Comparing outcomes in renal replacement therapy: how should we correct for case mix?

The need to evaluate the effectiveness of clinical practice to justify expensive therapy in the face of financial constraints in all areas of health care delivery makes it necessary to identify groups of patients who are likely to benefit most from treatment. Various risk stratification methods have been used for analyzing survival probabilities for patients receiving renal replacement therapy. Complicated risk stratification methods produce large numbers of risk groups of small sizes, which makes comparison between individual centers difficult. We compared three simple methods of risk stratification, that divided patients into low-, medium-, and high-risk groups, in a cohort of 1,407 patients who commenced renal replacement therapy in five European countries during a 7-year period. Method 1 considered age (>55 years) and diabetes alone; method 2 used a higher age limit (>70 years) and comorbid illnesses, including those other than diabetes; and method 3 used only the number of comorbidities (none, 1, or > or =2) for stratification. Kaplan-Meier survival curves were constructed for comparison between risk groups and Cox's regression model used to assess strength of relationship with mortality. Although patient survival was significantly different between the low-, medium-, and high-risk groups using all three methods, Cox's regression analysis showed that method 2 provided the greatest discrimination between risk groups. In predicting mortality, method 2 (based on comorbidities and age) showed the highest sensitivity and specificity (84% and 80%, respectively) compared with method 1 (80% and 74%) and method 3 (64% and 82%). Validation of this approach in other populations in a prospective study is required before this method, which takes into account the influences of both age and comorbidity for risk stratification, can be used for comparing survival data and for presenting results of renal replacement therapy.

Regulation of intestinal vitamin D receptor expression in experimental uraemia: effects of parathyroidectomy and administration of PTH

In this study, the effects of PTH on binding of [3H]-1,25-dihydroxyvitamin D3 [1,25(OH)2D3] and on vitamin D receptor (VDR) mRNA concentration were assessed in intestinal mucosa of subtotally nephrectomized rats (Nx) and in intestinal mucosa of sham-operated rats with normal kidney function (Intact). Intestinal 1,25(OH)2D3 binding capacity of Intact remained unchanged (i) after parathyroidectomy (PTx), (ii) after administration of PTH for up to 6 days, and (iii) after PTx and subsequent administration of PTH (n = 4 experiments). In contrast, PTx of subtotally nephrectomized animals (Nx-PTx) decreased 1,25(OH)2D3 binding capacity from 757 +/- 95 fmol/mg protein in Nx to 417 +/- 42 in Nx-PTx (P < 0.01, n = 5). PTH administration had no effect on intestinal 1,25(OH)2D3 binding capacity in Nx. However, PTH administration to Nx-PTx resulted in re-elevation of 1,25(OH)2D3 binding capacity to a level (790 +/- 113 fmol/mg protein) which was comparable to Nx. Kd-values remained unaltered under all experimental conditions. The intestinal VDR mRNA concentration (normalized to beta-actin mRNA) was decreased, on average, by 23% in Nx-PTx (P < 0.05 versus Nx). In further experiments, 1,25(OH)2D3 was administered to Nx-PTx. This resulted in upregulation of 1,25(OH)2D3 binding capacity as compared to vehicle-treated Nx-PTx (562 +/- 90 fmol/mg protein versus 249 +/- 32, P < 0.01). The latter results could indicate that PTH-mediated stimulation of residual renal 1,25(OH)2D3 production was involved in PTH-mediated up-regulation of intestinal 1,25(OH)2D3 binding capacity in Nx-PTx. To rule out this possibility, PTH was administered to totally nephrectomized and parathyroidectomized rats (TNx-PTx). Since PTH caused an approximately 80% increase (P < 0.05) in intestinal 1,25(OH)2D3 binding capacity under those experimental conditions a mediator role of 1,25(OH)2D3 could be excluded. Functional significance of decreased intestinal 1,25(OH)2D3 binding capacity in Nx-PTx as compared to Nx was demonstrated by significantly lower 1,25(OH)2D3-mediated stimulation of intestinal 25(OH)D3-24-hydroxylase activity in Nx-PTx (209 +/- 68 pmol/mg protein) than in Nx (385 +/- 81, P < 0.01). The modulation of intestinal 1,25(OH)2D3 binding capacity was not correlated with changes in calcium, phosphate or 1,25(OH)2D3 serum concentrations under our experimental conditions. Taken together, intact parathyroid gland function was required to maintain adequate intestinal VDR expression in experimental uraemia (but not in normal animals). The mechanism of the modulation of intestinal VDR by PTH remains to be elucidated although an indirect effect of PTH on VDR expression in intestinal mucosa seems most likely.

