ANGIOTENSIN CONVERTING ENZYME INHIBITORS IN … · Initially normotensive diabetic patients with microalbuminuria or overt albuminuria. Types of intervention ... Lisinopril was used

ANGIOTENSIN CONVERTING ENZYME INHIBITORS IN NORMOTENSIVE DIABETIC PATIENTS WITH MICROALBUMINURIA

Lovell HG

Date of most recent substantive amendment: 01 April 1999

This review should be cited as: Lovell HG. Angiotensin converting enzyme inhibitors in normotensive diabetic patients with microalbuminuria (Cochrane Review). In: The Cochrane Library, Issue 1, 2004.

Chichester, UK: John Wiley & Sons, Ltd.

ABSTRACT

Background Renal disease is a serious complication of diabetes mellitus.

Objectives To examine whether the progression of early diabetic renal disease to end-stage renal failure may be slowed by the use of angiotensin converting enzyme inhibitors for reasons other than their antihypertensive properties, so that they have value in the treatment of normotensive diabetics with microalbuminuria.

Search Strategy Medline was searched for English language reviews and randomised controlled trials. Personal reference lists, and reference lists of retrieved studies were also used.

Selection Criteria Randomised controlled trials with separate identifiable results for initially normotensive diabetic patients, who received angiotensin converting enzyme inhibitors for at least one year and were compared with controls.

Data collection and analysis Meta-analyses were performed on the results of 12 randomised controlled trials with a variety of patient inclusion and exclusion criteria. One further study met all conditions for inclusion but did not provide data in useable form for meta-analyses.

Main Results Albumin excretion rate fell for patients on angiotensin converting enzyme inhibition in 12 of the 13 studies but did so for only two of the 13 groups on placebo. Treatment provided a significant reduction in albumin excretion rate in both insulin dependent diabetes mellitus and non insulin dependent diabetes mellitus. Treatment with either captopril, enalapril or lisinopril reduced albumin excretion rate in comparison with control patients.

A significantly greater lowering of blood pressure was experienced by initially normotensive patients in the angiotensin converting enzyme inhibitor than in the placebo group. Average glycosylated haemoglobin fell a little in the treated patients and rose in the controls, the difference being just significant. The difference in changes in glomerular filtration rate did not reach statistical significance.

Reviewers' conclusions Inhibition of angiotensin converting enzyme can arrest or reduce the albumin excretion rate in microalbuminuric normotensive diabetics, as well as reduce or prevent an increase in blood pressure. But, given the drop in blood pressure in patients on angiotensin converting

enzyme inhibitors, it is not certain that the reduction of albumin excretion rate is due to a separate renal effect. A direct link with postponement of end-stage renal failure has not been demonstrated. There appear to be no substantial side effects.

This review should be cited as: Lovell HG Angiotensin converting enzyme inhibitors in normotensive diabetic patients with microalbuminuria (Cochrane Review). In: The Cochrane Library, Issue 1, 2004. Chichester, UK: John Wiley & Sons, Ltd.

BACKGROUND

GLOSSARY Albustix positive proteinuria: Albustix testing uses a 'stick' that is dipped into a sample of urine to test for protein. It is only positive when there is much more protein than usual. Angiotensin converting enzyme inhibitors: Drugs that inhibit the conversion of angiotensin I to angiotensin II. Diabetic nephropathy: A type of kidney disease that may occur in people with diabetes. End stage renal failure: A deficiency of renal function which is fatal without transplantation or dialysis. Glomerulus: The filtration unit of the kidney. Glomerular filtration rate: Rate at which the kidney filters out waste products, usually 125 ml/minute, although it decreases in renal failure. Glomerular filtration rate is therefore used as a measure of renal failure. Heterogeneity: A measure of the extent to which studies vary. Homogeneous studies are in accord; heterogeneous studies differ from each other. Meta-analysis: Statistical techniques which combine the results of several trials to produce a joint estimate. Microalbuminuria: Higher than normal amounts of protein in the urine but not enough to be detected by albustix. Once persistent, is a sign of early renal disease. Normotensive: Having normal blood pressure. Renal: Relating to the kidney.

MICROALBUMINURIA A patient is said to have overt proteinuria if urine tests show more than 300mg/l of albuminuria. Sensitive assays have shown that levels of albuminuria too low to be detected by conventional dipstick urinalysis can be present for months or years before reaching the 300mg/l level. This phenomenon is termed microalbuminuria and is defined as an excretion rate of 20 to 200 micrograms/min (30 to 300mg/day). Assuming an average daily urine output of one to 1.5 litres, this gives concentrations of 20 to 300 mg/l (Liou 1995).

DIABETIC NEPHROPATHY About 20-35% of diabetic patients develop persistent proteinuria, a decline in glomerular filtration rate, and increased arterial blood pressure, which collectively constitute the clinical syndrome of diabetic nephropathy (Andersen 1983; Borch-Johnsen 1985; Krolewski 1985; Parving 1989; Bilous 1995). The relationship between arterial blood pressure and diabetic nephropathy seems to be a complex one: nephropathy increasing blood pressure and blood pressure accelerating the course of nephropathy (Parving 1993). The presence of nephropathy is closely associated with the increased morbidity and mortality in insulin dependent diabetes mellitus (Knowles 1971; Andersen 1983; Borch-Johnsen 1985; Krolewski 1985; Parving 1988). The high mortality is due to an excess of cardiovascular mortality (Borch-Johnsen 1987) and to end stage renal failure (Knowles 1971; Andersen 1983; Borch-Johnsen 1985; Krolewski 1985). Albuminuric diabetics may be 20 times more likely to die of cardiovascular disease than are non-albuminuric diabetics (Borch-Johnsen et al 1985). On average, death occurs five to 10 years after the start of persistent proteinuria (Knowles 1971; Andersen 1983; Krolewski 1985). Long term angiotensin converting enzyme inhibitor treatment of insulin dependent diabetic patients with nephropathy has been associated with a low rate of decline in kidney function (Lewis 1993; Mulec 1994).

PREDICTION OF NEPHROPATHY The insulin dependent diabetes mellitus subpopulation at risk of developing nephropathy may be identified fairly accurately by the detection of microalbuminuria (Parving 1993). Microalbuminuria is also a strong predictor of all causes of mortality, particularly cardiovascular ones, in non insulin dependent patients (Parving 1993). Incipient diabetic nephropathy is considered to be present if persistent microalbuminuria is found in two of three urine samples collected consecutively, preferably within six months (Parving 1993).

