Comparative Effects of Irbesartan on Ambulatory and Office Blood Pressure: A substudy of ambulatory blood pressure from the Irbesartan in Patients with Type 2 Diabetes and Microalbuminuria Study

The IRMA2 study protocol has been described in detail previously (3). In brief, 590 type 2 diabetic patients with microalbuminuria and hypertension were enrolled in a multinational, randomized, double-blind, placebo-controlled study of irbesartan (150 or 300 mg daily) and were followed for 2 years. The primary outcome was the time to onset of diabetic nephropathy, defined as persistent albuminuria in overnight specimens with a UAE rate >200 μg/min and ≥30% higher than the baseline level. Target trough blood pressure was <135/85 mmHg for all three groups at 3 months after randomization. Additional antihypertensive drugs used included diuretics, β-blockers, calcium-channel blockers (except dihydropyridines), and α-blockers. These agents were added when the target blood pressure was not reached after 3 months on the study drug.

In the present substudy, we included 50 type 2 diabetic patients with persistent microalbuminuria and hypertension who were recruited from the Steno Diabetes Center for enrollment in the IRMA2 study. OBP and ABP were measured after a single-blind, 3-week run-in screening period during which all antihypertensive treatment was discontinued and replaced by placebo. OBP and ABP were measured again 2 years after randomization.

Trough OBP (phase I/V) was measured using an appropriate cuff with a sphygmomanometer in the sitting position after at least 10 min of rest. Two readings were taken 2 min apart and read to the nearest 2 mmHg; the average was used for calculation.

A 24-h ABP was measured with the Takeda TM2420 device (version 6 and 7; A&D Medical, Tokyo, Japan) (13). Cuff size 20–31 was used in patients with an upper arm circumference <32 cm and cuff size 28–36 was used in patients with an upper arm circumference >32 cm. Once yearly, the devices were tested on the bench by the local A&D agent. All devices had a variation of <±3 mmHg for systolic and diastolic blood pressure compared with a mercury sphygmomanometer before and after the study. The device was set up to measure blood pressure every 15 min during the day (7:00 a.m. to 11:00 p.m.) and every 30 min during the night (11:00 p.m. to 7:00 a.m.). Patients were instructed not to restrict their daily activities during the monitoring period, except during an ongoing measurement where patients were instructed not to move their arm. Before the start of the monitoring period, the automatic readings were cross-checked against manually measured blood pressure by auscultation to ascertain correct measurements of the device. Values were averaged for each hour before calculating the mean 24-h, day, and night ABP. Mean ABP was calculated as follows: diastolic blood pressure + [(systolic blood pressure - diastolic blood pressure)/3]. Any 24-h ABP measurement was accepted if at least 50% of the programmed pressures were measured successfully for each hour during the whole 24-h monitoring interval. To eliminate bias, no manual editing of data was performed. Previous recordings of 24-h ABP during ordinary life conditions on two occasions 2–4 months apart in 63 diabetic patients revealed coefficients of variation for systolic/diastolic blood pressure as 10/8% (24-h), 10/7% (daytime), and 13/13% (nighttime) (14).

The smoothness index was obtained by calculating the average blood pressure values for each hour of the 24-h monitoring period, both before and during treatment. From these, all hourly changes in blood pressure induced by treatment were obtained, and the average of these hourly values was divided by its standard deviation (15).

Results are presented as means ± SD, unless otherwise stated. Albuminuria was log transformed before analysis because of its positively skewed distribution. A one-way ANOVA was used to test for differences among treatment groups. Student’s t test for paired samples was used to determine changes in blood pressure and albuminuria from baseline to end of follow-up within each group. Linear regression analysis was performed with the relative change in the UAE rate as the dependent variable and the smoothness index, baseline levels, and absolute reductions after 2 years of OBP, 24-h ABP, day ABP, and night ABP as independent variables. Data were analyzed using the SPSS 10.0 (SPSS, Chicago, IL)

Of the 50 patients originally included in the IRMA2 study at the Steno Diabetes Center, 43 patients were included in the present study. During the study, one patient withdrew because of the side effects of the study medication and six (three, two, and one from the placebo, irbesartan 150 mg [Irb150], and irbesartan 300 mg [Irb300] group, respectively) were excluded from the analysis as they did not have 24-h ABP with at least 50% of the programmed pressures measured successfully for each hour during the whole 24-h monitoring interval at both the baseline visit and end of the study.