1,25(OH)2-vitamin D3 reduces spontaneous and hypocalcemia-stimulated pulsatile component of parathyroid hormone secretion

To investigate the effects of 1,25(OH)2-vitamin D3 (1,25(OH)2D3) on pulsatile parathyroid hormone (PTH) release, minute-to-minute intact PTH secretion was examined in nine healthy adults under baseline conditions and during hypocalcemia (sodium citrate clamp) before and after 5 d of oral 1,25(OH)2D3 treatment (1 microgram/d). In addition, acute effects of 1,25(OH)2D3 were examined by a single intravenous bolus of 2 micrograms of 1,25(OH)2D3. Pulsatile and tonic PTH secretion rates were calculated by the multiparameter deconvolution technique. During baseline, 68% of circulating PTH was attributable to tonic, and 32% to pulsatile, secretion. During induction of hypocalcemia, a selective increase in pulsatile secretion (+1100%), mediated by a combined increase in burst frequency and burst mass, was observed. During the subsequent steady-state hypocalcemic period, burst frequency and mass decreased and tonic secretion increased to 3 times the baseline level. Intravenous 1,25(OH)2D3 did not affect the temporal pattern of PTH secretion. In contrast, oral 1,25(OH)2D3 decreased baseline plasma PTH by 30% without a detectable change in Ca2+. This suppression was accounted for mainly by a decrease in PTH burst frequency. During hypocalcemia induction, a significantly lower (30%) increase in burst mass occurred compared with the pretreatment study. During steady-state hypocalcemia, PTH burst mass (-45%) and pulsatile (-50%) and total (-35%) secretion rate were lower than before treatment. In conclusion, acute hypocalcemia selectively increases pulsatile PTH release by stimulating both burst frequency and mass via a Ca2+ rate-sensitive mechanism. Oral 1,25-(OH)2D3 suppresses pulsatile baseline PTH release and reduces the pulsatile secretory capacity of the parathyroids during a hypocalcemic stimulus.

Association of a human G-protein beta3 subunit variant with hypertension

Hypertension is a common disorder of multifactorial origin that constitutes a major risk factor for cardiovascular events such as stroke and myocardial infarction. Previous studies demonstrated an enhanced signal transduction via pertussis toxin-sensitive G proteins in lymphoblasts and fibroblasts from selected patients with essential hypertension. We have detected a novel polymorphism (C825T) in exon 10 of the gene encoding the beta3 subunit of heterotrimeric G proteins (GNB3). The T allele is associated with the occurrence of a splice variant, GNB3-s (encoding G beta3-s), in which the nucleotides 498-620 of exon 9 are deleted. This in-frame deletion causes the loss of 41 amino acids and one WD repeat domain of the G beta subunit. By western-blot analysis, G beta3-s appears to be predominantly expressed in cells from individuals carrying the T allele. Significant enhancement of stimulated GTPgammaS binding to Sf9 insect cells expressing G beta3-s together with G alpha(i)2 and G gamma5 indicates that this splice variant is biologically active. Genotype analysis of 427 normotensive and 426 hypertensive subjects suggests a significant association of the T allele with essential hypertension.

Genetic abnormalities in parathyroid nodules of uremic patients

The molecular pathway of autonomous growth of the parathyroid glands in uremic patients is poorly understood. We have analyzed 71 parathyroid lesions from 24 patients with refractory hyperparathyroidism for allelic loss at chromosomes 1, 3, 6, 9, 11, 12, 13, 15, and 17 and at the X chromosome. Microsatellite analysis was performed using 24 highly polymorphic markers. Deletions at chromosomes 1, 3, 6, 11, 12, and 13 and at the X chromosome were detected in only 10 of 67 nodules (15%). No allelic loss of the p16 and p53 tumor suppressor genes or the extracellular calcium receptor gene was found. The X-chromosome inactivation assay revealed a monoclonal pattern in 58% of hyperplastic nodules in females. Our results indicate monoclonal growth in the majority of hyperplastic nodules and suggest that some of these lesions might be considered precursors for adenoma development.