POSSIBLE EFFECTS OF ANGIOTENSIN CONVERTING ENZYME INHIBITOR ANTIHYPERTENSIVE TREATMENT Non-insulin dependent diabetic patients with early nephropathy and mild to moderate hypertension demonstrate a progressive increase in urinary protein excretion. Administration of captopril has resulted in prompt control of hypertension and reversal of the increase in urinary protein excretion (Vora 1996). Antihypertensive treatment prevents overt proteinuria in insulin dependent diabetes mellitus patients with microalbuminuria (Passa 1992). It may be that angiotensin converting enzyme inhibitors contribute to a slowing of progression by contributing an antiproteinuric effect not necessarily related to the effects of systemic blood pressure (Marre 1988; Mathiesen 1991; Valentino 1991; Liou 1995). If so it may be advantageous to use them before the patient has become hypertensive. One meta-analysis, of studies lasting at most one year, showed that the treatment of diabetic patients with angiotensin converting enzyme inhibitors not only effectively reduced high blood pressure but also reduced microalbuminuria/proteinuria and, in addition, exhibited an antihyperglycaemic effect by improving blood sugar levels (Bergemann 1992). Another meta-analysis, of studies lasting four or more weeks, concluded that as angiotensin converting enzyme inhibitors exert a specific antiproteinuric effect even without a change in systemic blood pressure, they are superior to other agents in treating microalbuminuria or overt proteinuria in initially normotensive or mildly hypertensive diabetic patients (Böhlen 1994). In a study (of both normotensives and hypertensives) designed to determine whether captopril was associated with an effect independent of its role as an antihypertensive agent, Lewis et al concluded that the beneficial effect of captopril was not explained by small differences in the level of blood pressure control between captopril and placebo groups (Lewis 1993). They concluded that captopril slows the progression of diabetic nephropathy by a mechanism that is independent of its antihypertensive properties. In proteinuric diabetic patients with normal blood pressure and normal renal function, captopril acutely lowered albuminuria without lowering blood pressure (Elving 1992). Parving et al (Parving 1994) concluded, from a meta-analysis based on randomised controlled trials lasting more than one year, that angiotensin converting enzyme inhibition arrests the progressive rise in albuminuria in normotensive insulin dependent diabetic patients with nephropathy (Parving 1989). But mean arterial pressure fell by 3 (SE 2) mmHg in the captopril treated group and rose by 6 (SE 1) mmHg in the controls. Albuminuria declined by 11% in the captopril group and rose by 55% in the controls. The glomerular filtration rate declined by 3.1 (SE 2.8) ml/min/1.73 m2 with captopril and by 6.4 (SE 3.1) ml/min/1.73 m2 in the controls. No severe side effects occurred. They concluded that, the crucial question, requiring longer term follow up, had not been answered - namely can the rate of decline in the glomerular filtration rate be reduced or even prevented in normotensive diabetic nephropathy by captopril or other hypotensive regimens?

SIDE EFFECTS Known side effects of angiotensin converting enzyme inhibitors are persistent dry cough (EUCLID 1997), in some 5% to 20% of patients (Israili 1992; Molyneaux 1996), and exacerbation of inflammation (Vane 1995). Angiotensin converting enzyme inhibitors may affect potassium levels and should not be used in combination with potassium sparing diuretics or potassium supplements (British Nat For 1997).

OBJECTIVES

It may be possible to slow or halt the progression of diabetes to end-stage renal failure by the use of angiotensin converting enzyme inhibition for reasons other than their antihypertensive properties. If so, it may be desirable to treat normotensive diabetics exhibiting microalbuminuria

with angiotensin converting enzyme inhibitors. This has been investigated by examining published randomised controlled trials on initially normotensive patients.

CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW

Types of studies Randomised controlled trials of angiotensin converting enzyme inhibitors versus placebo, lasting for at least one year.

Types of participants Initially normotensive diabetic patients with microalbuminuria or overt albuminuria.

Types of intervention The angiotensin converting enzyme inhibitors captopril and enalapril were each used in six studies. Lisinopril was used in one study. The meta-analyses exclude one of the captopril studies.

Types of outcome measures Study changes in blood pressure, glycosylated haemoglobin, glomerular filtration rate, and albumin excretion rate. Longer term outcomes were not available.

SEARCH STRATEGY FOR IDENTIFICATION OF STUDIES

See: Cochrane Metabolic and Endocrine Disorders Group search strategy

MEDLINE was searched for English language reviews and randomised controlled trials using the terms 'diabetes & ACE inhibitors'; 'diabetes & renal failure'; and 'diabetes & microalbuminuria'. Interest was confined to randomised controlled trials in which normotensive diabetic patients received angiotensin converting enzyme inhibitors for at least one year. Reference lists of retrieved papers were checked. Personal reference lists were also used. A draft of the first version of this review was presented to a scientific meeting of the Scottish Study Group for the Care of the Young Diabetic which, inter alia, provided an opportunity to check comprehensiveness of retrieved studies. Readers are invited to suggest studies, particularly in other languages, which should be considered for inclusion when this report is next updated.

METHODS OF THE REVIEW

Where publications included appropriate raw data, means and variances were recalculated. Otherwise a measure of location could be either mean or median or midpoint of a confidence interval or range. Variances were sometimes calculated from standard deviations, standard errors, confidence intervals or ranges. In the latter case this involved a, sometimes questionable, assumption of distribution symmetry. Not all publications specify whether '±' is followed by a standard deviation or standard error. Some papers, or groups of papers contained conflicting figures. For some study variables, published standard errors and standard deviations were reconciled by assuming that not all patients contributed to the given data. Data from Hansen et al (1994) were used to supplement those of Viberti et al (1994), but only when there was no risk of double counting. Generally, meta-analyses were carried out using inverse variance weighted, rather than standardised, means but the latter were employed to reduce heterogeneity and logarithmic transformation to counteract skewness. In order to assess changes, an 'effect' is defined as the difference between study duration increment for the angiotensin converting enzyme inhibition and control groups. Unless raw data were available, the standard deviation of change has been

calculated assuming (questionably) that start and end measures of location are independent. This was necessitated by the summary nature of the data examined and leads to a conservative estimate of standardised average change. Since data may be a mixture of arithmetic mean, geometric mean and median using two units of measurement, caution is advisable.

DESCRIPTION OF STUDIES

Twelve studies were found to satisfy the criteria for inclusion in a meta-analysis. Data and descriptions may have been extracted from more than one publication but without multiple counting. The Ahmad 1997 and Laffel 1995 studies were not included in the previous review and that of Sano 1995 now includes extended data. Ravid 1994 has also been extended but only the first five years met our criteria for inclusion. Usual diabetic diets were specified with only Bakris 1994 mentioning salt restriction. The control group for Hallab 1993 was given a diuretic, otherwise controls received no treatment or an inert placebo unless there was more than one control group, in which case only the placebo control group has been included here. If patients became hypertensive, antihypertensive therapy was sometimes added. Definitions of normotensive varied for the studies included in this report, although none had blood pressure exceeding 160/95 mmHg, which is the World Health Organisation definition in the absence of multiple risk factors. Most studies explicitly excluded co-morbidities but the presence of diabetes could make the lower WHO guideline blood pressure of 140/90 mmHg inappropriate. See table 1 for the definitions of 'normotensive' used in the different trials.

Both insulin dependent and non-insulin-dependent diabetes mellitus patients are included in randomised controlled trials usually of captopril (25 to 100 mg/day) or enalapril (5 to 20 mg/day). Lisinopril was used in one study, the dosage being chosen to reduce glomerular filtration rate to at most 140 mL/min. Sample sizes in individual studies were often small. The numbers on angiotensin converting enzyme inhibition and placebo who completed their respective courses were 237 and 234 out of a total of 538 patients admitted to studies in the meta-analyses and 304 and 304 out of 681 patients overall. Most patients initially had microalbuminuria but some had overt albuminuria. When the age of onset of diabetes was specified, it was always less than 41 years.