All 43 patients included in the study were Caucasians, and the baseline characteristics were comparable in the three groups (Table 1).

At baseline, the present subpopulation compared well with the remaining patients in the overall IRMA2 trial except for systolic OBP, which was slightly higher among patients in the present substudy. The substudy also had an increased prevalence of smokers (Table 1).

OBP at baseline was 157 ± 15/89 ± 7, 156 ± 15/91 ± 11, and 159 ± 16/90 ± 9 mmHg in the placebo, Irb150, and Irb300 groups (NS), respectively. The 24-h ABP at baseline was 148 ± 13/83 ± 11, 148 ± 16/82 ± 7, and 147 ± 16/81 ± 10 mmHg (NS), respectively (Table 2, Fig. 1). At baseline, diastolic night ABP was lower in the Irb300 group as compared to the other two groups (P = 0.04), but there was no statistically significant difference among the three groups in 24-h, day, or systolic night ABP.

We found that 2 years after randomization, OBP was significantly reduced in all three groups, with no statistically significant differences among groups (Table 2, Fig. 1). Reductions in 24-h, day, and night ABP were comparable in all three groups, but were generally less pronounced, especially at night, than reductions in OBP (Table 2).

At baseline, the night-to-day ratio of mean arterial pressure (MAP) averaged 0.92 ± 0.08, 0.91 ± 0.07, and 0.87 ± 0.08 in the placebo, Irb150, and Irb300 groups, respectively (NS). No statistically significant changes were seen 2 years after randomization, when the night-to-day ratio of MAP was 0.90 ± 0.06, 0.90 ± 0.10, and 0.92 ± 0.10, respectively (NS).

The mean hourly reductions in systolic/diastolic blood pressure during the 24-h ABP monitoring period were 12 ± 25/10 ± 15, 6 ± 22/7 ± 14, and 9 ± 21/8 ± 14 mmHg in the placebo, Irb150, Irb300 groups, respectively (NS). There was no statistical significant difference in the systolic/diastolic smoothness index among the placebo (0.57± 0.83), Irb150 (0.51± 0.67), and Irb300 groups (0.62 ± 0.75). There was no significant difference in antihypertensive or glucose- or lipid-lowering medication among the three treatment groups at the end of the study (Table 3).

At baseline, the UAE rate was (geometric mean with 95% CI) 43 (32–57), 46 (30–70), and 59 (42–85) μg/min in the placebo, Irb150, and Irb300 groups, respectively (NS). Then 2 years after randomization, the mean reduction in UAE was 0% (−86 to 42) % in the placebo group, 38% (−14 to 66) in the Irb150 group, and 73% (59 to 82) in the Irb300 group (overall, P < 0.01).

Linear regression analysis revealed a significant correlation between the systolic and diastolic smoothness index and relative reduction of albuminuria from baseline to end of follow-up (an increase in diastolic smoothness index of 1 equaled a 15% reduction in albuminuria [P = 0.01], and an increase in systolic smoothness index of 1 equaled an 8% reduction in albuminuria [P = 0.05]). There was no significant correlation between relative change in albuminuria and baseline levels or absolute reductions after 2 years of OBP, 24-h ABP, day ABP, and night ABP.

In the present IRMA2 substudy of hypertensive type 2 diabetic patients with persistent microalbuminuria, we found no significant differences in blood pressure reductions, as assessed by trough OBP, for 24-h, day, and night ABP, among treatment groups 2 years after randomization to placebo or irbesartan 150 or 300 mg daily. Despite comparable blood pressure reductions, the UAE rate was significantly lower at follow-up in patients treated with irbesartan, particularly those given 300 mg daily.

Among the 50 patients included at the Steno Diabetes Center for the IRMA2 study, 6 patients were excluded from the present substudy as they did not have successful 24-h ABP at both the baseline visit and follow-up. The missing information regarding 24-h ABP in these patients may have affected the results of our study if the patients not included differed from those included in the study. However, this did not seem to be the case, as trough OBP at baseline in the six patients excluded from the study averaged 154/87 mmHg and was reduced to 145/83 mmHg at the end of follow-up, which corresponded well to the OBP values seen in the patients included in the study.

