Effect of Antihypertensive Therapy on the Kidney in Patients with Diabetes: A Meta-Regression Analysis

  1. Bertram L. Kasiske, MD;
  2. Roberto S. N. Kalil, MD;
  3. Jennie Z. Ma, MS;
  4. Minjen Liao, MD; and
  5. William F. Keane, MD
  1. From Hennepin County Medical Center, Minneapolis, Minnesota. Requests for Reprints: Bertram L. Kasiske, MD, Division of Nephrology, Department of Medicine, Hennepin County Medical Center, 701 Park Avenue, Minneapolis, MN 55415. Acknowledgments: The authors thank Ms. Jan Lovick and Ms. Dee Lunzer for help in preparing the manuscript; and Thomas Louis, PhD, Chairman, Department of Biostatistics, University of Minnesota, for helpful suggestions and guidance.
     
    Next Section

    Abstract

    Objective: To assess the relative effect of different antihypertensive agents on proteinuria and renal function in patients with diabetes.

    Data Sources: We used MEDLINE and bibliographies in recent articles to identify studies of the effects of antihypertensive agents on renal function in patients with diabetes.

    Study Selection: We selected 100 controlled and uncontrolled studies that provided data on renal function, proteinuria, or both, before and after treatment with an antihypertensive agent.

    Data Extraction: Data on blood pressure, renal function, proteinuria, patient characteristics (for example, age, sex, and type of diabetes), and study design (for example, random allocation and the use of a placebo) were extracted from selected studies.

    Data Synthesis: Multiple linear regression analysis indicated that angiotensin-converting enzyme (ACE) inhibitors decreased proteinuria independent of changes in blood pressure, treatment duration, and the type of diabetes or stage of nephropathy, as well as study design (P < 0.0001). Reductions in proteinuria from other antihypertensive agents could be entirely explained by changes in blood pressure. Blood pressure reduction in itself was associated with a relative increase in glomerular filtration rate (regression coefficient [±SE], 3.70 ±.92 mL/min for each reduction of 10 mm Hg in mean arterial pressure; P = 0.0002); however, compared with other agents, ACE inhibitors had an additional favorable effect on glomerular filtration rate that was independent of blood pressure changes (3.41 ± 1.71 mL/min; P = 0.05).

    Conclusion: Angiotensin-converting enzyme inhibitors can decrease proteinuria and preserve glomerular filtration rate in patients with diabetes. These effects occur independent of changes in systemic blood pressure.

    The incidence and prevalence of renal failure from diabetes have grown alarmingly during the past decade [1]. The incidence of hypertension associated with nephropathy is high in patients with type I and type II diabetes [2, 3], and hypertension can contribute to a progressive deterioration in renal function in patients with diabetic nephropathy [4, 5]. Indeed, antihypertensive treatment has been shown to retard the rate of declining renal function in patients with diabetic nephropathy [6-8]. Moreover, experiments in animal models of diabetes have shown that antihypertensive agents may vary in their ability to prevent renal injury [9-20].

    The effect of different antihypertensive agents on glomerular filtration rate and proteinuria in patients with diabetes has been examined in many clinical trials, with conflicting results. Several possible reasons for the differing results can be advanced. First, different agents were studied, including various angiotensin-converting enzyme (ACE) inhibitors, calcium antagonists, β-blockers, and other agents alone or in combination. Second, study samples differed with regard to the presence or absence of hypertension, the type of diabetes, and the stage of nephropathy. Third, although some investigations were well controlled, most were not, and the duration of treatment varied from days to years. To weigh and interpret the different results of these studies, we codified the data from each study and carried out a “meta-regression” analysis [21]. We used multiple linear regression to determine the extent to which differences in agents, patient characteristics, study design features, and the duration of treatment might have affected the renal response to antihypertensive therapy in patients with diabetes.

    Previous SectionNext Section

    Methods

    Studies

    Using MEDLINE and bibliographies in recent publications to identify clinical trials that examined the effects of antihypertensive agents on blood pressure, glomerular filtration rate, renal blood flow, urine protein excretion, and urine albumin excretion in patients with diabetes, we found 101 studies with extractable data [4,6-8,22-118]. For nine studies, data on blood pressure were estimated from figures. In one instance, data on urine albumin excretion were extracted from a figure. We excluded one study, much larger than the others, that included only blood pressure response and did not indicate whether the standard deviation or standard error was reported [90]. Of the 100 studies analyzed, 87 have been reported since 1985. Fifty-two studies had only one experimental group, 35 had two groups, 9 had three groups, 3 had four groups, and 1 had seven groups. The analysis included 168 experimental groups, totalling 2494 patients.

    Study End Points

    The end points we examined included mean arterial pressure (calculated as one third of the pulse pressure plus the diastolic pressure [mm Hg]); glomerular filtration rate (mL/min); renal blood flow (mL/min); filtration fraction (glomerular filtration rate divided by the renal blood flow); urine protein excretion defined as either albumin or, in its absence, total protein excretion (mg/mL); and urine albumin excretion alone (mg/24 h). Each end point, except for urine protein and albumin excretion, was defined by the change from baseline, that is, the value after treatment or placebo minus the value at baseline within each experimental group. Because the frequency distributions of the changes in urine protein and albumin excretion were not normal, these end points were transformed using the natural logarithm (ln). For both urine protein excretion and albuminuria, the end point analyzed was the natural logarithm of the value after treatment or placebo minus the natural logarithm of the baseline value (ln [treatment] −ln [baseline]).

    We estimated the variances for each end point using the standard deviations of the values before and after treatment for each experimental group [119]. Thus, Var(X − Y) = Var(X) + Var(Y) −2 γ xy x radical(Var[X]) x radical (Var[Y]), where X and Y were the means of the treatment and baseline measurements, respectively, and γXY was the correlation coefficient between X and Y estimated from the experimental group means across all studies. Complete data for calculating the end point and its variance were available as follows: mean arterial pressure, 147 experimental groups; glomerular filtration rate, 74 groups; renal blood flow, 35 groups; filtration fraction, 34 groups; urine protein excretion, 78 groups; and urine albumin excretion alone, 55 groups.

    Independent Explanatory Variables

    To examine the extent to which differences in study end points could be explained by differences in the agents used, characteristics of the patients studied, or study design features, several independent explanatory variables were used to describe each experimental group. The specific agent (for example, enalapril, captopril, nifedipine), the class of agent (for example, ACE inhibitor, calcium antagonist, β-blocker), and the duration of treatment were used to characterize each group. The patient characteristics examined were the type of diabetes (type I or type II), the presence or absence of diagnosed hypertension, the World Health Organization (WHO) stage of diabetic nephropathy, the duration of diabetes, age, and gender. Study design features that indicated how well each experimental group had been controlled were also examined. These design features included whether the study allocated patients randomly, used a single or double-blinded design, included the use of a placebo, incorporated a wash-in phase, used an untreated control group (placebo or not), or used stratification for hypertension, diabetes stage, or other patient characteristics. In addition, publication of the study in abstract form only was also included as an independent variable to explain differences among the results of the studies.

    In general, data for most of the independent explanatory variables were complete for most of the experimental groups. Exceptions included variables describing the type, stage, and duration of diabetes, and the proportion of male patients in each group: Data on type, stage, and duration of diabetes were available for 153 (91%), 102 (61%), and 81 (48%) experimental groups, respectively; data on age and gender were available for 112 (67%) and 122 (73%) experimental groups, respectively.