METHODOLOGICAL QUALITY

The 12 studies had a variety of inclusion/exclusion criteria but all were randomised control studies in which normotensive diabetics received an angiotensin converting enzyme inhibitor for at least one year. Randomisation was reported to have been concealed in only two of these trials. Blinded outcome was mentioned for most but the use of intention-to-treat analyses were less in evidence. Six studies reported steps taken to establish compliance. No study continued until patients experienced end-stage renal failure. Reasons for withdrawal were generally given but in three studies it was not possible to identify the treatment group of patients withdrawing prior to the end of the study. The reported reasons for dropouts show little difference between treated and control patients.

The quality of studies was assessed using a score with seven components, based on Kleijnen 1991. This allocates 30 of the available 100 points for the numbers of patients in the groups compared. Only the study excluded from the meta-analyses and one other were large enough to obtain more than ten of these and most scored zero. Maximum scores available for other aspects were patient characteristics (10), randomisation (20), intervention descriptions (5), double blinding (20), relevance and description of outcomes (10), and checkability of analysis (5). The average score of meta-analyses studies was 55 (range 34-78).

RESULTS

Except where otherwise stated, these results refer to studies included in the meta-analyses.

BASELINE DATA The proportions of males were not significantly heterogeneous and overall, at 54% and 58% did not differ substantially for the treatment and control groups. Duration of diabetes was not significantly heterogeneous or different for the 11 studies with estimates, the treatment and control means being 12.3 and 12.7 years, respectively. Age on admission to the 12 combined studies varied because of patient selection criteria used but, if it were considered reasonable to perform a test, overall the studies would not be found to be significantly heterogeneous. Less surprisingly, because of de facto within study matching, the average ages of all those receiving and those not receiving angiotensin converting enzyme inhibition were the same, at 41 years. Body mass index was available for eight of the studies, some of which stated an upper limit for trial acceptance. Most studies had a (baseline) lower average body mass index in their treatment group; pooled mean body mass indices being 23.8 and 23.2 for patients on placebo and angiotensin converting enzyme inhibition, respectively. The difference has a 95% confidence interval from 0.4 to 0.9 and is highly significant. Those studies that provided data on baseline insulin dose exhibited no significant heterogeneity or difference between treated and untreated groups. The studies do not exhibit significant heterogeneity in initial mean systolic, diastolic or arterial blood pressures even though the latter may have been the subject of varying definitions. The pooled mean initial blood pressures were non-significantly higher for those treated with angiotensin converting enzyme inhibitors (130/78 mmHg) than for those on placebo (128/78 mmHg). But the five mean arterial pressures of those assigned to angiotensin converting enzyme inhibitors had a significantly higher average (96.7 compared to 95.4 mmHg), the excess having 95% confidence interval 0.8 to 2.9 mmHg. The ten studies for which adequate estimates were available did not have significantly heterogeneous or different initial glycosylated haemoglobin, means being 8.9% and 8.7% for the angiotensin converting enzyme inhibition and control groups, respectively. Glomerular filtration rates could be compared for patients entering seven studies. Again there was no significant heterogeneity between the studies and the pooled means were not significantly different for angiotensin converting enzyme inhibitor and control groups, at 125 and 128 ml/min/1.73m2, respectively. Baseline albumin excretion rates that had been measured in mg per day were not significantly heterogeneous or different, averaging 134 and 129 for treatment and control. But those measured in microg per minute (treatment mean 96; control mean 115) were significantly heterogeneous. After logarithmic transformation, to counteract distributional skewness, neither heterogeneity or difference between treated and untreated groups remained significant.

END OF STUDY DATA The seven studies with estimates for systolic and diastolic blood pressure were very highly significantly heterogeneous, as were the five with end of study arterial pressures (p<0.001). This reflects different policies on blood pressure targets and the addition of other antihypertensive treatments. Pooled mean blood pressures for the treated group were lower than for the untreated group for all three types of pressure (127/76 and 97 mmHg for treated; 129/79and101 mmHg for control), but mean blood pressures for individual studies did not all follow the same pattern and the heterogeneity renders the differences nonsignificant. End of study glycosylated haemoglobin showed significant heterogeneity between ten studies (Hansen et al (1994) being used vice Viberti et al (1994)). (Chisquared was 20 with nine degrees of freedom; p<0.025). Values for the angiotensin converting enzyme inhibitor (8.8%) and placebo (8.7%) groups were, however, very close. End of study glomerular filtration rate was homogeneous for five studies compared, also with no significant difference between the two groups of patients (treatment 121; control 119 ml/min/1.73m2). End-of-study albumin excretion rates measured in microg per minute were very significantly heterogeneous but this was entirely removed by (skewness and variance reducing) logarithmic transformation. The overall treatment mean (98 mg per day; 71 microg per minute) was significantly lower than the corresponding control group mean (285 mg per day; 187 microg per minute).

STUDY CHANGES IN BLOOD PRESSURE The studies for which initial and final systolic and diastolic blood pressure were available did not

have homogeneous data (p<0.001) although, curiously, those with mean arterial blood pressure were not so different (p>0.1) despite not having the same definition in all studies. Estimated mean changes in systolic and diastolic pressures during the course of trials were downward for those on angiotensin converting enzyme inhibition (130/78 to 127/76 mmHg) and upwards for controls (128/78 to 129/79 mmHg) despite the use of other antihypertensives in some cases. Mean arterial pressure rose little for the angiotensin converting enzyme inhibitor group (96.7 to 97.0 mmHg) and not as much as for the controls (95 to 101 mmHg). Thus overall a significantly greater antihypertensive effect was experienced by initially normotensive patients in the angiotensin converting enzyme inhibitor than in the placebo group. The relatively large study excluded from meta-analysis (Laffel 1995) had results consistent with the above.

STUDY CHANGES IN GLYCOSYLATED HAEMOGLOBIN Where beginning and end of study data were available, the differential effect on glycosylated haemoglobin was not significantly heterogeneous. The pooled mean effect was just significant with an average fall from 8.89% to 8.85% for the treated groups and a rise from 8.71% to 8.74% in the control groups. Glycaemic control did not differ between treatment and control groups in Laffel 1995.

STUDY CHANGES IN GLOMERULAR FILTRATION RATE Where beginning and end of study data were available, the differential effects were not significantly heterogeneous and pooled mean effects were not significantly different for glomerular filtration rate.