The 24-h blood pressure measurements were unfortunately not available from other centers participating in the IRMA2 trial, which would have increased the power and generalizability of the present findings. However, the present subpopulation compared well with the overall IRMA2 study population with respect to diastolic blood pressure and UAE rate at baseline, race, age, sex, BMI, HbA_1c, serum creatinine, and cholesterol. Systolic blood pressure was slightly higher at baseline in the present substudy than in the overall study population, but the reductions in OBP were similar to those seen in the overall IRMA2 study, where OBP was reduced from 150/90 mmHg in all three groups to 143/83 mmHg in the placebo and Irb150 groups and to 141/83 mmHg in the Irb300 group during the study.

At baseline, risk factors for elevated blood pressure and progression of microalbuminuria were balanced in the three groups in the present study, including age, sex, BMI, smoking, and UAE rate. The three treatment groups had comparable OBP and ABP at randomization, except for diastolic night blood pressure, which probably by chance was significantly lower in the Irb300 group as compared to the placebo and Irb150 groups. However, diastolic night ABP at the end of the study was similar in the three groups.

At follow-up, we found no indications of a greater diurnal blood pressure reduction in patients receiving irbesartan. Blood pressure reductions during nighttime were also comparable among groups, and differences in nocturnal blood pressure could therefore not explain the greater reduction in overnight UAE rate in patients receiving irbesartan.

In the present study, reductions in ABP were generally less pronounced, particularly during nighttime, as compared to reductions in OBP. The lower reduction of ABP versus OBP is a common finding in hypertensive trials, where patients are selected on the basis of an elevated OBP (8). At follow-up, regression to the mean will make patients have a lower OBP and will lead to an overestimation of the effect of the antihypertensive intervention. Because patients are not selected by an elevated ABP at baseline, the reduction in ABP during the study will not be affected by regression to the mean, and therefore a relatively greater reduction is seen in OBP than ABP. Contributing to the overestimation of the reduction in OBP after antihypertensive treatment is the ∼10% of type 2 diabetic patients with microalbuminuria who have white coat hypertension at baseline (16), which to some extent may attenuate over a 2-year follow-up period and lead to a “false” reduction in OBP that is not seen by ABP measurements.

The lower reduction in ABP than OBP in the present study would from a pathophysiological point of view imply that the effect of blood pressure lowering by irbesartan may be overestimated and that nonhemodynamic effects may play a greater role in preventing the development of diabetic nephropathy in the overall IRMA2 study.

In contrast to our findings, the HOPE substudy demonstrated a greater reduction in ABP as compared to OBP after antihypertensive intervention. In the HOPE substudy, the 24-h APB was significantly reduced in patients receiving ramipril as compared to placebo, mainly because of a pronounced lowering of blood pressure during nighttime, whereas OBP and day ABP were not significantly reduced by ramipril. The discrepancy between the results of our study and those of the HOPE substudy may well be explained by differences in design: type of medication, timing of drug intake, patients included (hypertensive versus normotensive), blood pressure treatment, and goals. In particular, a major difference between the HOPE and IRMA2 trials may relate to HOPE being an add-on treatment study with no specified treatment goals for blood pressure whereas the IRMA2 trial involved titration to a target blood pressure of 135/85 mmHg and allowed other antihypertensive drugs to be added on. The addition of other antihypertensive agents also explains the blood pressure reduction demonstrated in the placebo group in the IRMA2 study.

In conclusion, reductions in OBP and 24-h blood pressure patterns were comparable among groups in patients randomized to placebo or irbesartan 150 or 300 mg daily in this IRMA2 substudy. Despite similar blood pressure levels, the UAE rate was reduced significantly only in patients treated with irbesartan. The renoprotective effects of irbesartan demonstrated in the overall IRMA2 trial could not be attributed to effects on blood-pressure-lowering patterns over a 24-h period in the present IRMA2 substudy.

We gratefully appreciate the work of the IRMA2 study group, which made this substudy possible.