    Univariate Analysis

    We examined the mean effect of each antihypertensive agent or each combination of agents (included in two or more experimental groups) on mean arterial pressure, glomerular filtration rate, renal blood flow, urine protein excretion, and urine albumin excretion by pooling all experimental groups using the same agent or combination. We also examined the effect of each class of agent for each of the study end points. The mean effect of each different agent or class of agent was compared with the mean effects of the other agents or classes as well as with those of “no antihypertensive drug treatment.” The differences between these agent-specific groups were tested using analysis of variance.

    Multiple Linear Regression Analysis

    We used multiple linear regression analysis to determine the relative magnitude and independent effects of different agents, treatment duration, patient characteristics, and study features on each of the study end points. We analyzed the change in mean arterial pressure, glomerular filtration rate, urine protein excretion, and urine albumin excretion separately. For each of these dependent variables, we tested several independent explanatory variables. The change in blood pressure was also included as a possible independent explanatory variable to determine the extent to which changes in glomerular filtration rate, urine protein excretion, and urine albumin excretion could be explained by changes in systemic blood pressure. Regression models were weighted by the inverse of the variance of the change in measured end point.

    For the regression analysis of each end point, we included all experimental groups that had data on the effect of treatment (treatment or placebo value minus the baseline value) and data that allowed an estimate of the variance of the change in the end point. In the case of independent variables for which data were missing, models were tested using only cases with complete data for all variables, and separately using mean substitution of missing data. When it was found that a variable with missing data did not enter into either of these models, the variable was dropped from subsequent models to permit the inclusion of all experimental groups. Thus, for each end point that was analyzed, the final regression model included virtually all experimental groups that measured that end point.

    Meta-Analysis of Randomized, Controlled Trials

    Only 12 randomized, controlled trials could be subjected to a meta-analysis of treatment effects calculated using a parallel control group. Because all but 1 of the 12 trials examined the effects of an ACE inhibitor [101], only treatment effects of ACE inhibitors could be analyzed [22, 30-33, 35, 44, 49-51, 57]. The treatment effect of each study was defined as Δ = ΔT − ΔC, where ΔT and ΔC were the changes in end points for the treatment and control groups, respectively. For example, in the case of mean arterial pressure, Δ = (mean arterial pressure after treatment − mean arterial pressure before treatment) −(mean arterial pressure after placebo − mean arterial pressure before placebo). In the case of urine albumin excretion, values were transformed using the natural logarithm before calculating the treatment effect. As in the regression analysis, treatment effects were weighted by the inverse variance. The pooled treatment effect and pooled estimates of 95% CIs were calculated as described by Cappuccio and coworkers [120]. The results were considered significant at α = 0.05 when the 95% CI did not include 0.

    Statistical Conventions

    Values in the text and tables are expressed as the mean ±SD unless otherwise indicated. All differences were considered significant at P < 0.05. The analysis was carried out using the Statistical Package for the Social Sciences (SPS) software [121].

    Previous SectionNext Section

    Results

    Study Characteristics

    The mean number of patients in an experimental group was 15 ± 10 (range, 4 to 111 patients). The mean age of the patients in each group was 48 ± 12 years (range, 14 to 72 years). The mean proportion of men in each group was 54% ± 21% (range, 0% to 100%). The duration of the study was more than 6 months in 27% of the experimental groups and more than 1 year in 13% of the groups. The patients studied all had type I diabetes in 49% of the groups, type II in 32%, and a mixture of both types in 11%; the type of diabetes was not indicated in 9% of the groups. Patients in 17% of the groups had WHO stage 3 (microalbuminuria) diabetic nephropathy, whereas patients in 35% of the groups had clinical nephropathy (stage 4 or 5). The WHO stage was not indicated in 39% of the groups. Patients in 78% of the experimental groups had hypertension. Fifty percent of the groups were randomized, 15% were randomized and controlled, 36% were blinded in some fashion, 18% were placebo-controlled, 32% used a placebo-controlled wash-in phase, and 17% stratified patients in experimental groups by clinical characteristics. In 13% of studies, data had been published only in abstract form.

    Antihypertensive Agents

    An ACE inhibitor was investigated in 46% of the experimental groups. Captopril and enalapril were investigated in 26% and 15% of the groups, respectively. Calcium antagonists were used in 14% of the experimental groups and β-blockers in 21% of the groups. In 77% of the experimental groups that were receiving an antihypertensive agent, that agent was used alone. Only 8% of the groups received either no treatment or placebo.

    Blood Pressure

    Treatments with ACE inhibitors, calcium antagonists, and β-blockers had a similar effect on mean arterial pressure Table 1 and Figure 1. A small decline in blood pressure was observed in the untreated control groups. The reduction in pressure with the different ACE inhibitors was also similar (data not shown). The reduction in blood pressure was less in studies that were randomized, were placebo-controlled, or included a separate control group. Blood pressure decreased by 12 ± 7 mm Hg in studies that randomly allocated patients and by 15 ± 9 mm Hg in studies that did not randomly allocate patients (P < 0.05). Antihypertensive agents reduced blood pressure by 6 ± 4 mm Hg in randomized, controlled studies and by 14 ± 8 mm Hg in uncontrolled studies (P < 0.05). Interestingly, the magnitude of blood pressure reduction tended to be greater when the results appeared only in abstract form (17 ± 11 mm Hg) compared with the reduction found in studies reported in a peer-reviewed journal (13 ± 7 mm Hg) (P = 0.06).

    View this table:
    Table 1. Effects of Antihypertensive Agents on Changes in Blood Pressure, Renal Function, Proteinuria, and Albuminuria*
    Figure 1. Not shown are three groups that received combination therapy with an angiotensin-converting enzyme inhibitor and a calcium antagonist, two groups that received an angiotensin-converting enzyme inhibitor plus a β-blocker, and one group that received a calcium antagonist plus a β-blocker. Bars represent 95% CIs.
    View larger version:
    Figure 1. Not shown are three groups that received combination therapy with an angiotensin-converting enzyme inhibitor and a calcium antagonist, two groups that received an angiotensin-converting enzyme inhibitor plus a β-blocker, and one group that received a calcium antagonist plus a β-blocker. Bars represent 95% CIs. The change in mean arterial blood pressure for each experimental group included in the regression analysis.

    The regression analysis, which simultaneously took into account the effects of different agents as well as patient and study characteristics, confirmed that the mean arterial-pressure-lowering effects of the different antihypertensive agents were similar (Table 2). Not surprisingly, patients with hypertension had a greater reduction in arterial pressure than did patients who were normotensive. Patients with type II diabetes also had a greater reduction in blood pressure relative to patients with type I diabetes. Studies in which patients were randomly allocated to experimental groups showed relatively less blood pressure reduction. Other explanatory variables failed to independently influence the reported changes in mean arterial pressure. In particular, none of the specific agents or classes of agents had a greater effect on mean arterial pressure than any other agent. Similarly, the duration of therapy and the WHO stage of diabetic nephropathy had no independent effect on the response to treatment. Thus, with regard to the magnitude of the change in mean arterial pressure, it did not matter what agents were used, how long therapy lasted, or whether the patients had early or more advanced nephropathy.