STUDY CHANGES IN ALBUMIN EXCRETION RATE The 12 studies were extremely heterogeneous with regard to the differing effects of placebo and angiotensin converting enzyme inhibition on albumin excretion rate ( chisquared with 11 degrees of freedom = 120; p < 0.001); more so for those measured in microg/min (chisquared with 4 degrees of freedom = 104; p< 0.001) than for those in mg/day (chisquared with 6 degrees of freedom = 15; p < 0.025). This is partly because albumin excretion rate has a very skewed distribution. Logarithmic transformation reduces but does not eliminate the heterogeneity overall (chisquared with 11 degrees of freedom = 30; p < 0.005), or separately for those measured as mg/day (chisquared with six degrees of freedom = 15; p < 0.025), or microg/min (chisquared with four degrees of freedom = 15; p < 0.005). Albumin excretion rate fell for patients on angiotensin converting enzyme inhibition in 11 of the 12 studies but did so for only two of the 12 groups on placebo. The estimated effect of angiotensin converting enzyme inhibition was highly significant (p < 0.001) whether fixed, random, weighted or standardised models were used. The pattern of effects is the same and significant in both measurement subgroups. In seven studies the patients were exclusively described as type 1 or insulin dependent and in four as solely type 2 or non-insulin-dependent. Angiotensin converting enzyme inhibition provided a significant reduction in albumin excretion rate in both groups as well as in the remaining study, which included both types of diabetic patient. Captopril, enalapril and lisinopril were used in five, six and one of the combined studies, respectively. All three angiotensin converting enzyme inhibiting drugs significantly reduced albumin excretion rate in comparison with controls. Captopril was prescribed for insulin dependent subjects in Laffel 1995, where albumin excretion rate increased at an annual rate of 11.8% (95% confidence interval -3.3% to 29.1%) in the placebo group while it declined by 17.9% (95% confidence interval -29.6% to -4.3%) in the captopril group (p=0.004); results consistent with those of the meta-analyses.

DISCUSSION

There appears to be a rise in blood pressure associated with the change from normoalbuminuria through microalbuminuria (Mathiesen 1991) to macroalbuminuria. Patients may start and some remain normotensive, although with higher pressures than normoalbuminurics (Viberti 1984) and an increased proportion of hypertensives (Parving 1993). This is true of both insulin dependent and non insulin dependent diabetes mellitus patients (Parving et al 1993). Eventually diabetic patients become hypertensive and hypertensive diabetic patients benefit from the use of antihypertensive drugs, including angiotensin

converting enzyme inhibitors, since reduction of blood pressure slows the progression to end-stage renal failure (Mogensen 1982; Parving 1983; Parving 1987; Mathiesen 1989). When systemic blood pressure can be lowered by 20%, as is desirable and achievable in more severely hypertensive patients, angiotensin converting enzyme inhibitors, conventional therapy, and several calcium antagonists all have a reported antiproteinuric action (Böhlen 1994).

Intraglomerular hypertension may exist whether or not systemic hypertension is present (Anderson 1988). Assuming that intraglomerular hypertension is an independent contributor to progressive renal injury, drugs that specifically lower intraglomerular pressures independent of their effect on systemic blood pressure might prove especially beneficial in reducing the rate of progression of renal insufficiency. The angiotensin converting enzyme inhibitors are such a group of drugs. The proposed beneficial effect in renal disease is predicated on their ability to predominately alter efferent arteriolar tone and consequently decrease intraglomerular pressure independent of their effect on systemic blood pressure (Sirmon 1991).

The EUCLID study (EUCLID 1997) found that lisinoplil had scarcely any effect on the albumin excretion rate of those starting with normoalbuminuria but considerable effect on the albumin excretion rate of those with microalbuminuria. The purpose of treating microalbuminuric diabetic patients with angiotensin converting enzyme inhibitors is not to reduce albumin excretion rate or to prevent its progression per se, but to prevent a future decline in glomerular filtration rate which otherwise would be expected in the majority of these patients (Mogensen 1995) and lead to end-stage renal failure and cardiovascular death (Passa 1992). In the natural history of diabetes the decline in glomerular filtration rate usually occurs with the development of overt proteinuria. At the stage of normo- or microalbuminuria, glomerular filtration rate is generally well preserved and may even be at the supranormal level (Mogensen et al 1992). A meta-analysis of controlled and noncontrolled studies, lasting at least 6 months, on diabetics and non-diabetics found that, in addition to the long-term beneficial effects of antihypertensive agents, angiotensin converting enzyme inhibitors have beneficial effects on proteinuria and glomerular filtration rate that are independent of blood pressure reductions (Maki 1995).

Definitions of normotensive varied for the studies included in this report (see description of studies), but reduction of albumin excretion rate associated with lowered blood pressure appears to be no respecter of thresholds.

Albuminuria, like hypertension, is defined with reference to an underlying continuous variable. Microalbuminuria was defined either as 30 to 300 mg/day or as 20 to 200 microg/min, usually found in at least two out of three determinations. Albumin excretion rates were obtained by differing techniques, recorded sometimes as mg/24h and sometimes as microg/min, and reported as arithmetic or geometric means or medians, perhaps after transformation. But the beneficial effect of angiotensin converting enzyme inhibition in limiting its rise is clear. None of the studies lasted long enough to establish a relationship with end-stage renal failure, but after an open follow-up extension of their study Ravid et al (Ravid 1996) concluded that treatment with enalapril resulted in an absolute risk reduction of 42% for nephropathy to develop during seven years (95% confidence interval 15% to 69%; p<0.001)

The early use of angiotensin converting enzyme inhibition on normotensive microalbuminuric diabetics appears to produce intermediate benefits on both blood pressure and albumin excretion rate. It has been suggested that treatment at the onset of microalbuminuria is an extremely cost-effective therapy (Rodby 1996).

REVIEWERS' CONCLUSIONS

Implications for practice Inhibition of angiotensin converting enzyme can arrest and even reduce the albumin excretion rate in microalbuminuric normotensive diabetics. This is accompanied by some reduction or prevention of increase in systemic blood pressure and it is not possible to be certain that reduction of albumin excretion rate is due to a separate renal effect. There appear to be no

substantial side effects. A direct link with postponement of end-stage renal failure has not been demonstrated.

Implications for research Studies which are sufficiently longterm to examine directly the relationship between early use of angiotensin converting enzyme inhibitors and end stage renal failure are still desirable. Ideally these should be carried out in representative groups of patients with both insulin dependent diabetes mellitus and non insulin dependent diabetes mellitus.

ACKNOWLEDGEMENTS

The core funding for SHPIC reviews comes from the Management Executive of the Scottish Health Service. The work of converting the initial review into Cochrane format was supported by a research grant from the Chief Scientist Office of the SHS.

POTENTIAL CONFLICT OF INTEREST

None known.

NOTES

This update was done quite a while before the establishment of the Metabolic and Endocrine Disorders Group, but never published. As two new studies are included in the review we feel that it is appropriate to include the new version. Readers should realise, however, that a new update is due now and that further work will be necessary on this review in the near future.

TABLES

Characteristics of included studies Study Ahmad 1997

Methods Five year prospective randomised single-blind placebo controlled trial.

Participants Type 2 nonobese diabetics (120 recruited; 103 analysed). Persistent AER 20-200 microg/min on two consecutive visits, and normal renal function. Age 43-55 years and known duration of diabetes < 15 years. No evidence of non-diabetic renal, systemic, cardiac, hepatic or urinary tract diseases. BMI < 27 kg/m2. GF <= 90 ml/min.

Interventions Three months pretreatment then randomised to enalapril 10 mg/day or placebo (similar but not identical tablets). Diet plus oral hypoglycemic agents (60), diet alone (26), or insulin (17). If systolic BP 145 mmHg or more or diastolic BP 95 mmHg or more treatment with long-acting nifedipine was initiated.

