Background

This study examined the effects of stent placement for renal artery stenosis on the function of treated and contralateral kidneys.

Methods

Eighteen patients who underwent stent placement for unilateral renal artery stenosis presenting with hypertension and/or renal failure were studied before angiography and stent placement and at their one-year follow-up. Renal vein blood samples were taken at both sides, at each side simultaneously with a sample from the aorta, to measure the plasma renin concentration and the concentrations of ¹³¹I-hippuran and ¹²⁵I-thalamate during constant systemic infusion of these radiochemicals. This allowed an assessment of the single-kidney contributions to the total renin secretion, effective renal plasma flow (¹³¹I-hippuran clearance) and glomerular filtration rate (¹²⁵I-thalamate clearance).

Results

At the one-year follow-up, the vein-to-artery renin ratio at the treated side had decreased to normal, from 1.65 0.131 to 1.23 0.076 (mean SEM; P = 0.011), indicating an improved renal blood flow. Contralaterally it rose from 1.09 0.042 to 1.17 0.029 (P = 0.055) at follow-up. The extraction ratio of ¹³¹I-hippuran improved at the treated side (0.48 0.049 to 0.62 0.034; P = 0.003) and contralaterally (0.67 0.033 to 0.73 0.026; P = 0.043). The extraction ratio of ¹²⁵I-thalamate, which equals filtration fraction, improved at both sides (0.12 0.014 to 0.17 0.012 at the treated side, P = 0.001; 0.18 0.013 to 0.22 0.011 contralaterally, P = 0.002). Two-kidney effective renal blood flow and glomerular filtration rate remained unchanged.

Conclusion

Renal artery stenting was capable of causing improvement of glomerular filtration rate of the treated kidney, although the overall glomerular filtration rate did not change.

METHODS

Patients

The study group comprised 18 patients (12 males, 6 females; aged 56 10.7 years; mean SD) who underwent stent placement for angiographically-proven symptomatic unilateral atherosclerotic renal artery stenosis of 50% luminal diameter reduction, and were studied by means of renal vein blood sampling. These patients were selected from a group of 40 consecutive patients undergoing stent placement who were referred to our hospital in the years 1997 and 1998. The length of the treated kidney, as determined by ultrasound, was>8 cm in all patients (10 0.3 cm; mean SEM). Renal scintigraphy with mercaptoacetythiglycine (MAG₃) and other single-kidney renal function studies were performed within one month before stent placement. These tests were repeated one year after the procedure, prior to the angiographic follow-up examination. Medication regimen and blood pressure were recorded before the intervention and at the follow-up examination. The study protocol was approved by the Local Committee of Human Research and written informed consent was obtained from all patients.

Renal scintigraphy

Scintigraphy was performed with the patient in the supine position and using the posterior view. Patients received 25 or 50 mg of captopril orally at one hour before the examination. After intravenous administration of 74 MBq ^99mTc-MAG₃, data were collected in 10-second frames during a 20-minute period. The number of total counts accumulated in the kidney during the first 60 seconds was quantified and taken as a measure of the renal plasma flow to the affected and contralateral kidneys. This method is based on the assumption that the early part of the time-activity curve is determined solely by the renal plasma flow. To avoid mentioning absolute data on renal blood flow, the ratio between the counts of the affected and the contralateral kidney was calculated. To avoid interference of the oral captopril with subsequent tests used in this study to evaluate renal function, these tests were performed several days apart.

Renal vein blood sampling

Concentrations of active renin and ¹³¹I-hippuran and ¹²⁵I-thalamate in the left and right renal veins were measured and assayed as described by Wenting et al³. The method involves constant infusion of ¹³¹I-hippuran and ¹²⁵I-thalamate through an antecubital vein. In order to test whether the steady state was reached, blood samples were taken 60, 90 and 105 minutes after starting infusion. The steady state was reached whenever the plasma concentrations of two blood samples were the same. After reaching the steady state, a renal venous blood sample was taken at each side simultaneously with a sample from the abdominal aorta. Samples from these sites were also used for renin measurements. Additional blood samples were taken at 15-minute intervals from a peripheral vein to estimate the total (two-kidney) clearances of hippuran and thalamate. All blood samples were immediately centrifuged and measurements were made in plasma. Determination of blood oxygen saturation served to assure that blood samples were taken from the renal veins. Single-kidney extraction ratios of ¹³¹I-hippuran and ¹²⁵I-thalamate were defined as (A - V)/A * 100%, where A is the activity in the abdominal aorta and V is the activity in renal vein. The total clearances of hippuran and thalamate were taken as a measure of total effective renal plasma flow (ERPF) and glomerular filtration rate (GFR), respectively. The concentration of active renin in plasma was measured by radioimmunoassay⁴. The normal value of the vein-to-artery ratio of renin is 1.24. In case of renal artery stenosis, the vein-to-artery ratio at the treated side is considered elevated with values>1.48. Contralateral suppression of renin secretion is defined as a vein-to-artery renin ratio <1.13⁵.