    View this table:
    Table 2. Independent Correlates of Changes in Blood Pressure

    Glomerular Filtration Rate

    Glomerular filtration rate was measured by creatinine clearance in 19 experimental groups (26%), by inulin in 8 (11%), by Chromium-51-EDTA in 23 (31%), by iothalamate in 7 (9%), and by other techniques in 17 (23%). The method used for filtration rate estimation did not appear to affect the results. Univariate analysis showed no statistically significant differences in the effects of different antihypertensive agents on glomerular filtration rate (see Table 1; Figure 2). Angiotensin-converting enzyme inhibitors and calcium antagonists appeared to have a more favorable effect on glomerular filtration rate compared with other antihypertensive agents or “no treatment,” but these differences were not statistically significant (see Table 1). However, in the regression analysis, ACE inhibitors were uniquely associated with a relative increase in filtration rate that was independent of effects from mean arterial pressure reduction, treatment duration, and other study variables (Table 3). The blood pressure reduction and the presence of hyperfiltration (glomerular filtration rate >110 mL/min) were also associated with relative increases in renal function. However, of all the specific agents and classes of agents examined, only ACE inhibitors exerted effects on renal function that were independent of mean arterial pressure reduction.

    View this table:
    Table 3. Independent Correlates of Changes in Glomerular Filtration Rate*
    Figure 2. Not shown is one group that received combination therapy with an angiotensin-converting enzyme inhibitor and a calcium antagonist and one group that received combination therapy with a calcium antagonist and a β-blocker. Bars represent 95% CIs.
    View larger version:
    Figure 2. Not shown is one group that received combination therapy with an angiotensin-converting enzyme inhibitor and a calcium antagonist and one group that received combination therapy with a calcium antagonist and a β-blocker. Bars represent 95% CIs. The change in glomerular filtration rate for each experimental group included in the regression analysis.

    Renal Blood Flow

    Relatively few studies measured renal blood flow (see Table 1). A small increase in renal blood flow occurred in patients treated with antihypertensive agents compared with controls, but this difference was not statistically significant. Overall, the effects of different agents on renal blood flow appeared to be similar, but few studies examined the effects of agents other than ACE inhibitors. Similarly, the different agents did not differ regarding their effect on filtration fraction, nor was a difference in effect seen between treated patients and controls (see Table 1). In the regression analysis, none of the explanatory variables were associated with changes in renal blood flow or filtration fraction. However, the number of studies analyzed may have been too small to draw meaningful conclusions from the multiple linear regression analysis of renal blood flow and filtration fraction.

    Proteinuria and Urine Albumin Excretion

    The greatest reductions in urine protein and urine albumin excretion occurred in patients treated with ACE inhibitors (see Table 1; Figure 3). Moreover, the effects of different ACE inhibitors on urine protein and urine albumin excretion were not significantly different. In particular, no difference was observed between the effects of enalapril and captopril, despite their structural differences [data not shown]. Interestingly, patients with hyperfiltration (glomerular filtration rate >110 mL/min) tended to have more marked reductions in protein and albumin excretion than patients who had a lower filtration rate (≤ 110 mL/min) (urine protein excretion, 0.49 ± 0.81 compared with 0.61 ± 0.38, P = NS; urine albumin excretion, 0.29 ± 0.59 compared with 0.61 ± 0.38, P = 0.07; changes were calculated as ln [treatment] − ln [baseline]). In general, the effects of calcium antagonists, β-blockers, and other antihypertensive agents on urine protein excretion and urine albumin excretion were intermediate between the effects of “no treatment” and the effects of ACE inhibitors (see Table 1).

    Figure 3. Not shown is one group that received combination therapy with an angiotensin-converting enzyme inhibitor and a calcium antagonist. ln (treatment/baseline) = ln (treatment) − ln (baseline).
    View larger version:
    Figure 3. Not shown is one group that received combination therapy with an angiotensin-converting enzyme inhibitor and a calcium antagonist. ln (treatment/baseline) = ln (treatment) − ln (baseline). The change in urine protein excretion for each experimental group included in the regression analysis.

    In the regression analysis, ACE inhibitors caused significant reductions in urine protein excretion that were independent of mean arterial pressure changes and other explanatory variables (Table 4). An additional independent lowering of protein excretion occurred in patients with higher baseline mean arterial pressure. However, once blood pressure was taken into account, calcium antagonists, β-blockers, and other agents had no independent effect on urine protein excretion. A greater reduction in urine protein excretion occurred among persons with hyperfiltration. However, alterations in glomerular filtration rate did not otherwise correlate with the changes in proteinuria. In general, the regression results were similar when urine albumin excretion alone was examined separately. Indeed, of all the agents studied, only ACE inhibitors had an independent effect on urine albumin excretion alone.

    View this table:
    Table 4. Independent Correlates of Changes in Urine Protein Excretion*

    Analysis Restricted to Randomized, Controlled Trials

    We examined the effects of ACE inhibition in a separate analysis of the 11 randomized, controlled studies. For mean arterial pressure, the treatment effect was −3.05 mm Hg (95% CI, −4.64 to −1.46 mm Hg; P < 0.05). For glomerular filtration rate, the treatment effect was −2.85 mL/min (95% CI, −6.55 to 0.78 mL/min; P > 0.05), whereas for renal blood flow it was 5.0 mL/min (95% CI, −45.3 to 55.3 mL/min; P > 0.05). The ACE inhibitor treatment effect for urine albumin excretion (ln [treatment effect]) was −0.59 (95% CI, −0.35 to −0.82,P < 0.05). Thus, in this separate analysis of treatment effects in randomized, controlled trials, the effects of ACE inhibitors on glomerular filtration rate and renal blood flow were not significantly different from control values. In contrast, ACE inhibitors caused a significant reduction in urine albumin excretion.

    Previous SectionNext Section

    Discussion

    Although many different agents have been shown to be effective in reducing blood pressure in patients with diabetes, results have conflicted regarding the renal effects of specific antihypertensive agents. The differing results of these studies could be due to differences in the agents themselves, differences in patients, differences in the duration of treatment, or even differences in study design.

    We used a meta-regression analysis to examine why different studies produced different results. Rather than restricting the regression analysis to the 12 randomized, controlled trials, all but 1 of which examined only ACE inhibitors, we included in the regression analysis 100 studies that measured any changes in end points. This allowed us to compare the effects of other agents with those of ACE inhibitors and to explore correlations between study results and several important patient and study design characteristics.

    Our meta-analysis differed from the more traditional type in that we could not compare the measured change in each end point with the expected change in a random sample of patients drawn from the same population. Instead, we weighted the regression for the change in each end point with the inverse variance of that change to take into account random variation from study to study. Thus, relatively more weight was given to studies when the measured change was less variable. In addition, because the calculation of the variance took sample size into account, larger studies also received more weight than smaller studies.

    Most investigations found that ACE inhibitors reduced proteinuria, urine albumin excretion, or both [7, 22-59]. However, many studies found no change in proteinuria [6, 38, 40, 60-70], and one study found that urine protein excretion increased after treatment with an ACE inhibitor [71]. Whether the effects of ACE inhibitors on proteinuria in patients with diabetes are unique to this class of agent or represent a nonspecific effect of blood pressure reduction has been controversial. Studies that examined the effects of calcium antagonists, for example, found that protein excretion decreased [29, 37, 40, 43, 72], remained unchanged [24, 40, 52, 53, 67, 73], or increased [30, 72]. Studies of other agents and combinations also found that protein excretion decreased [4, 51, 74-81], or remained unchanged [81, 82].

    Our analysis showed that, among all antihypertensive agents, ACE inhibitors had a unique ability to decrease proteinuria independent of the reduction in proteinuria caused by changes in systemic blood pressure. The reductions in protein excretion by other agents were less impressive. Moreover, the regression analysis suggested that, unlike reductions from ACE inhibitors, the reductions in proteinuria from other agents could be entirely attributed to decreases in systemic blood pressure.