Outcomes Albumin excretion rate, blood pressure, fasting plasma glucose, glycosylated haemoglobin, glomerular filtration rate, renal plasma flow, urinary urea, diabetic retinopathy.

Notes

Allocation concealment

C

Study Bakris 1994

Methods Parallel design lasting 18 months. Seven dropouts; four non compliant, one with dizzyness and orthostasis, two because of schedule conflicts with their jobs. All subjects were placed on a 120 mEq sodium, 1.0 g per kg protein diet.

Participants Fifteen out of 22 normotensive insulin dependent diabetic patients with elevated glomerular filtration rates.

Interventions Patients were prospectively randomised to lisinopril or placebo.

Outcomes Fasting blood sugar; glycosylated haemoglobin A1c; glomerular filtration rate; renal blood flow; 24h urine for microalbumin, sodium, and urea nitrogen; renal ultrasound for renal size; blood pressure.

Notes


B

Study Bilo 1993

Methods Duration one year. Patients were randomised in double blinded order.

Participants Twentyfour normotensive, insulin dependent subjects with microalbuminuria (urinary albumin excretion between 30 and 300 mg in at least two out of three 24-hour urine collections). Onset of diabetes before age 41.

Interventions Patients received either placebo, 50 mg captopril daily or 20 mg nifedipine 'retard' daily

Outcomes Glomerular filtration rate; effective plasma renal flow; blood glucose; urinary albumin loss; glycosylated haemoglobin A1c.

Notes


B

Study Chase 1993

Methods Double blind placebo controlled trial lasting two years. Diabetic nephropathy (albumin excretion rate of 20 to 200microg/min on at least three of four initial overnight urine collections). Seen every three months.

Participants Sixteen type I diabetics with early diabetic nephropathy and normal blood pressures. 13 subjects also had diabetic retinal changes. None had previously received angiotensin converting enzyme inhibition or other hypertensives.

Interventions Captopril 50mg BID or placebo.

Outcomes Eye grades; renal albumin excretion; blood pressure measurements.

Notes Raw data given.


B

Study Hallab 1993

Methods Randomised, double blind, double dummy parallel study for one year after a three month, single blind placebo period

Participants Twentyone normotensive insulin dependent diabetics aged 18-60 with persistent microalbuminuria (30-300 mg/24h)

Interventions Enalapril 20mg or hydrochlorothiazide 25 mg. Patients were instructed to follow a constant isocaloric diet.

Outcomes Urinary albumin excretion; systolic, diastolic and mean arterial pressure; plasma active renin; aldosterone concentrations; effective plasma renal flow; glomerular filtration rate; glycosylated haemoglobin

Notes Raw data given.


A

Study Laffel 1995

Methods A prospective, randomised, double-blind, placebo-controlled (blocks of two), trial lasting two years.

Participants One hundred and fortythree subjects from 26 centres in the United States and Canada aged 14 to 57 years, with IDDM for four to 33 years before 45 years of age with persistent albumin excretion 20 to 200 microg/min confirmed by ketonuria and need for insulin therapy.

Interventions Captopril 50 mg or placebo BID. Diet and insulin management were not altered unless diet exceeded the 50-60% carbohydrate, 30% fat, 20% protein guideline or HbA1c exceeded 11.5%. If blood pressure was 140/90 mmHg or more at two consecutive visits, prazosin or clonidine was prescribed.

Outcomes Progression to proteinuria, albumin excretion rate, blood pressure, serum creatinine, glycihaemoglobin and creatinine clearance. HbA1c, urea excretion and blood pressure were examined for their confounding effects on albumin excretion rate.

Notes Results given as regression analyses and diagrams. No suitable end of study data available for meta-analyses.


D

Study Marre 1988

Methods Randomised trial lasting one year, double blind for six months then single blind, after a run in period of three months

Participants Twenty normotensive type I or type II diabetics aged 20-60 with persistent microalbuminuria (30-300 mg/24h)

Interventions Twenty mg enalapril or placebo. Patients were instructed to follow a constant isocaloric diet with no restriction on sodium intake.

Outcomes Weight; urinary urea excretion; total glycosylated haemoglobin; systolic, diastolic and mean arterial pressure; albumin excretion rate; glomerular filtration rate; total renal resistance; fractional albumin clearance

Notes Some raw data provided.


A

Study Mathiesen 1991

Methods Open randomised controlled study of four years duration.

Participants Fortyfour normotensive insulin dependent diabetics with persistent microalbuminuria ( at least two out of three urines of 30-300 mg/24h). Aged under 50 with diabetes onset below 41.

Interventions Initially 25 mg captopril; increased to 100 mg; then thiazide added versus placebo.

Outcomes Albuminuria; kidney function; systolic, diastolic and arterial blood pressure.

Notes Some raw data given.


B

Study Parving 1989

Methods Open randomised study for one year.

Participants Thirtytwo normotensive insulin dependent diabetics with nephropathy.

Interventions Captopril 25-100 mg/day or no treatment. Usual diabetic diet without sodium or protein restriction.

Outcomes Albuminuria; mean arterial blood pressure; glomerular filtration rate; glycosylated haemoglobin A1c.

Notes


B

Study Ravid 1994

Methods Randomised, double blind, placebo controlled trial lasting five years.

Participants Ninetyfour normotensive aged under 50 type II diabetics with microalbuminuria (30-300 mg/h) and normal renal function.

Interventions For five years, enalapril maleate 10mg/day or placebo. Long acting nifedipine added if blood pressure rose above 140/90 mmHg. After five years, patients could choose, and some received additional ahtihypertensive treatment.

Outcomes Albumin excretion rate; blood pressure; serum creatinine; fasting blood glucose; glycosylated haemoglobin; retinopathy; total cholesterol, HDL, LDL, and triglyceride.

Notes A two year follow-up of the original five year trial was no longer double blind.


B

Study Sano 1995

Methods A four year randomised prospective study.

Participants Sixtytwo normotensive non insulin dependent diabetics aged 45-70 years, with normal renal function, persistent microalbuminuria (AER 20-300 mg/day on 3-4 separate occasions), serum creatinine < 106.1 mumol/l, HbA1c < 10%, and supine blood pressure < 150/90 mmHg.

Interventions Enalapril 5 mg/day or no treatmentl.

Outcomes Blood pressure; 24h urinary albumin excretion; N-acetyl-beta-glucosaminidase, beta2-microglobulin; Creatinine clearance; serum glycosylated haemoglobin A1c; total cholesterol; and triglycerides.

Notes Data from Sano 1995 replace those from Sano 1994 used in the first edition of this review. Twentysix hypertensives, well controlled with nifedipine, were reported in Sano 1994 but excluded from the review.


B

Study Stornello 1989

Methods Double blind randomised design lasting one year.

Participants Sixteen normotensive, age 40-55, type II diabetics with persistent proteinuria

(> 300 mg/24h) on at least three occasions.

Interventions Enalapril 5 mg or placebo.