The renoprotective effect of angiotensin-converting enzyme inhibitors in experimental chronic renal failure is not dependent on enhanced kinin activity

BACKGROUND

Angiotensin-converting enzyme (ACE) inhibitors have been shown to ameliorate the progression of glomerulosclerosis both in experimental models of uraemia and in patients with renal failure. It has not been documented, however, whether this is due to a decrease in angiotensin II generation or is a consequence of elevated local level of bradykinin.

METHODS

Morphometric investigation of renal tissue was performed in 5/6 nephrectomized (SNx) rats, i.e. untreated or treated with the ACE inhibitor ramipril (SNx-RAM), the B2 kinin receptor antagonist HOE 140 (SNx-HOE), or a combination of both (SNx-RAM + HOE) over 8 weeks. A further group of SNx received delayed treatment with ramipril from week 5 onward (SNx-RAMD). In addition, a sham-operated (SHAM) control group was studied.

RESULTS

Systolic blood pressure was significantly lower in both SNx-RAM and SNx-RAM + HOE groups compared to (untreated) SNx. The glomerulosclerosis index (GSI) was substantially higher in the (untreated) SNx group (0.24 +/- 0.04) vs SHAM (0.02 +/- 0.01). A significantly higher GSI was found in the SNx-HOE group (0.45 +/- 0.08) as compared to (untreated) SNx. However, in the SNx-RAM, SNx-RAM + HOE, and SNx-RAMD groups, the GSI was lowered to a similar extent (0.1 +/- 0.02, 0.09 +/- 0.02, and 0.07 +/- 0.01 respectively). In addition, a concomitant attenuation of tubulointerstitial damage was noted in all the above groups.

CONCLUSION

Increased kinin activity does not appear to play a major role in the renoprotective effect of ACE inhibitors in the remnant kidney model.

Association of chronic kidney graft failure with recipient blood pressure. Collaborative Transplant Study

Immunological rejection is the most important cause of kidney transplant failure. Recently, nonimmunological causes of long-term allograft failure have become more widely appreciated. In primary chronic renal disease, blood pressure is of overriding importance for long-term renal function. The role of blood pressure in determining long-term transplant outcome has not yet been established. We studied the influence of blood pressure post-transplantation on long-term kidney graft outcome in 29,751 patients. Outpatient blood pressure measurements were recorded and reported to the Collaborative Transplant Study. Graft and patient survival rates were analyzed over seven years in relation to blood pressure. Increased levels of systolic and diastolic blood pressure post-transplantation were associated with a graded increase of subsequent graft failure (P < 0.0001). Chronic graft failure was significantly associated with blood pressure even when patient death was censored (P < 0.0001). Cox regression analysis established increased blood pressure as an independent risk factor for graft failure. We conclude that post-transplant blood pressure is a highly significant predictor of long-term kidney graft outcome. Whether aggressive lowering of blood pressure improves long-term transplant outcome will have to be studied prospectively.

Long-term progression of chronic renal insufficiency in the AIPRI Extension Study. The Angiotensin-Converting-Enzyme Inhibition in Progressive Renal Insufficiency Study Group

The Angiotensin-converting-enzyme Inhibition on Progressive Renal Insufficiency (AIPRI) Study showed that the ACE inhibitor benazepril provides protection against loss of renal function in patients with chronic renal insufficiency (CRI) caused by various renal diseases. As a result of unexpectedly low mortality in the placebo group, there was a substantial imbalance in mortality during the course of this study (8 patients on benazepril vs. 1 on placebo). The aim of the extension study was to follow-up the patients from the AIPRI core study until autumn 1996, focusing on CRI progression and mortality. Data collection was post hoc. Patients were treated according to investigators' usual practices, without knowledge of the core study trial medication or (initially) the core trial results. A new primary efficacy parameter was defined as the time from the start of core study treatment to the occurrence of the first event in the combined composite end-point of dialysis, renal transplantation or death related to renal disease. Serial serum creatinine levels and all-cause mortality were also recorded. The median total follow-up for core + extension periods was 6.6 years. Many patients from both treatment groups (64% on benazepril and 61% on placebo) received ACE inhibitors during follow-up. In the intention-to-treat analysis of the core + extension data, only 79 of 300 patients from the benazepril group, compared to 102 of the 283 patients from the placebo group needed dialysis or renal transplantation, or died related to renal disease (P < 0.013, log-rank test). The mortality imbalance seen in the core trial was not evident with the longer follow-up (25 deaths in the benazepril and 23 in the placebo group, before dialysis). These data clearly demonstrate a long-term beneficial effect in patients randomized to take benazepril during the core study, but because treatment during the extension period was not randomized, the results of this intention-to-treat analysis need to be interpreted with care.