Stent placement

Pre-intervention digital subtraction angiograms were obtained using aortic-flush injections. The stenosis was then crossed with a 5 F selective catheter and the lesion was predilated with an angioplasty balloon. A Palmaz stent (Johnson & Johnson Interventional Systems, New York, NY, USA) was placed, and the procedure was considered technically successful when both intravascular ultrasound and angiography showed complete stent-vessel wall apposition, complete lesion-covering, and <20% residual diameter stenosis⁶. If necessary, additional stent dilation was performed. During the procedure the patients received 5000 IU of heparin, and heparin infusion was continued for 48 hours (20,000 to 30,000 IU per day). Oral acetylsalicylic acid (100 mg daily) was started at the day of the procedure and continued during the entire follow-up period.

Calculations

Two-kidney ERPF was calculated by dividing the infusion rates of ¹³¹I-hippuran by the peripheral venous plasma concentration of ¹³¹I-hippuran. Similarly, two-kidney GFR was calculated by dividing the infusion rate of ¹²⁵I-thalamate by the peripheral venous plasma concentration of ¹²⁵I-thalamate. In addition, single-kidney RPF, ERPF and GFR were calculated by using single-kidney data derived from MAG₃ scintigraphy, as explained in the Appendix.

Statistical analysis

Because a normal distribution could not be assumed, the effect of intervention on single-kidney function parameters was analyzed using the non-parametric Wilcoxon signed ranks test at P < 0.05. Analysis was repeated for the subgroup of patients (N = 9) who used the same medication both pre-stenting and after one year, to test whether differences in medication had influenced the results. We also examined whether there was any difference in clinical parameters between the 18 patients studied and the excluded patients. Finally, using regression analysis we assessed whether there was any relationship between the measured functional parameters and the clinical data on blood pressure and renal function.

RESULTS

Of the total group of 40 patients who underwent renal artery stent placement, 18 had complete recordings of single-kidney functions before stent placement and at follow-up. Of the 22 patients who were excluded, 4 had become dialysis dependent during the follow-up period, 4 had bilateral renal artery stenosis, and 4 refused follow-up investigation. In the remaining 10 patients the clinician did not adhere to the protocol. Stent placement was angiographically successful in all patients. Angiography at follow-up (12 months in 16 patients and 6 months in 2 patients) showed restenosis 50% (compared to the distal main renal artery diameter) in 4 patients. (Results of the patients with restenosis were also included in the analysis.) An overview of data of each patient individually is given in Tables 1 and 2. Systolic blood pressure decreased from 170 5.5 mm Hg (mean SEM) before stent placement to 146 4.4 mm Hg at follow-up. Diastolic blood pressure decreased from 95 2.4 mm Hg to 82 1.6 mm Hg. The mean amount of antihypertensive drugs decreased from 2.9 0.28 DDDs before stent placement to 2.1 .40 DDDs at follow-up. Two patients were cured from hypertension (normotensive without medication).

Table 1 - Overview of clinical data of each patient individually.

Full table

Table 2 - Overview of single-kidney function data of each patient.

Full table

At follow-up, the renal vein-to-artery renin ratio at the treated side had decreased from 1.65 0.131 to normal, 1.23 0.076 (mean SEM; P = 0.011; Table 3). Contralaterally, the vein-to-artery renin ratio was 1.09 0.042 before stenting and 1.17 0.029 (P = 0.055) at follow-up. Peripheral renin values decreased significantly from 251 114.0 ng angiotensin I/(mL * h) before stenting to 94 55.1 ng angiotensin I/(mL * h) at one-year follow-up (P = 0.008).

Table 3 - Measured and calculated single-kidney function parameters obtained before stent placement and at one-year follow-up in 18 patients.