Outcomes Blood pressure; heart rate; urinary albumin excretion; plasma renin activity and aldosterone; blood glucose; serum fructosamine; urine urea; body weight.

Notes


B

Study Viberti 1994

Methods Multicentre randomised (in blocks of two), double blind, placebo controlled trial of two years duration. Albumin excretion rate in range 20-200 microg/min at screening and at least two of three consecutive determinations.

Participants Ninetytwo normotensive insulin dependent diabetics, aged 18-55 onset before 39, with persistent microalbuminuria. None had been treated with antihypertensives.

Interventions Captopril 50 mg or placebo twice per day. No change in diet. If hypertension later diagnosed, stepped treatment included a betablocker, a diuretic and a vasodilator.

Outcomes Albumin excretion rate; blood pressure; glycosylated haemoglobin; urinary urea nitrogen excretion; glomerular filtration rate.

Notes Compliance monitored. A subset used for some comparisons.


B

Characteristics of excluded studies

Study Reason for exclusion

Anonymous 1996

To avoid double counting. This is a reanalysis of the Laffel and Viberti studies which have been included.

EUCLID 1997

This study included some patients with normoalbuminuria. Also results were in diagrams rather than useable tables and correspondence did not elicit further data.

Heinemann 1996

The study lasted less than a year.

Melbourne 1991

This study used both hypertensive and normotensive patients. It was excluded because the numbers of normotensive patients randomised to perindopril and nifedipine are not given and so data for the normotensives could not be separately identified.

Mogensen 1992a

Only an abstract was available and this provided insufficient information.

Molyneaux 1996

treatment lasted less than a year and not all patients were normotensive.

Trevisan 1995

This study lasted less than a year and included some mildly hypertensive patients.

ADDITIONAL TABLES

Table 01 Definition of 'normotensive' used in trial

Trial Definition

Ahmad 1997 <=140/90 mmHg sitting, after 15 minutes rest on two consecutive examinations

Bakris 1994 <140/90 mmHg implied; > 110/60 mmHg stated

Bilo 1993 <145/90 mmHg on two separate occasions; no antihypertensives

Chase 1993 <=141/90 mmHg; no antihypertensive treatment

Hallab 1993 <160/95 mmHg (supine); no antihypertensive treatment

Laffel 1995 <140/90 mmHg in the absence of antihypertensive therapy

Marre 1988 <160/95 mmHg (supine) for 3 months; no antihypertensives

Mathiesen 1991

diastolic < 95 mmHg; but all subjects were < 160/90 mmHg

Parving 1989 mean of at least 3 <= 150/90 mmHg; no antihypertensives

Ravid 1994 <=140/90 mmHg and mean BP <107 mmHg twice, consecutively

Sano 1995 <150/90 (supine) over a long period; no antihypertensives

Stornello 1989 <140/90 (supine) on 3 consecutive visits; no antihypertensives

Viberti 1994 <160/95 (35+ years), <145/90 (<35 years); no antihypertensives

REFERENCES

References to studies included in this review Ahmad 1997{published data only} Ahmad J, Siddiqui MA, Ahmad H. Effective postponement of diabetic nephropathy with enalapril in normotensive type 2 diabetic patients with microalbuminuria. Diabetes Care 1997;20:1576-81.

Bakris 1994{published data only} Bakris GL, Slataper R, Vicknair N, Sadler R. ACE inhibitor mediated reductions in renal size and microalbuminuria in normotensive, diabetic subjects. J Diab Comp 1994;8:2-6.

Bilo 1993{published data only} Bilo H, Kluitman E, van Ballegooie E, van Loon BJP, Bakker K, Michels B, Gans R, Donker A. Long term use of captopril or nifedipine in normotensive microalbuminuric patients with insulin-dependent diabetes mellitus. Diabetes Research 1993;23:115-122.

Chase 1993{published data only} Chase HP, Garg SK, Harris S, Hoops S, Jackson WE, Holmes DL. Angiotensin-converting enzyme inhibitor treatment for young normotensive diabetic subjects: a two-year trial. Ann Ophthamol 1993;25:284-9.

Hallab 1993{published data only} Hallab M, Gallois Y, Chatellier G, Rohmer V, Fressinaud P, Marre M. Comparison of reduction in microalbuminuria by enalapril and hydrochlorothiazide in normotensive patients with insulin dependent diabetes. Br Med J 1993;306:175-82.

Laffel 1995{published data only} Laffel L, Gans JD, McGill JB. Captopril decreases the rate of progression of renal disease in normotensive, insulin-dependent diabetes mellitus (IDDM) patients with microalbuminuria. [abstract]. J Am Soc Nephrol 1993;3:304.

Laffel LMB, McGill JB, Gans JD. The beneficial effect of angiotensin-converting enzyme inhibition on diabetic nephropathy in normotensive, IDDM patients with microalbuminuria. North American Microalbuminuria Study Group. Am J Med 1995;99(5):497-504.

Marre 1988{published data only}

Marre M, Chatellier G, Leblanc H, Guyene TT, Menard J, Passa P. Prevention of diabetic nephropathy with enalapril in normotensive diabetics with microalbuminuria. Br Med J 1988;297:1092-5.

Marre M, Leblanc H, Suarez L, Guyenne TT, Ménard J, Passa P. Converting enzyme inhibition and kidney function in normotensive diabetic patients with persistent microalbuminuria. Br Med J 1987;294:1448-52.

Mathiesen 1991{published data only} Mathiesen ER, Hommel E, Giese J, Parving HH. Efficacy of captopril in postponing nephropathy in normotensive insulin dependent diabetic patients with microalbuminuria. Br Med J 1991;303:81-7.

Parving 1989{published data only} Parving HH, Hommel E, Nielsen MD, Giese J. Effect of captopril on blood pressure and kidney function in normotensive insulin dependent diabetics with nephropathy. Br Med J 1989;299:533-6.

Ravid 1994{published data only} Ravid M, Lang R, Rachmani R, Lishner M. Long-term renoprotective effect of angiotensin-converting enzyme inhibition in non-insulin-dependent diabetes mellitus; a 7-year follow-up study. Arch intern Med 1996;156:286-9.

Ravid M, Neumann L, Lishner M. Plasma lipids and the progression of nephropathy in diabetes mellitus type II: effect of ACE inhibitors. Kidney Int 1995;47:907-10.

Ravid M, Savin H, Jutrin I, Bental T, Katz B, Lishner M. Long-term stabilizing effect of angiotensin-converting enzyme inhibition on plasma creatinine and on proteinuria in normotensive type II diabetic patients. Ann Intern Med 1993;118:577-81.

Ravid M, Savin H, Jutrin I, Bental T, Lang R, Lishner M. Long-term effect of ACE inhibition on development of nephropathy in diabetes mellitus type II. Kidney Int 1994;45(suppl 45):S161-S164.

Sano 1995{published data only} Sano T, Hotta N, Kawamura T, Matsumae H, Chaya S, Sasaki H, Nakayama M, Hara T, Matsuo S, Sakamoto N. Effects of long-term enalapril treatment on persistent microalbuminuria in normotensive type 2 diabetic patients; results of a 4-year, prospective, randomized study. Diabetic Medicine 1996;13:120-4.