    The salutary effect of ACE inhibitors on proteinuria was seen whether or not the patient had hypertension, whether or not the patient had microalbuminuria or clinically apparent proteinuria, and whether the patient had type I or type II diabetes. The effects of structurally different ACE inhibitors, especially enalapril and captopril, the two most frequently studied agents, were similar. Interestingly, the reductions in proteinuria after therapy with ACE inhibitors were greater in patients with an increased glomerular filtration rate (hyperfiltration) at baseline. Surprisingly, the duration of therapy with ACE inhibitors did not affect the magnitude of change in proteinuria which remained similar after days, months, and even years of therapy. These data do not allow us to determine mechanisms whereby ACE inhibitors reduced proteinuria. However, we can speculate that early functional declines in proteinuria, perhaps from alterations in glomerular hemodynamic function, may have been subsequently replaced by declines attributable to a reduction in glomerular structural damage.

    As with proteinuria, studies examining the effect of specific antihypertensive therapy on renal function in patients with diabetes also produced conflicting results. The glomerular filtration rate after ACE inhibition for different durations was either unchanged [22, 26-29, 31,33-35, 37-45, 48-52, 57, 58, 60, 61-67, 69, 70, 84, 85] or decreased [6, 7, 22, 24, 32, 38, 53, 59]. Two studies reporting a decrease in the glomerular filtration rate also reported a decrease in the rate of functional decline [6, 7]. However, other studies found no change in the rate of decline in renal function after ACE inhibitor therapy [32, 49]. Studies with calcium antagonists found no change [29, 37, 40, 43, 53, 67, 72, 86] or a decrease in glomerular filtration rate [24, 52, 72]. At least one study using calcium antagonists found a decrease in the rate of decline in renal function [84]. Studies using other agents found either no change [51, 74, 81, 82, 87] or a decrease in renal function [4, 8, 75-78]. Three of these studies found a decrease in the rate of functional decline [8, 76, 78].

    The meta-regression analysis showed that ACE inhibitors had a unique, specific, beneficial effect on glomerular filtration rate in patients with diabetes. Indeed, the relative increase in renal function from ACE inhibition was independent of the salutary effects of changes in blood pressure. In contrast, other antihypertensive agents had no effect on glomerular filtration rate once the beneficial effects of mean arterial pressure reduction were taken into account. Thus, mean arterial pressure reduction from ACE inhibitor therapy caused a significantly greater improvement in glomerular filtration rate than did a comparable pressure reduction from other agents. Patients with hyperfiltration also had relatively less reduction in glomerular filtration rate, perhaps reflecting an earlier or more indolent stage of diabetic nephropathy.

    The effects of ACE inhibitors on renal function were not limited to patients with type I or type II diabetes, patients with hypertension, or patients with early or more advanced diabetic nephropathy. In addition, the magnitude of the effect of ACE inhibition was independent of the duration of therapy. Thus, the relative increase in glomerular filtration rate after ACE inhibition was not significantly different in patients treated for short or prolonged periods of time.

    Overall, the filtration fraction tended to decrease in patients receiving ACE inhibitor therapy. Whether this effect resulted from changes in hydraulic pressures that govern filtration or from a change in the filtration barrier is unclear. Studies of the glomerular filtration barrier in diabetic patients suggested that ACE inhibitors alter the permeability characteristics of the glomerular capillary [45]. Studies in animal models of type I diabetes found that ACE inhibition reduced glomerular capillary pressure [12, 13, 17, 19, 20]. However, in the obese Zucker rat model of type II diabetes, ACE inhibitor therapy reduced albuminuria and glomerular injury while causing little alteration in glomerular capillary pressure [18]. Nevertheless, in these experiments, a trend toward reduced glomerular pressure was observed, which may have combined with a decrease in glomerular size to reduce capillary wall tension and thereby to decrease glomerular injury [18]. The effects of ACE inhibitors on specific determinants of filtration in humans with diabetes are unknown.

    In general, the results of the separate meta-analysis of the randomized, controlled trials agreed with the broader results of the meta-regression. Thus, ACE inhibitors significantly reduced mean arterial pressure and urine albumin excretion and tended to increase glomerular filtration rate in the randomized, controlled trials. Few of the study design and quality features had much effect on the results of the meta-regression. Only random allocation affected changes in mean arterial pressure, and no other study features affected changes in glomerular filtration rate, urine protein excretion, or urine albumin excretion. It is possible that the large number of studies analyzed in the meta-regression negated effects of poor study design, with positive and negative biases that canceled each other out.

    The meta-regression allowed us to compare the results of virtually all published investigations on the subject. However, our findings should be interpreted with caution. By comparing changes in the end points between different experimental and control groups from different studies, there is a real possibility that confounding differences due to variability in patient study populations and in study design were not detected. An attempt to control for these differences was made by including covariates that described the patients studied and the study design features. However, some of the patient characteristics, (for example, age, duration of diabetes, stage of nephropathy) may not have accurately accounted for true differences between the groups because they reflect group averages rather than individual data. In addition, some potentially important characteristics were not always reported, and these missing data could have also biased the results. Moreover, other details (how often patients were seen in clinic, how diet was managed, and so forth) were not reported in most studies and were not included in the regression analysis. Any of these potentially important differences between the groups could have influenced the results. Thus, in a very real sense, the results of the meta-regression are only as good as the studies that were included.

    Because the regression analysis included only published studies, the results could also have been influenced by publication bias. If a sufficiently large number of studies were carried out but not published because the results differed from the existing literature, then the sample of studies used in the present meta-analysis may not have been representative of all published and unpublished studies. The estimation procedure used to calculate the variance of the change in each end point may have also introduced errors that could have decreased the power of the analysis to detect significant differences, especially for agents or patient populations that were not used in many studies. Thus, it is possible that the inclusion of additional studies with other agents or groups of patients could ultimately show effects not seen in the current analysis.

    Finally, our study did not address possible beneficial or adverse metabolic effects of treatment with different antihypertensive agents. The effect of agents on the kidney should not be the sole criterion used in selecting antihypertensive therapy for diabetic patients. The effect of agents on serum lipid levels and glucose metabolism, for example, should also be considered. However, our results suggest that ACE inhibitors have a unique and potentially beneficial effect on the kidney in patients with diabetes. If future investigations show that renal structure is also spared by the use of ACE inhibitors, a major advance will have been made in the treatment of hypertension and renal disease in patients with diabetes.