Apolipoprotein A, fibrinogen, age, and history of stroke are predictors of death in dialysed diabetic patients: a prospective study in 412 subjects

BACKGROUND

Diabetic patients with end-stage renal failure (ESRD) have a high cardiovascular morbidity and mortality. The underlying mechanisms are not completely elucidated. The aim of our study was to define predictors of death in diabetic patients with end-stage renal disease.

PATIENTS AND METHODS

We performed a prospective study in 35 dialysis centres in Germany between 1985 and 1994. To evaluate predictors and risk factors in this population we examined 412 diabetic patients at the time of admission to dialysis treatment (peritoneal dialysis (PD) or haemodialysis (HD)). Classification of the type of diabetes was done according the criteria of the National Diabetes Data Group [1,2]. Items assessed at the time of admission were coronary artery disease (CAD), peripheral occlusive disease (POD), and stroke. CAD was defined as a history of myocardial infarction with the corresponding changes in the ECG or luminal narrowing by more than 50% in at least one coronary artery upon coronarangiography; POD was defined as claudication and/or brachial-tibial ratio (BTR) less than 0.9 or a history of amputation. Assessment of the nutritional state comprised body mass index, skinfold thickness of the upper arm and lateral thorax area, and urea concentration. Cholesterol, HDL, LDL, apolipoprotein A (ApoA-I) and B (ApoB), triglycerides, lipoprotein(a) (Lp(a)), and fibrinogen were measured. As an index of disturbed cardiac innervation beat-to-beat variation was measured. Outcome measurements were causes of death (i.e. cardiac and non-cardiac) and time of survival.

RESULTS

One hundred and eighty of 412 (44%) patients died during the observation period. Patients who died were older (61 +/- 12 versus 53 +/- 15 years P < 0.0001), had lower skin fold thickness (13.1 +/- 6.0 versus 15.1 +/- 7.3 mm P < 0.04), lower ApoA-I (100 +/- 35 versus 111 +/- 32 mg/dl P < 0.005) and higher fibrinogen (515 +/- 146 versus 451 +/- 155 mg/dl P < 0.02). Type II diabetic patients had a lower mean survival time than type I (34 versus 66 months P < 0.0006). The mode of renal replacement therapy (PD or HD) had no adverse effect on survival time. Survivors less frequently had a history of CAD, POD and stroke than non-survivors. In multivariate analysis ApoA-I, fibrinogen, age and stroke were independent predictors of cardiac and non-cardiac death in diabetic patients with end-stage renal failure. Lipid values and nutritional state did not independently predict the overall and cardiovascular mortality.

CONCLUSION

This study in dialysed diabetic patients identifies several predictors of death, some of which are susceptible to intervention.

Renal expression of two rat kallikrein genes under diabetic conditions

OBJECTIVE

We have reported that bradykinin (BK) excretion is increased in severely diabetic rats, independent of the activity of the main renal kinin-forming enzyme, true kallikrein (KLK). To further investigate the relationship between renal BK excretion and renal KLK in diabetes we studied the regulation of the renal kallikrein-like gene, rat kallikrein 7 (rKLK7), as well as of the KLK encoding gene, rKLK1, in streptozotocin-induced (STZ) diabetic rats.

METHODS

Experiments were performed in STZ-induced diabetic male Wistar rats and their non-diabetic controls (n = 7 each group). Twelve weeks after STZ injection, urinary KLK activity, glomerular filtration rate and total protein excretion were determined. After extraction of total renal cortical RNA, specific oligonucleotides were used to generate a reverse transcription-polymerase chain reaction (RT-PCR) products of renal cortical rKLK1 and rKLK7 messenger (m)RNA. Southern blot analysis of these RT-PCR products were hybridized with appropriate gene-specific oligonucleotide probes.