Sano T, Kawamura T, Matsumae H, Sasaki H, Nakayama M, Hara T, Matsuo S, Hotta N, Sakamoto N. Effects of long-term enalapril treatment on persistent microalbuminuria in well-controlled hypertensive and normotensive NIDDM patients. Diabetes Care 1994;17:420-4.

Stornello 1989{published data only} Stornello M, Valvo EV, Scapellato L. Angiotensin converting enzyme inhibition in normotensive type II diabetics with persistent mild proteinuria. J Hypertens 1989;7(suppl 6):S314-S315.

Viberti 1994{published data only} Hansen KW, Klein F, Christensen PD, Sorensen K, Andersen PH, Moller J, Pedersen EB, Christiansen JS, Mogensen CE. Effects of captopril on ambulatory blood pressure, renal and cardiac function in microalbuminuric type 1 diabetic patients. Diabete Metab 1994;20:485-93.

Viberti G, Mogensen CE, Groop LC, Pauls JF, for the European Microalbuminuria Captopril Study Group. Effect of captopril on progression to clinical proteinuria in patients with insulin-dependent diabetes mellitus and microalbuminuria. JAMA 1994;271:275-9.

References to studies excluded from this review Anonymous 1996 The Microalbuminuria Captopril Study Group. Captopril reduces the risk of nephropathy in IDDM patients with microalbuminuria. Diabetologia 1996;39:587-93.

EUCLID 1997 The EUCLID Study Group. Randomised placebo-controlled trial of lisinopril in normotensive patients with insulin-dependent diabetes and normoalbuminuria or microalbuminuria. Lancet 1997;349:1787-92.

Heinemann 1996 Heinemann L, Janicke I, Bender R, Berger M, Sawicki PT. Effects of enalapril and nitrendipine on exercise albuminuria in normotensive type I diabetic patients with incipient nephropathy. Horm Metab Res 1996;28:549-52.

Melbourne 1991 Melbourne diabetic nephropathy study group. Comparison between perindopril and nifedipine in hypertensive and normotensive diabetic patients with microalbuminuria. Br Med J 1991;302:210-6.

Mogensen 1992a

Mogensen CE. Captopril (c) delays progression to overt renal disease in insulin-dependent diabetes mellitus (IDDM) patients with microalbuminuria [abstract]. J Am Soc Nephrol 1992;3(suppl 3):336.

Molyneaux 1996 Molyneaux LM, Willey KA, Yue DK. Indapamide is as effective as captopril in the control of microalbuminuria in diabetes. J Cardiovasc Pharmacol 1996;27:424-7.

Trevisan 1995 Trevisan R, Tiengo A. For the north-east italy microalbuminuria study group. Effect of low-dose ramipril on microalbuminuria in normotensive or mild hypertensive non-insulin-dependent diabetic patients. Am J Hypertens 1995;8:876-83.

Additional references Andersen 1983 Andersen AR, Christianisen JS, Andersen JK, Kreiner S, Deckert T. Diabetic nephropathy in type I (insulin dependent) diabetes; an epidemiological study. Diabetologia 1983;25:496-501.

Anderson 1988 Anderson S, Brenner BM. Therapeutic benefit of converting enzyme inhibitors in progressive renal disease. Am J Hyper 1988;1(suppl):380-3.

Bergemann 1992 Bergemann R, Wohler D, Weidmann P, Betzin J, Nawrath T. Verbesserte glucoseeinstellung und mikroalbuminurie/proteinurie bei diabetikern unter ACE-hemmer-behandlung: eine metaanalyse publizieter studien aus den jahren 1985 bis 1990. Schweiz Med Wschr 1992;122:1369-76.

Bilous 1995 Bilous R. Diabetes: should you be screening for microalbuminuria?. Practitioner 1995;239:343-5.

Borch-Johnsen 1985 Borch-Johnsen K, Andersen PK, Deckert T. The effect of proteinuria on relative mortality in type I (insulin dependent) diabetes mellitus. Diabetologia 1985;28:590-6.

Borch-Johnsen 1987 Borch-Johnsen K, Kreiner S. Proteinuria - value as a predictor of cardiovascular mortality in insulin dependent diabetes mellitus. Br Med J 1987;294:1651-4.

British Nat For 1997 British Medical Association and the Royal Pharmaceutical Society of Great Britain. British National Formulary. Vol. 33, , 1997:86-7.

Böhlen 1994 Böhlen L, de Courten M, Weidmann P. Comparative study of the effect of ACE-inhibitors and other antihypertensive agents on proteinuria in diabetic patients. Am J Hypertens 1994;7:84S-92S.

Elving 1992 Elving LD, Wetzels JFM, DE Noble E, Hoitsma AJ, Berden JHM. Captopril acutely lowers albuminuria in normotensive patients with diabetic nephropathy. Am J Kidney Dis 1992;20:559-63.

Israili 1992 Israili ZH, Hall WD. Cough and angioneurotic edema associated with angiotensin-converting enzyme inhibitor therapy: a review of the literature and pathophysiology. Ann Intern Med 1992;117:234-42.

Kleijnen 1991 Kleijnen J, Knipschild P, ter Reit G. Clinical trials of homeopathy. Br Med J 1991;302:316-23.

Knowles 1971 Knowles HC Jr. Long term juvenile diabetes treated with unmeasured diet. Trans Assoc Am Physicians 1971;84:95-101.

Krolewski 1985 Krolewski AS, Warram JH, Christleib AR, Busick EJ, Kahn CR. The changing natural history of nephropathy in type I diabetes. Am J Med 1985;78:785-94.

Lewis 1993 Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of angiotensin-converting-enzyme inhibition on diabetic nephropathy. N Engl J Med 1993;329:1456-62.

Liou 1995

Liou HH, Huang HP, Campese VM. Effect of long-term therapy with captopril on proteinuria and renal function in patients with non-insulin-dependent diabetes and with non-diabetic renal diseases. Nephron 1995;69:41-8.

Maki 1995 Maki DD, Ma JZ, Louis TA, Kasiske BL. Long-term effects of antihypertensive agents on proteinuria and renal function. Arch Intern Med 1995;155:1073-80.

Mathiesen 1989 Mathiesen ER, Borch-Johnsen K, Jenses DV, Deckert T. Improved survival in patients with diabetic nephropathy. Diabetologia 1989;32:884-6.

Mogensen 1982 Mogensen CE. Long-term antihypertensive treatment inhibiting progression of diabetic nephropathy. Br Med J 1982;285:685-8.

Mogensen 1992b Mogensen CE, Hansen KW, Osterby R, Damsgaard EM. Blood pressure elevation versus abnormal albuminuria in the genesis and prediction of renal disease in diabetes. Diabetes Care 1992b;15:1192-1204.

Mogensen 1995 Mogensen CE. Should ACE inhibitors be used in normotensive micro-albuminuric IDDM?. Diabetes Reviews International 1995;3:2-6.

Mulec 1994 Mulec H, Johnsen SA, Björck S. Long-term enalapril treatment in diabetic nephropathy. Kidney Int 1994;45(suppl 45):S141-S144.