    Previous Section
     

    References

    1. 1.↵
      United States Renal Data System. USRDS 1990 Annual Data Report. Bethesda, Maryland: The National Institutes of Health, National Institute of Diabetes and Digestive Diseases; 1990.
    2. 2.↵
      Parving HH. Impact of blood pressure and antihypertensive treatment on incipient and overt nephropathy, retinopathy, and endothelial permeability in diabetes mellitus. Diabetes Care. 1991; 14: 260-9.
    3. 3.↵
      Norgaard K, Feldt-Rasmussen B, Borch-Johnsen K, Saelan H, Deckert T. Prevalence of hypertension in type 1 (insulin-dependent) diabetes mellitus. Diabetologia. 1990; 33:407-10.
    4. 4.↵
      Mogensen CE. Progression of nephropathy in long-term diabetics with proteinuria and effect of initial anti-hypertensive treatment. Scand J Clin Lab Invest. 1976; 36:383-8.
    5. 5.↵
      Hasslacher C, Ritz E, Terpstra J, Gallasch G, Kunowski G, Rall C. Natural history of nephropathy in type I diabetes. Relationship to metabolic control and blood pressure. Hypertension. 1985; 7(Suppl II):II74-8.
    6. 6.↵
      Bjorck S, Nyberg G, Mulec H, Granerus G, Herlitz H, Aurell M. Beneficial effects of angiotensin converting enzyme inhibition on renal function in patients with diabetic nephropathy. Br Med J (Clin Res Ed). 1986; 293:471-4.
    7. 7.↵
      Parving HH, Hommel E, Smidt UM. Protection of kidney function and decrease in albuminuria by captopril in insulin dependent diabetics with nephropathy. BMJ. 1988; 297:1086-91.
    8. 8.↵
      Mogensen CE. Long-term antihypertensive treatment inhibiting progression of diabetic nephropathy. Br Med J (Clin Res Ed). 1982; 285:685-8.
    9. 9.↵
      Cooper ME, Allen TJ, MacMillan PA, Clarke BE, Jerums G, Doyle AE. Enalapril retards glomerular basement membrane thickening and albuminuria in the diabetic rat. Diabetologia. 1989; 32:326-8.
    10. 10.↵
      Jackson B, Cubela R, Debrevi L, Whitty M, Johnston CI. Disparate effects of angiotensin converting enzyme inhibitor and calcium blocker treatment on the preservation of renal structure and function following subtotal nephrectomy or streptozotocin-induced diabetes in the rat. J Cardiovasc Pharmacol. 1987; 10(Suppl 10): S167-9.
    11. 11.↵
      Rabkin R, Petersen J, Kitaji J, Marck B, Murphy W, Muirhead EE. Effect of antihypertensive therapy on the kidney in spontaneously hypertensive rats with diabetes (Abstract). Kidney Int. 1984; 36:205.
    12. 12.↵
      Fujihara CK, Padilha RM, Santos MM, Zatz R. Role of glomerular hypertension, glomerular hypertrophy and lipid deposition in the genesis of glomerular sclerosis of experimental diabetes (Abstract). Kidney Int. 1990; 37:506.
    13. 13.↵
      Anderson S, Rennke HG, Garcia DL, Brenner BM. Short and long term effects of antihypertensive therapy in the diabetic rat. Kidney Int. 1989; 36:526-36.
    14. 14.↵
      Bank N, Klose R, Aynedjian HS, Nguyen D, Sablay LB. Evidence against increased glomerular pressure initiating diabetic nephropathy. Kidney Int. 1987; 31:898-905.
    15. 15.↵
      Yoshida K, Yasujima M, Kohzuki M, Sato M, Tsunoda K, Kudo K, et al. Calcium channel blocker and angiotensin converting enzyme inhibitor in normotensive diabetic rats and hypertensive diabetic rats (Abstract). In: Proceedings of the International Congress of Nephrology. Tokyo, Japan: International Society of Nephrology; 1990:160A.
    16. 16.↵
      Cooper ME, Allen TJ, O'Brien RC, Papazoglu D, Clarke BE, Jerums G, et al. Nephropathy in model combining genetic hypertension with experimental diabetes Enalapril versus hydralazine and metoprolol therapy. Diabetes. 1990; 39:1575-9.
    17. 17.↵
      Dunn BR, Zatz R, Rennke HG, Meyer TW, Anderson S, Brenner BM. Prevention of glomerular capillary hypertension in experimental diabetes mellitus obviates functional and structural glomerular injury. J Hypertens. 1986; 4(Suppl 5):S251-4.
    18. 18.↵
      Schmitz PG, O'Donnell MP, Kasiske BL, Katz SA, Keane WF. Renal injury in obese Zucker rats: Glomerular hemodynamic alterations and the effects of enalapril. Am J Physiol (In press).
    19. 19.↵
      Zatz R, Dunn BR, Meyer TW, Anderson S, Rennke HG, Brenner BM. Prevention of diabetic glomerulopathy by pharmacological amelioration of glomerular capillary hypertension. J Clin Invest. 1986; 77:1925-30.
    20. 20.↵
      Anderson S, Rennke HG, Zayas MA, Brenner BM. Chronic calcium channel blockade fails to lower glomerular pressure and albuminuria in diabetic rats (Abstract). Clin Res. 1991; 39:247A.
    21. 21.↵
      Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev. 1987; 9:1-30.
    22. 22.↵
      Hommel E, Parving HH, Mathiesen E, Edsberg B, Damkjaer Nielsen M, Giese J. Effect of captopril on kidney function in insulin-dependent diabetic patients with nephropathy. Br Med J (Clin Res Ed). 1986; 293:467-70.
    23. 23.↵
      Taguma Y, Kitamoto Y, Futaki G, Ueda H, Monma H, Ishizaki M, et al. Effect on captopril on heavy proteinuria in azotemic diabetics. N Engl J Med. 1985; 313:1617-20.
    24. 24.↵
      Holdaas H, Hartmann A, Lien MG, Nilsen L, Jervell J, Fauchald P, et al. Contrasting effects of lisinopril and nifedipine on albuminuria and tubular transport functions in insulin dependent diabetics with nephropathy. J Intern Med. 1991; 229:163-70.
    25. 25.↵
      Parving HH, Hommel E, Edsberg B, Mathiesen E, Damkjaer Nielsen M, Giese J. The effect of captopril on glomerular filtration rate and albuminuria in insulin-dependent diabetic patients with nephropathy (Abstract). Postgrad Med J. 1986; 62(Suppl 1):65.
    26. 26.↵
      Hollander E. Effect of antihypertensive (captopril) treatment on proteinuria in diabetic nephropathy. Ther Hung. 1988; 36:191-5.
    27. 27.↵
      Romero R, Sanmarti A, Salinas I, Teixido J, Foz M. Utilidad de los inhibidores de la enzima conversiva de la angiotensina en el trateamiento de la nefropatia diabetica (Usefulness of angiotensin converting enzyme inhibitors in the treatment of diabetic nephropathy.) Med Clin (Barc). 1988; 90:494-6.
    28. 28.↵
      Stornello M, Valvo EV, Vasques E, Leone S, Scapellato L. Systemic and renal effects of chronic angiotensin converting enzyme inhibition with captopril in hypertensive diabetic patients. J Hypertens. 1989; 7(Suppl 7):S65-7.
    29. 29.↵
      Baba T, Murabayashi S, Takebe K. Comparison of the renal effects of angiotensin converting enzyme inhibitor and calcium antagonist in hypertensive type 2 (non–insulin-dependent) diabetic patients with microalbuminuria: a randomized, controlled trial. Diabetologia. 1989; 32:40-4.
    30. 30.↵