RESULTS

After 12 weeks, the rats showed hyperglycemia, proteinuria and a reduced glomerular filtration rate. Renal kininogenase was reduced, as indicated by a reduction in the expression of rKLK1, as well as of the KLK-related gene, rKLK7.

CONCLUSIONS

Our data show that the expression of the two principal renal KLK genes is downregulated in the renal cortex of STZ-diabetic rats. We suggest that under severe diabetic conditions the rise in urinary BK excretion is not related to activation of the renal kinin-forming enzyme system.

A polymorphism in the gene for the angiotensin II type 1 receptor is not associated with hypertension

BACKGROUND

A mutation in the gene for the angiotensin II type 1 (AT1) receptor (A1166C) has been reported to be associated with primary hypertension.

OBJECTIVE

To determine whether this observation could be confirmed with a different population sample.

DESIGN

We examined 414 individuals with primary hypertension and 172 normotensive controls.

METHODS

The mutation in the gene for the AT1 receptor was detected using restriction polymerase chain reaction.

CONCLUSIONS

We detected no association of the AT1 receptor polymorphism with hypertension, but a trend towards a decreased prevalence of the 1166C allele among hypertensive patients with a late age at diagnosis (> or = 50 years) was observed.

Genetic determinants of diabetic renal disease and their impact on therapeutic interventions

Approximately 30% of patients with type 1 and type 2 diabetes develop diabetic nephropathy. Apart from metabolic control, genetic predisposition plays an important role in its genesis. Analysis of intermediate phenotypic markers showed that the activity of Na/Li- and Na+/H(+)-countertransport is increased in patients with diabetic nephropathy. The renin-angiotensin system is of crucial importance as a system for therapeutic intervention and as genetic marker for susceptibility to renal disease. Consequently, the analysis of molecular genetic markers has focused on a polymorphism in the gene for the angiotensin II converting enzyme (ACE). However, the analysis of the I/D-polymorphism with respect to development of diabetic nephropathy in type 1 and type 2 diabetes has yielded conflicting results, at least in type 1 diabetes. These discrepant results may be due to differences in definition, sample size and ethnic background of the patients. In IgA glomerulonephritis it has been shown that the DD genotype (which is correlated with higher serum and tissue ACE activity compared to II genotype) is associated with a more rapid deterioration of renal function. The same adverse effect of the DD genotype could also be demonstrated in patients with diabetic nephropathy. Two studies examined the response to treatment according to the different genotypes, with contradictory results. A Japanese study showed a more pronounced reduction in proteinuria under ACE inhibitor treatment in patients with DD genotype, whereas a Danish study showed that patients with the DD genotype exhibited a steeper decline in renal function despite ACE inhibitor treatment. The data available for other candidate genes are fragmentary and negative throughout.

[Hypertension and the kidney]

In the past 30 years, the paradigms to explain the relation between hypertension and progression of renal disease have changed in a dramatic fashion. Thirty years ago it was believed that high blood pressure was necessary to permit the damaged kidney to maintain its function. In contrast, today solid evidence is available that blood pressure is the most important factor accelerating progressive loss of renal function. Furthermore, on the basis of recent observations, target blood pressure during antihypertensive therapy has been substantially lowered. Finally, it could be documented that for interfering with progressive loss of renal function, ACE (angiotensin converting enzyme) inhibitors and calcium channel blockers are superior to alternative antihypertensive agents (nephroprotection).

Rapidly progressive IgA nephropathy with anti-myeloperoxidase antibodies benefits from immunosuppression

CLINICAL OBSERVATIONS

Three patients with previous pulmonary infections were recently admitted with rapidly progressive renal failure. Renal biopsy showed crescentic glomerulonephritis with deposits of IgA, C3c and C3d. Serology disclosed P-ANCA with high-titer anti-myeloperoxidase antibodies. Two out of three patients became dialysis dependent despite immunosuppression with methylprednisolone and cyclophosphamide. Renal function improved in both patients after 2 weeks and 9 months, respectively, permitting termination of hemodialysis. All patients benefited from immunosuppressive treatment which is currently still being continued.