Full table

The renal extraction ratios of ¹³¹I-hippuran improved at the treated side (0.48 0.049 to 0.62 0.034; P = 0.003) and contralaterally (0.67 0.033 to 0.73 0.026; P = 0.043). The extraction ratio of ¹²⁵I-thalamate, which equals the filtration fraction, improved at both sides (0.12 0.014 to 0.17 0.012 at the treated side; P = 0.001; and 0.18 0.013 to 0.22 0.011 contralaterally, respectively; P = 0.002). The total (two-kidney) ERPF did not change (323 23.1 to 333 27.8 mL/min; P = NS), and the same was true for the total effective renal blood flow (575 45.6 to 564 66.7 mL/min; P = NS) and total GFR (76 7.2 to 77 7.8 mL/min; P = NS).

The calculated RPF value of the treated kidney was 226 24.5 mL/min before stent placement and 212 19.8 mL/min at one-year follow-up (P = NS; Table 3). Calculated RPF of the contralateral kidney decreased significantly from 342 33.2 to 284 22.3 mL/min (P = 0.013). The calculated ERPF of the treated kidney was 106 15.6 mL/min before stenting and 136 15.4 mL/min at follow-up (P = 0.088). The calculated ERPF of the contralateral kidney was 227 19.2 before stenting and 213 19.4 mL/min at follow-up (P = 0.039). The flow improvement of the treated kidney, indicated by these calculations, is in agreement with the normalization of the renal vein-to-artery renin ratio of the treated kidney. Calculated GFR of the treated side rose significantly from 27 4.5 to 38 4.7 mL/min (P = 0.020) and did not change contralaterally (59 7.1 to 64 6.4 mL/min; P = NS).

The conclusions of the analysis in the subgroup of patients (N = 9) with unchanged medication regimen were consistent with those of the entire patient group. It appeared that the 18 included patients had better baseline renal function than excluded patients with a serum creatinine of 104 9.4 vs. 166 23.1 mol/L (P = 0.019). There was no difference in clinical outcome between patients who developed and those who did not develop restenosis. Finally, we found no relationship between the parameters of kidney length and measured functional data with the (change of) clinical parameters in this study.

DISCUSSION

Evaluation of the effect of renal artery stent placement on renal function is hampered by limited data on the function of the kidneys separately. In the present study single-kidney function was determined before and one year after renal artery stent placement.

There was a decrease in the vein-to-artery renin ratio of the treated kidney and an increase contralaterally. This finding extends previous studies on the effects of balloon angioplasty without stenting, which reported that the lateralization of renin secretion to the stenosed kidney together with contralateral suppression had normalized 6 to 39 months after the procedure^7,8. Under steady state conditions, an increased vein-to-artery renin ratio of a kidney perfused by a stenosed renal artery is a reflection of decreased renal blood flow rather than increased renin secretion. Consequently, the normalization of the vein-to-artery renin ratio observed in our study after one year primarily reflects a lasting increase of blood flow to the treated kidney. We observed a decrease in aortic renin concentration, which indicates a decrease of the two-kidney release of renin. Since the renin release of the contralateral kidney was suppressed before the intervention, the observed change in arterial renin levels can be considered as evidence indicating a decrease in renin release by the treated kidney.

Data on single-kidney function measurements before and after renal artery stent placement are, to the best of our knowledge, not available in the literature. The present study shows that the extraction ratios of ¹³¹I-hippuran and ¹²⁵I-thalamate of the treated kidney improved at one-year follow-up. Before intervention, the extraction ratio of ¹³¹I-hippuran was impaired, which is in accordance with the literature³. This may be explained by a redistribution of renal blood flow in the presence of a renal artery stenosis in such a way that the sites where ¹³¹I-hippuran is excreted by the tubules are bypassed. An alternative explanation is a reduction in urine flow resulting in some tubular reabsorption of ¹³¹I-hippuran. The increase in ¹²⁵I-thalamate extraction of the treated kidney reflects an increase of filtration fraction and, probably, is due to an increase in GFR. Remarkably, both the affected and the contralateral kidneys experienced an improvement in function after renal artery stent placement. This was in accordance with data of Farmer et al, who performed single-kidney GFR measurements using ⁵¹CrEDTA and ^99mTc scintigraphy⁹. The explanation of this finding is at present unclear.