Parving 1983 Parving H-H, Andersen AR, Smidt UM, Svendsen PA. Early aggressive antihypertensive treatment reduces rate of decline in kidney function in diabetic nephropathy. Lancet 1983;1:1175-9.

Parving 1987 Parving HH, Andersen AR, Smidt UH, Hommel E, Mathiesen ER, Svendsen PA. Effect of antihypertensive treatment on kidney function in diabetic nephropathy. Br Med J 1987;294:1443-7.

Parving 1988 Parving HH, Hommel E, Mathiesen E, et al. Prevalence of microalbuminuria, arterial hypertension, retinopathy, and neuropathy in patients with insulin dependent diabetes. Br Med J 1988;296:156-60.

Parving 1993 Parving HH, Rossing P. Calcium antagonists and the diabetic hypertensive patient. Am J Kidney Dis 1993;21(suppl 3):47-52.

Parving 1994 Parving HH, Rossing P. The use of antihypertensive agents in prevention and treatment of diabetic nephropathy. Current opinion in nephrology and hypertension 1994;3:292-300.

Passa 1992 Passa P. Does antihypertensive treatment prevent progression of microalbuminuria to overt proteinuria in insulin-dependent diabetic patients?. J Am Soc Nephrol 1992;3:S91-S96.

Rodby 1996 Rodby RA, Firth LM, Lewis EJ. The collaborative study group. An economic analysis of captopril in the treatment of diabetic nephropathy. Diabetes Care 1996;19:1051-61.

Sirmon 1991 Sirmon MD, Kirkpatrick WG. Diabetic nephropathy: new directions in management. Renal Failure 1991;13:51-9.

Valentino 1991 Valentino VA, Wilson MD, Weart W, Bakris GL. A perspective on converting enzyme inhibitors and calcium channel antagonists in diabetic renal disease. Arch Intern Med 1991;151:2367-72.

Vane 1995 Vane J. Trumping the ACE. Lancet 1995;346:916.

Viberti 1984 Viberti CG, Wiseman MJ, Bending JJ. Diabetic nephropathy: a preventable complication?. Proc EDTA-ERA 1984;21:621-9.

Vora 1996

Vora JP, Leese GP, Peters JR, Owens DR. Longitudinal evaluation of renal function in non-insulin-dependent diabetic patients with early nephropathy: effects of angiotensin-converting enzyme inhibition. J Diab Comp 1996;10:88-93.

GRAPHS

To view a graph or table, click on the outcome title of the summary table below. To view graphs using MetaView, click on the "Show metaview" link at the top of the graph.

01 ACE inhibition versus placebo

Outcome title No. of studies

No. of participants Statistical method Effect size

01 baseline systolic blood pressure

9 382 Weighted Mean Difference (Fixed) 95% CI

0.80 [-0.48, 2.09]

02 baseline diastolic blood pressure


-0.39 [-1.45, 0.66]

03 baseline mean arterial blood pressure


1.87 [0.84, 2.89]

04 baseline glycosylated haemoglobin


0.02 [-0.16, 0.20]

05 baseline glomerular filtration rate


-1.81 [-5.93, 2.31]

06 baseline albumin excretion rate


-7.92 [-15.10, -0.74]

07 sex (proportion of males)

11 464 Peto Odds Ratio 95% CI

0.87 [0.60, 1.26]

08 years of diabetes 11 508 Weighted Mean Difference (Fixed) 95% CI

-0.19 [-0.66, 0.28]

09 age at start of study 12 524 Weighted Mean Difference (Fixed) 95% CI

-0.52 [-1.26, 0.22]

10 body mass index 8 451 Weighted Mean Difference (Fixed) 95% CI

-0.68 [-0.92, -0.45]

11 baseline insulin dose


0.04 [-0.03, 0.11]

12 end of study systolic 7 294 Weighted Mean -3.74 [-5.51,

blood pressure Difference (Fixed) 95% CI

-1.98]

13 end of study diastolic blood pressure


-5.67 [-7.04, -4.30]

14 end of study mean arterial pressure


-2.78 [-4.03, -1.53]

15 end of study mean glycosylated haemoglobin


-0.09 [-0.23, 0.06]

16 end of study glomerular filtration rate


0.91 [-4.02, 5.84]

17 end of study albumin excretion rate


-153.89 [-168.90, -138.89]

18 change in systolic blood pressure


-3.49 [-5.77, -1.22]

19 change in diastolic blood pressure


-5.89 [-7.55, -4.22]

20 change in mean arterial blood pressure


-3.98 [-5.67, -2.30]

21 change in glycosylated haemoglobin


-0.36 [-0.60, -0.12]

22 change in glomerular filtration rate


0.71 [-6.10, 7.52]

23 change in albumin excretion rate


-136.95 [-154.20, -119.69]

25 ln change in albumin excretion rate


-6.53 [-7.35, -5.70]

26 ln change in AER by type of diabetes


-6.53 [-7.35, -5.70]

27 ln change in AER by drug used


-6.33 [-7.16, -5.51]

29 ln end of study AER 12 503 Weighted Mean Difference (Fixed) 95% CI

-1.09 [-1.88, -0.30]

30 ln baseline albumin excretion rate


0.00 [-0.67, 0.67]

31 change in systolic BP by duration of study


-3.49 [-5.77, -1.22]

32 change in diastolic BP by duration of study


-5.89 [-7.55, -4.22]

33 ln change in AER by duration of study


-6.60 [-7.43, -5.78]

COVER SHEET

Title Angiotensin converting enzyme inhibitors in normotensive diabetic patients with microalbuminuria

Reviewer(s) Lovell HG

Contribution of reviewer(s) Information not supplied by reviewer

Issue protocol first published Information not available

Issue review first published Information not available

Date of most recent amendment Information not available

Date of most recent SUBSTANTIVE amendment

01 April 1999

Most recent changes Information not supplied by reviewer

Date new studies sought but none found

Information not supplied by reviewer

Date new studies found but not yet included/excluded


Date new studies found and included/excluded


Date reviewers' conclusions section amended


Contact address Mr Howard Lovell Retired 2 Ardencaple Drive Helensburgh G84 8PS Dunbartonshire UK tel: 01436 673679

Cochrane Library number CD002183

Editorial group Cochrane Metabolic and Endocrine Disorders Group

Editorial group code HM-ENDOC

SOURCES OF SUPPORT

External sources of support

• No sources of support supplied

Internal sources of support

• Scottish Health Purchasing Information Centre UK

Index Terms

Medical Subject Headings (MeSH) Albuminuria [complications] [drug therapy]; Angiotensin-Converting Enzyme Inhibitors [therapeutic use]; Diabetes Mellitus [complications] [drug therapy]; Diabetic Nephropathies [prevention & control]; Kidney Failure, Chronic [prevention & control]; Randomized Controlled Trials Mesh check words: Human

Copyright © 2004 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. All rights reserved. No part of The Cochrane Library may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, UK, or emailed to [email protected], or faxed to (+44) 1243 770620.

Documents

ANGIOTENSIN CONVERTING ENZYME INHIBITORS IN … · Initially normotensive diabetic patients with microalbuminuria or overt albuminuria. Types of intervention ... Lisinopril was used