CONCLUSION

The data suggest that early immunosuppression is beneficial in patients presenting with crescentic rapidly progressive IgA GN and anti-myeloperoxidase antibodies, which may represent a novel subset of crescentic IgA GN associated with high-titer anti-myeloperoxidase antibodies constituting an overlap group between microscopic polyangiitis and IgA GN.

Angiotensin II affects basal, pulsatile, and glucose-stimulated insulin secretion in humans

Angiotensin II (Ang II) modulates the tissue response to insulin (insulin sensitivity), but the effect of Ang II on the secretion of insulin has not been investigated thus far. Nineteen healthy volunteers (17 male; mean age, 26+/-1 years) were studied. In a double-blind, randomized, placebo-controlled study, seven volunteers were allocated on three occasions in random order after an overnight fast to three interventions: (1) solvent (placebo) infusion; (2) infusion of 1.0 ng Ang II x kg(-1) x min(-1) (subpressor dose); and (3) infusion of 5.0 ng Ang II x kg(-1) x min(-1) (pressor dose). Frequent blood samples (each minute) were obtained for estimation of plasma insulin concentrations over a period of 120 minutes to assess basal and pulsatile insulin secretion. In an ancillary study, plasma glucose and insulin levels were measured after an oral glucose tolerance test while solvent (placebo) or Ang II was infused in 12 fasting healthy volunteers. Plasma insulin concentrations were measured immunoenzymatically (enzyme-linked immunosorbent assay). Insulin secretion pulses were analyzed with the deconvolution technique, and the regularity of insulin secretion was analyzed with the approximate entropy technique. Plasma insulin half-life was assessed using the hyperinsulinemic euglycemic clamp method. The pressor dose of Ang II reduced total, basal, and pulsatile insulin secretion, and this effect was highly significant (P<.01). The subpressor dose tended to suppress insulin secretion. The burst frequency (number of peaks) and the regularity of insulin secretion were not affected by administration of Ang II. After the oral glucose load, the insulinemic response was significantly lower and plasma glucose concentrations were significantly higher with infusion of Ang II compared with placebo. Ang II affects both the basal (nonpulsatile) and the pulsatile component of spontaneous insulin secretion and the glucose-stimulated insulin secretion in humans. This observation is of potential interest with respect to the interaction of Ang II and insulin, eg, in the genesis of hyperinsulinemia and hypertension.

Abnormal pulsatile secretion of growth hormone in non-insulin-dependent diabetes mellitus

OBJECTIVE

Studies of GH secretion in patients with non-insulin dependent diabetes mellitus (NIDDM) have produced conflicting results. We aimed to differentiate the effects of obesity and metabolic control on pulsatile GH secretion in patients with NIDDM.

DESIGN

Blood sampling every 15 min from 22.00 hours to 08.00 hours after a fasting period of at least 3 h.

PATIENTS

13 male NIDDM patients, 9 healthy control subjects matched for age and BMI, and 6 lean subjects matched for age.

MEASUREMENTS

Measurement of GH by a novel ultrasensitive chemiluminescence assay. Analysis of concentration vs time profiles by a multiparameter deconvolution technique.

RESULTS

GH burst frequency was increased in the NIDDM (0.82 +/- 0.28 h-1) compared with both control groups (lean: 0.6 +/- 0.11; obese: 0.56 +/- 0.19). GH burst mass was decreased in patients (1.57 +/- 0.98 micrograms/l.min) and in obese controls (1.46 +/- 1.44) compared to lean controls (3.71 +/- 3.88). These differences resulted in a significantly higher nocturnal pulsatile GH secretion rate in the lean compared to the obese controls, whereas in the patient group enhanced GH burst frequency compensated for reduced burst mass. The characteristics of GH secretion were not related to nocturnal or early morning blood glucose concentrations. However, GH secretion rate was correlated positively with HbA1c (r = 0.57; P = 0.04), and negatively with plasma C peptide concentrations.

CONCLUSIONS

The specific increase in GH burst frequency previously described in insulin-dependent diabetes mellitus is also present in NIDDM. However, GH hypersecretion does not occur because GH burst mass is reduced in proportion to the degree of obesity. The effect of diabetes on the hypothalamic control of GH release appears to be determined by the quality of long-term glycaemic control.