Thus far, we have discussed our data in semiquantitative terms. In order to obtain quantitative information, we used calculations of single-kidney RPF, ERPF and GFR on the basis of the flow ratio (treated/contralateral kidney) as derived from the MAG₃ time-activity curve. The implicit assumption was that the extraction ratios of the intravenously infused radiofarmacon were not different between the treated and contralateral kidney. Our study, however, shows that the extraction ratio of ¹³¹I-hippuran at the treated side was smaller than the contralateral side. Therefore, our calculated data underestimated the true renal plasma flow of the treated kidney and, consequently, the GFR of the treated kidney. It should be noted that MAG₃-scintigraphy was performed with a captopril challenge, which is known to reduce the extraction ratio of ¹³¹I-hippuran³. The direction of changes as shown by our calculations, however, is in agreement with the conclusions derived from the renin measurements.

The two sets of data showed an increase of ERPF and GFR after stenting at the treated side. The total GFR was unchanged, which may be explained by the fact that differences in function of the contralateral kidney counterbalanced the beneficial effect on the treated kidney. Our finding that the filtration fraction of the contralateral kidney was increased after stent placement, suggests a proportionally larger decrease in RPF than in GFR. This was in agreement with the calculated data of the contralateral kidney.

Although renal artery stent placement resulted in an improvement of renal function in the patient group as a whole, on an individual basis there was a wide range of responses. These may have been caused by restenosis or by progressive renal disease. We found no significant differences in single-kidney function data between the patients with and without restenosis. Our analysis, however, was limited to a small group of patients.

The clinical management of renal artery stenosis remains a controversial subject. Our group recently found that there was no significant advantage of balloon angioplasty over medical treatment for the treatment of renal artery stenosis¹. Concurrently, a randomized study comparing balloon angioplasty with stent placement for renal artery stenosis found no differences in final clinical outcome². This might lead to the conclusion that there is no place for revascularization of renal artery stenosis. However, one should keep in mind that, especially in the first study, there was crossover in almost one half of the patients treated (that is, patients who were randomized for medical therapy were withdrawn from the study to undergo balloon angioplasty), which limits interpretation of the results. The present study was meant to investigate the pathophysiological effect of stent placement on the function of the two kidneys separately. These data are lacking in the literature and were not supplied by the mentioned randomized studies. The results of our study show that renal artery stent placement leads to a slight, but statistically significant improvement of single-kidney function. This study, however, may be biased in that the included patients had a better baseline renal function than the excluded patients. Yet, with regards to the current trend in the literature toward expressing the effect of revascularization in terms of stabilizing rather than improving renal function, the results of our present study should at least be considered to be positive. Further studies should be performed to elucidate the subgroup of patients who may clinically benefit from the intervention.

Some limitations of the present study should be addressed. First, the study group was small, which limits extrapolation of the results. Further studies with larger patient groups are mandatory to confirm our findings. Second, serial single-kidney measurements were available in only 18 of the 40 patients. This may have caused a selection bias, because, as mentioned earlier, included patients had a somewhat lower serum creatinine at baseline than the excluded patients. Third, as explained earlier in this article, the renal blood flow measurements derived from MAG₃-scintigraphy should be interpreted with caution. Fourth, the single-kidney measurements may have been influenced by the medication used at the time of renal vein blood sampling. This, however, seems less likely since the conclusions derived from the analysis in patients with unchanged medication were consistent with those derived from the entire patient group.

In conclusion, renal artery stent placement is capable of causing a slight improvement of glomerular filtration rate of the treated kidney, although the overall glomerular filtration rate did not change.

References

References

Appendices

APPENDIX

The renal plasma flow of the treated side (RPF_tr) and contralaterally (RPF_cl), the effective renal plasma flow of the treated side (ERPF_tr) and contralaterally (ERPF_cl), together with the glomerular filtration rate of the treated side (GFR_tr) and contralaterally (GFR_cl) were calculated as follows:

The extraction ratios can be written as follows:

(Eq. 1)

(Eq. 2)

(Eq. 3)

(Eq. 4)

In which the subscripts tr and cl denote the treated and contralateral sides, respectively.

The renal plasma flow ratio between the two kidneys is given by:

(Eq. 5)

In which Flowcounts is the area of the time-activity curve during the first 60 seconds, thereby assuming that this part of the curve is determined solely by the renal plasma flow.

From equations quation1, 2 and 5, it follows that:

(Eq. 6)

Total effective renal plasma flow is defined as:

(Eq. 7)

in which C-hip = total clearance of ¹³¹I-hippuran.

From equations quation1, 2, 5, 6 and 7 it follows that:

(Eq. 8)

(Eq. 9)

From equations quation1, 2, 8 and 9 it follows that:

(Eq. 10)

(Eq. 11)

From equations quation3, 4, 8 and 9, it follows that:

(Eq. 12)

(Eq. 13)