      Insua A, Ribstein J, Mimran A. Comparative effect of captopril and nifedipine in normotensive patients with incipient diabetic nephropathy. Postgrad Med J. 1988; 64(Suppl 3):59-62.
    31. 31.↵
      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.
    32. 32.↵
      Parving HH, Hommel E, Dampkjaer Nielsen M, Giese J. Effect of captopril on blood pressure and kidney function in normotensive insulin dependent diabetics with nephropathy. BMJ. 1989; 299: 533-6.
    33. 33.↵
      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):S314-5.
    34. 34.↵
      Pedersen MM, Schmitz A, Pedersen EB, Danielsen H, Christiansen JS. Acute and long-term renal effects of angiotensin converting enzyme inhibition in normotensive, normoalbuminuric insulin-dependent diabetic patients. Diabetic Med. 1988; 5:562-9.
    35. 35.↵
      Stornello M, Valvo EV, Puglia N, Scapellato L. Angiotensin converting enzyme inhibition with a low dose of enalapril in normotensive diabetics with persistent proteinuria. J Hypertens. 1988; 6(Suppl 4):S464-6.
    36. 36.↵
      Bjorck S, Mulee H, Johnsen SA, Nyberg G, Aurell M. Contrasting effects of enalapril and metoprolol on proteinuria in diabetic nephropathy. Br Med J. 1990; 300:904-7.
    37. 37.↵
      Stornello M, Valvo EV, Scapellato L. Hemodynamic, renal, and humoral effects of the calcium entry blocker nicardipine and converting enzyme inhibitor captopril in hypertensive type II diabetic patients with nephropathy. J Cardiovasc Pharmacol. 1989; 14:851-5.
    38. 38.↵
      Abu-Romeh SH, Nawz MK, Ali JH, Al-Suhaili AR, Abu-Jayyab AK. Short-term effect of angiotensin-converting enzyme inhibitor enalapril in incipient diabetic nephropathy. Clin Nephrol. 1989; 31: 18-21.
    39. 39.↵
      Rudberg S, Aperia A, Freyschuss U, Persson B. Enalapril reduces microalbuminuria in young normotensive type 1 (insulin-dependent) diabetic patients irrespective of its hypotensive effect. Diabetologia. 1990; 33:470-6.
    40. 40.↵
      Melbourne Diabetic Nephropathy Study Group. Comparison between perindopril and nifedipine in hypertensive and normotensive diabetic patients with microalbuminuria.BMJ. 1991; 302:210-6.
    41. 41.↵
      Brichard SM, Santoni JP, Thomas JR, Van de Voorde K, Ketelslegers JM, Lambert AE. Long term reduction of microalbuminuria after 1 year of angiotensin converting enzyme inhibition by perindropil in hypertensive insulin-treated diabetic patients. Diabete Metab. 1989; 16:30-6.
    42. 42.↵
      Slomowitz LA, Bergamo R, Grosvenor M, Kopple JD. Enalapril reduces albumin excretion in diabetic patients with low levels of microalbuminuria. Am J Nephrol. 1990; 10:457-62.
    43. 43.↵
      Bakris GL. Effects of diltiazem or lisinopril on massive proteinuria associated with diabetes mellitus. Ann Intern Med. 1990; 112:707-8.
    44. 44.↵
      Cook J, Daneman D, Spino M, Sochett E, Periman K, Balfe JW. Angiotensin converting enzyme inhibitor therapy to decrease microalbuminuria in normotensive children with insulin-dependent diabetes mellitus. J Pediatr. 1990; 117:39-45.
    45. 45.↵
      Morelli E, Loon N, Meyer T, Peters W, Myers BD. Effects of converting-enzyme inhibition on barrier function in diabetic glomerulopathy. Diabetes. 1990; 39:76-82.
    46. 46.↵
      Casado S, Carrasco MA, Arrieta FJ, Herrera JL. Effects of captopril in diabetic patients with different degrees of blood pressure and proteinuria (Abstract). Postgrad Med J. 1988; 64(Suppl 3):85.
    47. 47.↵
      Kisch ES. Captopril and proteinuria in diabetes mellitus. Isr J Med Sci. 1987; 23:833-4.
    48. 48.↵
      Stornello M, Valvo EV, Scapellato L. Comparative effects of enalapril, atenolol and chlorthalidone on blood pressure and kidney function of diabetic patients affected by arterial hypertension and persistent proteinuria. Nephron. 1991; 58:52-7.
    49. 49.↵
      Reams GP, Bauer JH, Klachko D. Decrease in proteinuria with ACE inhibition does not predict attenuation of progression of renal disease (Abstract). Clin Res. 1991; 39:247A.
    50. 50.↵
      Stornello M, Valvo EV, Scapellato L. Angiotensin converting enzyme inhibition in normotensive type II diabetics with persistent mild proteinuria (Abstract). J Nephrol. 1989; 2(Suppl 1):24.
    51. 51.↵
      Stornello M, Valvo EV, Scapellato L. Persistent albuminuria in normotensive type 2 diabetics: Comparative effect of converting enzyme inhibition and β-blocker (Abstract). J Nephrol. 1989; 2(Suppl 1):24.
    52. 52.↵
      Ferder L, Daccordi HA, Martello M, Panzalis M. ACE inhibitors versus calcium antagonist in the treatment of diabetic hypertensive patients (Abstract). Am J Hypertens. 1991; 4:87A.
    53. 53.↵
      Martello M, Daccordi HA, Ferder LF, Inserra F, Panzalis MM. Enalapril versus nifedipine in diabetic patients with hypertension and proteinuria (Abstract). Diabetes. 1991; 40:506A.
    54. 54.↵
      Boulton AJ, Fernando DJ, Kumar S, Hutchison A, Gokal R. Angiotensin converting enzyme inhibition in diabetic nephropathy: research to practice (Abstract). Diabetes. 1991; 40:508A.
    55. 55.↵
      Arroyo JA, Sola R, Mallafre JM, et al. Effects of hypoproteic diet and captopril on the proteinuria hyperexcretion in type 1 (insulin-dependent) diabetic patients (Abstract). Diabetes. 1991; 40:441A.
    56. 56.↵
      Zanella MT, Salgado BJL, Kohlmann J, Ribeiro AB. Converting enzyme inhibition: A therapeutical option for diabetic hypertensive patients (Abstract). Hypertension. 1987; 9:543.
    57. 57.↵
      Mathiesen ER, Hommel E, Giese J, Parving HH. Efficacy of captopril in postponing nephropathy in normotensive insulin dependent diabetic patients with microalbuminuria. BMJ. 1991; 303:81-7.
    58. 58.↵
      Gonzalez-Sicilia de Llamas L, Garcia-Aberola A, Lafuente Lopez-Herrera M, Fuente Jimenez T, Fernandez Pardo J, Hernandez Cascales J. Effects of captopril on diabetic nephropathy in hypertensive women. Eur J Clin Pharmacol. 1991; 41:405-9.
    59. 59.↵
      Wiegmann TB, Herron KG, Chonko AM, MacDougall ML, Moore WV. Effect of angiotensin-converting enzyme inhibition on renal function and albuminuria in normotensive type I diabetic patients. Diabetes. 1992; 41:62-7.
    60. 60.↵
      Passa P, LeBlanc H, Marre M. Effects of enalapril in insulin-dependent diabetic subjects with mild to moderate uncomplicated hypertension. Diabetes Care. 1987; 10:200-4.
    61. 61.↵
      D'Angelo A, Sartori L, Gambrao G, Giannini S, Malvasi L, Benetollo P, et al. Captopril in the treatment of hypertension in type I and type II diabetic patients. Postgrad Med J. 1986; 62(Suppl 1): 69-72.
