdom Prospective Diabetes Study, treatment with an ACE inhibitor (captopril) had similar effects on clinical outcomes in hypertensive type 2 diabetic patients compared with treatment with a β-blocker (atenolol) although the albuminuric status of these patients was not clearly defined.12 Similarly, the Enalapril Efficacy in Nephropathy Study also failed to confirm the renoprotective effect of ACE inhibitors in normotensive type 2 diabetic patients in the initial period, although ACE inhibitors reduced the rate of decline in renal function once albuminuria progressed.11 In our analysis, RAAS inhibition had neutral effects in patients with normoalbuminuria after adjustment for baseline cardiovascular and metabolic profile. It is plausible that beneficial effects (if any) conferred by RAAS inhibition in normoalbuminuric type 2 diabetic subjects may only become apparent with longer follow-up duration. Clearly future studies with longer term follow-up would be required to address this issue.
A remarkable finding of our study is that patients treated with ACE inhibitors, which were subsequently discontinued, had the highest annualized mortality compared with those who had never been treated with RAAS inhibitors. This reflects the high-risk group, which warranted ACE inhibitors, and this finding is consistent with those of the MICRO-Heart Outcomes Prevention Evaluation (HOPE) substudy.6 Our results also indicate that physicians should consider using Ang II antagonist as an alternative in high-risk patients who are unable to tolerate ACE inhibitors.
Interestingly, we found that malignancy was the main cause of death among patients who were not treated with RAAS inhibition. The underlying explanation for this observation is unclear, but our results are consistent with other previous study findings.18,19 Our study was not designed to investigate this issue; hence, further studies specifically designed to address this issue would be required.
Several potential limitations of this large-scale study warrant further discussion. First, this study cohort was recruited from 1995 to 2000, a time before publication of several landmark studies such as RENAAL, IDNT, and the Losartan Intervention For Endpoint reduction in hypertension (LIFE). Hence, most of the patients were treated with ACE inhibitors rather than Ang II antagonists (n=321). Nevertheless, the results were similar after we excluded this latter group of patients. Second, this study was not randomized, which caused a degree of selection bias. Third, our study results could be affected by confounding factors or bias in the analysis because the analysis did not take into account changes in antihypertensive therapy. Lastly, reasons for discontinuation of RAAS inhibition therapy were not clearly documented in nearly 40% of patients, which might confound the inference regarding the cause of death or morbidity. Despite the above limitations, and without any monitoring of drug compliance, we were able to show very significant beneficial results of RAAS inhibition in micro- and macroalbuminuric type 2 diabetic patients in this very large-scale cohort of Hong Kong Chinese, a population in which evidence for the beneficial effects of long-term RAAS inhibition is lacking.
In this prospective cohort analysis involving 3773 Chinese type 2 diabetic patients, we have confirmed the prognostic value of albuminuria in terms of all-cause mortality in this population. In support of results from major clinical trials, the beneficial effects of RAAS inhibitors can be translated in clinical practice, especially in high-risk patients with albuminuria, renal insufficiency, history of cardiovascular complications, and retinopathy. Discontinuation of these drugs in high-risk patients was associated with early death compared with those who continued with treatment. The effects of RAAS inhibition in low-risk, normoalbuminuric patients with type 2 diabetes require further evaluation.
We are most grateful to Professor H.H. Parving of Steno Diabetes Centre, Netherland and Dr Benny Zee, director of the Comprehensive Clinical Trial Centre of the Chinese University of Hong Kong for their critical appraisal. Special thanks are extended to Dr Fung Hong, deputy director, and Edwina Chu, senior statistician, of the Hong Kong Hospital Authority Headquarter for their assistance in retrieving the data on clinical outcomes and Albert Cheung, computer officer, of the Centre for Clinical Trials and Epidemiological Research for his assistance in data analysis. We thank all medical and nursing staff at the Prince of Wales Hospital Diabetes Centre for their commitment and dedication in implementing the structured diabetes care protocol and its continuous quality improvement. We are most grateful to Kevin H.M. Yu for computerizing and managing the Prince of Wales Hospital Diabetes Registry.
- Received May 18, 2004.
- Revision received April 6, 2004.
- Accepted June 17, 2004.
Okin PM, Devereux RB, Jern S, Kjeldsen SE, Julius S, Nieminen MS, Snapinn S, Harris KE, Aurup P, Edelman JM, Dahlof B. Regression of electrocardiographic left ventricular hypertrophy by losartan versus atenolol: the Losartan Intervention for Endpoint reduction in Hypertension (LIFE) study. Circulation. 2003; 108: 684–690.
Efficacy of atenolol and captopril in reducing risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 39. UK Prospective Diabetes Study Group. BMJ. 1998; 317: 713–720.
Piwernetz K, Home PD, Snorgaard O, Antsiferov M, Staehr-Johansen K, Krans M. For the DiabCare Monitoring Group of the St. Vincent Declaration Steering Committee. Monitoring the targets of the St. Vincent declaration and the implementation of quality management in diabetes care: the DiabCare initiative. Diabetic Medicine. 1993; 10: 371–377.
Laakso M, Pyorala K. Age of onset and type of diabetes. Diabetes Care. 1985; 8: 114–117.
Mogensen CE, Vestbo E, Poulsen PL, Christiansen C, Damsgaard EM, EiskjÆr H, FrØland A, Hansen KW, Nielsen S, Pedersen MM. Microalbuminuria and potential confounders. Diabetes Care. 1995; 18: 572–581.
Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972; 18: 499–502.