    62. 62.↵
      Drummond K, Levy-Marchal C, Laborde K, Kindermans C, Wright C, Dechaux M, et al. Enalapril does not alter renal function in normotensive, normoalbuminuric, hyperfiltering type 1 (insulin-dependent) diabetic children. Diabetologia. 1989; 32:255-60.
    63. 63.↵
      Elving LD, de Nobel E, van Lier HJ, Thien T. A comparison of the hypotensive effects of captopril and atenolol in the treatment of hypertension in diabetic patients. J Clin Pharmacol. 1989; 29:316-20.
    64. 64.↵
      Valvo E, Bedogna V, Casagrande P, Antiga L, Zamboni M, Bommartini F, et al. Captopril in patients with type II diabetes and renal insufficiency: Systemic and renal hemodynamic alterations. Am J Med. 1988; 85:344-8.
    65. 65.↵
      Rett K, Wicklmayr M, Tschollar W, Dietze G, Mehnert H. Role of angiotensin-converting enzyme inhibitors in early antihypertensive treatment in non-insulin dependent diabetes mellitus. Postgrad Med J. 1988; 64(Suppl 3):69-73.
    66. 66.↵
      Winocour PH, Waldek S, Anderson DC. Converting enzyme inhibition and kidney function in normotensive diabetic patients with persistent microalbuminuria (Letter). Br Med J. 1987; 295:391.
    67. 67.↵
      Corcoran JS, Perkins JE, Hoffbrand BI, Yudkin JS. Treating hypertension in non–insulin-dependent diabetes: A comparison of atenolol, nifedipine, and captopril combined with bendrofluazide. Diabetic Med. 1987; 4:164-8.
    68. 68.↵
      Mbanya JC, Thomas TH, Taylor R, Wilkinson R, Albeti KG. Effects of angiotensin converting enzyme inhibition on diabetic hypertension (Abstract). Diabetologia. 1987; 30:556A-7A.
    69. 69.↵
      Gall MA, Rossing P, Hommel E, Mathiesen ER, Gerdes U, Parving HH. Metabolic and haemodynamic effects of captopril, metoprolol and hydrochlorothiazide in hypertensive non–insulin-dependent diabetic (NIDDM) patients (Abstract). Diabetes. 1990; 39(Suppl 1): 124A.
    70. 70.↵
      Gambaro G, Morbiato F, Cicerello E, Del Turco M, Sartori L, D'Angelo A, et al. Captopril in the treatment of hypertension in type 1 and type 2 diabetic patients. J Hypertens. 1985; 3(Suppl 2):S149-51.
    71. 71.↵
      Hay U, Ludvik B, Gisinger C, Schernthaner G. Lack of influence of ACE-inhibition on macroproteinuria in diabetic nephropathy—a long-term trial over 6 months (Abstract). Postgrad Med J. 1988; 64(Suppl 3):88.
    72. 72.↵
      Demarie BK, Bakris GL. Effects of different calcium antagonists on proteinuria associated with diabetes mellitus. Ann Intern Med. 1990; 113:987-8.
    73. 73.↵
      Cruickshank JK, Anderson NM, Wadsworth J, Young SM, Jepson E. Treating hypertension in black compared with white non-insulin dependent diabetics: a double blind trial of verapamil and metoprolol. BMJ. 1988; 297:1155-9.
    74. 74.↵
      Christensen CK, Mogensen CE. Effect of antihypertensive treatment on progression of incipient diabetic nephropathy. Hypertension. 1985; 7(Suppl II):II109-13.
    75. 75.↵
      Parving HH, Andersen AR, Smidt UM, Friisberg B, Svendsen PA. Reduced albuminuria during early and aggressive antihypertensive treatment of insulin-dependent diabetic patients with diabetic nephropathy. Diabetes Care. 1981; 4:459-63.
    76. 76.↵
      Parving HH, Andersen AR, Hommel E, Smidt U. Effects of long-term antihypertensive treatment on kidney function in diabetic nephropathy. Hypertension. 1985; 7(Suppl II):II114-17.
    77. 77.↵
      Parving HH, Smidt UM. Hypotensive therapy reduces microvascular albumin leakage in insulin-dependent diabetic patients with nephropathy. Diabetic Med. 1986; 3:312-5.
    78. 78.↵
      Parving HH, Andersen AR, Smidt UM, Hommel E, Mathiesen ER, Svendsen PA. Effect of antihypertensive treatment on kidney function in diabetic nephropathy. Br Med J (Clin Res Ed). 1987; 294: 1443-7.
    79. 79.↵
      Friedman PJ, Dunn PJ, Jury DR. Metoprolol and albumin excretion in diabetes (Letter). Lancet. 1986; 2:1042-3.
    80. 80.↵
      Janka HU, Weitz T, Blumner E, van Michel A, Mehnert H. Hypertension and microalbuminuria in diabetic patients taking indapamide. J Hypertens. 1989; 7(Suppl 6):S316-7.
    81. 81.↵
      Hommel E, Mathiesen E, Edsberg B, Bahnsen M, Parving HH. Acute reduction of arterial blood pressure reduces urinary albumin excretion in type I (insulin-dependent) diabetic patients with incipient nephropathy. Diabetologia. 1986; 29:211-5.
    82. 82.↵
      Fogari R, Lazzari P, Zoppi A, Tettamanti F, Malamani G, Boari L. The effects of celiprolol in the short-term treatment of hypertensive patients with type 2 diabetes. Current Therapeutic Research. 1990; 47:879-88.
    83. 83.↵
      Kolendorf K, Bonnevie-Nielson V, Broch-Moller B. A trial of metoprolol in hypertensive insulin-dependent diabetic patients. Acta Med Scand. 1982; 211:175-8.
    84. 84.↵
      Ueda Y, Aoi W, Yamachika S, Shikaya T. Beneficial effects of angiotensin-converting enzyme inhibitor on renal function and glucose homeostasis in diabetics with hypertension. Nephron. 1990; 55(Suppl 1):85-9.
    85. 85.↵
      Escobar-Jiminez F, Fernandez Soto ML, Lobon JA, Aguilar M, Campos-Pastor MM, Garcia del Rio C. Effect of captopril at two different dosage regimens in hypertensive non-insulin dependent patients. Postgrad Med J. 1988; 64(Suppl 3):65-8.
    86. 86.↵
      Zander E, Conde N, Heinke P, Zander G, Strese J, Gums G, et al. The effect of antihypertensive treatment on kidney function in insulin-dependent (type I) diabetics with renal failure. Exp Clin Endocrinol. 1990; 95:70-6.
    87. 87.↵
      Levy P. Effects of prazosin on blood pressure and diabetic control in patients with type II diabetes mellitus and mild essential hypertension. Am J Med. 1989; 86(Suppl 1B):59-62.
    88. 88.↵
      Matthews DM, Wathen CG, Bell D, Collier A, Muir AL, Clarke BF. The effect of captopril on blood pressure and glucose tolerance in hypertensive non-insulin dependent diabetes. Postgrad Med J. 1986; 62(Suppl 1):73-5.
    89. 89.↵
      Dominguez JR, de la Calle H, Hurtado A, Robles RG, Sancho-Rof J. Effect of converting enzyme inhibitors in hypertensive patients with non–insulin-dependent diabetes mellitus. Postgrad Med J. 1986; 62(Suppl 1):66-8.
    90. 90.↵
      Aranda P, Aranda JM, Carmona JR, Nunez D. Captopril as first step therapy in elderly hypertensive diabetic patients (Abstract). Postgrad Med J. 1988; 64(Suppl 3):87.
    91. 91.↵
      Riobo P, Garcia-Robles R, Estopinan V, Hurtado A, Sancho-Rof J. Effect of long-term inhibition of converting enzyme on carbohydrate metabolism in non-insulin dependent diabetic patients with essential hypertension (Abstract). Postgrad Med J. 1988; 64(Suppl 3):63-4.
    92. 92.↵
      Paolisso G, Gambardella A, Verza M, Varricchio M. Quinapril reduces arterial blood pressure and improves metabolic control in elderly hypertensive diabetic patients. Diabete Metab. 1990; 16: 264-5.
    93. 93.↵
      Prince MJ, Stuart CA, Padia M, Bandi Z, Holland OB. Metabolic effects of hydrochlorothiazide and enalapril during treatment of the hypertensive diabetic patient. Arch Intern Med. 1988; 148:2363-8.
    94. 94.↵
      Eto M, Watanabe K, Otake Y, Morikawa A, Takebe T, Ishii K. Effects of captopril on blood pressure, metabolic control, plasma lipids and proteinuria in elderly hypertensive diabetics. Current Therapeutic Research. 1988; 43:427-34.
    95. 95.↵
      Bilo HJ, Westerman RF, Nicolaas-Merkus AM, Donker AJ. Effects of enalapril with and without hydrochlorothiazide in hypertensive patients with non–insulin-dependent diabetes mellitus. Diabetes Res. 1988; 9:21-5.
    96. 96.↵
      Matthews DM, Wathen CG, Bell D, Collier A, Roulston JE, Clarke BF, et al. The use of captopril and captopril plus frusemide as antihypertensive agents in non-insulin dependent diabetes. J Hum Hypertens. 1987; 1:19-23.
    97. 97.↵
      Whitcroft IA, Thomas JM, Rawsthorne A, Wilkinson N, Thompson H. Effects of α and β adrenoceptor blocking drugs and ACE inhibitors on long term glucose and lipid control in hypertensive non-insulin dependent diabetics. Horm Metab Res. 1990; 22(Suppl): 42-6.
    98. 98.↵
      Lanza G, Barbera R, Fontana S. Trattamento dell'ipertensione nel diabetico con captopril. (Treatment of hypertension in the diabetic with captopril). Minerva Med. 1985; 76:183-4.
    99. 99.↵
      Chellingsworth MC, Kendall MJ, Wright AD, Singh BM, Pasi J. The effects of verapamil, diltiazem, nifedipine and propranalol on metabolic control in hypertensives with non-insulin dependent diabetes mellitus. J Hum Hypertens. 1989; 3:35-9.
    100. 100.↵
      Andren L, Hoglund P, Dotevall A, Eggertsen R, Svensson A, Olson SO, et al. Diltiazem in hypertensive patients with type II diabetes mellitus. Am J Cardiol. 1988; 62:114G-20G.
    101. 101.↵
      Capewell S, Collier A, Matthews D, Hajducka C, Collier R, Clarke BF, et al. A trial of the calcium antagonist felodipine in hypertensive type 2 diabetic patients. Diabetic Med. 1989; 6:809-12.
    102. 102.↵
      Bhandarkar SD, Vernekar KS. Clonidine as an antihypertensive drug in diabetics. J Postgrad Med. 1978; 24:182-5.
    103. 103.↵
      Harrower AD, McFarlane G, Donnelly T, Gray CE. Effect of indapamide on blood pressure and glucose tolerance in non–insulin-dependent diabetes. Hypertension. 1985; 7(Suppl II):II161-3.
    104. 104.↵
      Chazan BI, Old JM, Venkataraman G, Snape J, Cameron EG. A β blocker-thiazide combination in the treatment of hypertension in diabetics. Br J Clin Pract. 1991; 393-8.
    105. 105.↵
      Coves MJ, Gomis R, Casamitjana R, Lienas V, Vilaradell E. Antihypertensive therapy with guanfacine induces elevated plasma growth hormone levels in diabetic patients. J Med. 1989; 20:291-6.
    106. 106.↵
      Benfield GF, Hunter KR. Oxprenolol, methyldopa and lipids in diabetes mellitus. Br J Clin Pharmacol. 1982; 13:219-22.
    107. 107.↵
      Groop L, Totterman KJ, Harno K, Gordin A. Influence of β-blocking drugs on glucose metabolism in patients with non-insulin dependent diabetes mellitus. Acta Med Scand. 1982; 211:7-12.
    108. 108.↵
      Guthrie GP Jr, Miller RE, Kotchen TA, Koenig SH. Clonidine in patients with diabetes and mild hypertension. Clin Pharmacol Ther. 1983; 34:713-7.
    109. 109.↵
      Roux P, Courtois H. Blood sugar regulation during treatment with indapamide in hypertensive diabetics. Postgrad Med J. 1981; 58(Suppl 2):70-2.
    110. 110.↵
      Raggi U, Palumbo P, Moro B, Bevilacqua M, Norbiato G. Indapamide in the treatment of hypertension in non–insulin-dependent diabetes. Isr J Med Sci. 1985; 7(Suppl II):II157-60.
    111. 111.↵
      Micossi P, Pollavini G, Raggi U, Librenti MC, Garimberti B, Beggi P. Effects of metoprolol and propranolol on glucose tolerance and insulin secretion in diabetes mellitus. Horm Metab Res. 1984; 16: 59-63.
    112. 112.↵
      Weber MA, Drayer JI, Deitch MW. Hypertension in patients with diabetes mellitus: treatment with a centrally acting agent. J Cardiovasc Pharmacol. 1984; 6:S823-9.
    113. 113.↵
      Castrignano R, D'Angelo A, Pati T, Al Awady M, Tronca R, Crepaldi G. A single-blind study of doxazosin in the treatment of mild-to-moderate essential hypertensive patients with concomitant noninsulin-dependent diabetes mellitus. Am Heart J. 1988; 116: 1778-84.
    114. 114.↵
      Maruyama H, Saruta T, Itoh H, Koyama K, Kido K, Itoh K, et al. Effect of α-adrenergic blockade on blood pressure, glucose, and lipid metabolism in hypertensive patients with non–insulin-dependent diabetes mellitus. Am Heart J. 1991; 121:1302-6.
    115. 115.↵
      Hauger-Klevene JH, Scornavacchi JC. Improvement of glucose tolerance in hypertensive diabetic patients treated with guanfacine one year. Eur J Clin Pharmacol. 1985; 29:391-3.
    116. 116.↵
      Dvorak I, Podrouzkova B, Kocourkova H, Spac J, Hemcova H, Blaha M. Guanfacine in the treatment of hypertensives with a pronounced response to exercise and diabetes mellitus. Cor Vasa. 1987; 29(Suppl 1):30-6.
    117. 117.↵
      Wallin JD, Krane NK, Bergman S, Morgan M. The use of transcutaneous clonidine hydrochloride in the patient with diabetes mellitus and mild hypertension. J Clin Hypertens. 1985; 4:315-21.
    118. 118.↵
      Wright AD, Barber SG, Kendall MJ, Poole PH. Beta-adrenoceptor-blocking drugs and blood sugar control in diabetes mellitus. Br Med J. 1979; 1:159-61.
    119. 119.↵
      DerSimonian R, Laird NM. Evaluating the effect of coaching on SAT scores: a meta-analysis. Harvard Education Review. 1983; 53: 1-15.
    120. 120.↵
      Cappuccio FP, Siani A, Strazzullo P. Oral calcium supplementation and blood pressure: an overview of randomized, controlled trials. J Hypertens. 1989; 7:941-6.
    121. 121.↵
      Norusis MJ. Statistical Package for the Social Sciences (SPS/PC+ V2.0): Base Manual. Chicago: SPS Inc.; 1988.
    « Previous | Next Article »Table of Contents

    This Article

    1. Ann Intern Med January 15, 1993 vol. 118 no. 2 129-138
    1. AbstractFREE
    2. Figures
    3. » Full Text
    4. Full Text (PDF)

    Rapid Responses

    1. Submit a response
    2. No responses published

    Navigate This Article

    Current Issue

    1. November 17, 2009, 151 (10)
    1. Alert me to current issues