Abnormal cation exchange in insulin-resistant patients with essential hypertension.

OBJECTIVES: To identify important factors that may contribute to abnormal glucose tolerance in elderly patients with treated hypertension with primary reference to changes in the following parameters: calculated insulin resistance, endogenous insulin processing and secretion; platelet cation concentration and membrane ATPase activity. DESIGN: Thirty-nine patients receiving antihypertensive therapy (including low-dose thiazide treatment) were compared to 13 normotensive, normoglycaemic control subjects. Total platelet cation concentration and membrane ATPase activity were measured and, following a 75-g oral glucose test, serum insulin, proinsulin and 31-32 des-proinsulin responses were measured in prospectively defined hypertensive patients with normal glucose tolerance (NG), impaired glucose tolerance (IGT) and diabetes mellitus (DM). RESULTS: Of the total patient cohort, seven patients manifested newly diagnosed DM, 18 had IGT and 14 NG. Among the three groups, no difference in duration of drug use (thiazides and beta-blockers) was noted; BMI and waist-to-hip ratio increased progressively from NG to IGT to overt DM. Compared to NG patients, serum insulin responses were significantly greater in the IGT (all time points) and DM (two-hour measurements) subjects. Proinsulin and 31-32 des-proinsulin serum responses were likewise significantly higher in the IGT and DM groups. The derived measure of insulin resistance in the hypertensive patients showed a significant increase in the progression from NG to IGT and DM. Mean total platelet potassium concentration was reduced in the DM compared to the IGT and the control groups, while platelet sodium, calcium and magnesium concentrations showed no significant differences. Platelet membrane magnesium ATPase activity was significantly higher in the normotensive control versus the hypertensive group. Sodium, potassium and calcium ATPase activity showed no significant differences among the subgroups. CONCLUSION: Our findings support the strong link between essential hypertension, insulin resistance/hyperinsulinaemia and regional adiposity. Beta-cell dysfunction (hypersecretion and abnormal insulin processing) is manifest in the progression from normality to overt diabetes. The use of antihypertensive therapy (low-dose thiazides and cardioselective beta-blockers) possibly added diabetogenic effect(s). The reduction in platelet total potassium concentration paralleled the diabetic state while a reduced membrane magnesium ATPase activity correlated with the hypertensive state.

Abnormal Cation Exchange in Insulin-Resistant Patients with Essential Hypertension : Cardiovascular Topics

Objectives: To identify important factors that may contribute to abnormal glucose tolerance in elderly patients with treated hypertension with primary reference to changes in the following parameters: calculated insulin resistance; endogenous insulin processing and secretion; platelet cation concentration and membrane ATPase activity. Design: Thirty-nine patients receiving antihypertensive therapy (including low-dose thiazide treatment) were compared to 13 normotensive; normoglycaemic control subjects. Total platelet cation concentration and membrane ATPase activity were measured and; following a 75-g oral glucose test; serum insulin; proinsulin and 31-32 des-proinsulin responses were measured in prospectively defined hypertensive patients with normal glucose tolerance (NG); impaired glucose tolerance (iGT) and diabetes mellitus (DM). Results: of the total patient cohort; seven patients manifested newly diagnosed DM; 18 had iGT and 14 NG. Among the three groups; no difference in duration of drug use (thiazides and beta-blockers) was noted; BMi and waist-to-hip ratio increased progressively from NG to iGT to overt DM. Compared to NG patients; serum insulin responses were significantly greater in the iGT (all time points) and DM (two-hour measurements) subjects. Proinsulin and 31-32 des-proinsulin serum responses were likewise significantly higher in the iGT and DM groups. The derived measure of insulin resistance in the hypertensive patients showed a significant increase in the progression from NG to iGT and DM. Mean total platelet potassium concentration was reduced in the DM compared to the iGT and the control groups; while platelet sodium; calcium and magnesium concentrations showed no Significant differences. Platelet membrane magnesium ATPase activity was significantly higher in the normotensive control versus the hypertensive group. Sodium; potassium and calcium ATPase activity showed no significant differences among the subgroups. Conclusion: our findings support the strong link between essential hypertension; insulin resistance / hyperinsulinaemia and regional adiposity. Beta-cell dysfunction (hypersecretion and abnormal insulin processing) is manifest in the progression from normality to overt diabetes. The use of antihypertensive therapy (low-dose thiazides and cardioselective beta-blockers) possibly added diabetogenic effect(s). The reduction in platelet total potassium concentration paralleled the diabetic state while a reduced membrane magnesium ATPase activity correlated with the hypertensive state

Hypertension guideline 2003 update.

OUTCOMES: Extensive data from many randomised controlled trials have shown the benefit of treating hypertension. The target blood pressure (BP) for antihypertensive management should be systolic BP < 140 mmHg, diastolic < 90 mmHg, with minimal or no drug side-effects. However, a lesser reduction will elicit benefit although this is not optimal. The reduction of BP in the elderly and in those with severe hypertension should be achieved gradually over 6 months. Stricter BP control is required for patients with end organ damage, co-existing risk factors and co-morbidity, e.g. diabetes mellitus. Co-existent risk factors should also be controlled. BENEFITS: Reduction in risk of stroke, cardiac failure, renal insufficiency and probably coronary artery disease. The major precautions and contraindications to each antihypertensive drug recommended are listed. RECOMMENDATIONS: Correct BP measurement procedure is described. Evaluation of cardiovascular risk factors and recommendations for antihypertensive therapy are stipulated. The total cardiovascular disease risk profile should be determined for all patients and this should inform management strategies. Lifestyle modification and patient education plays an essential role in the management strategy. Drug therapy: First line--low dose thiazide-like diuretics; second line--add one of the following: reserpine, or beta-blockers or ACE inhibitors or calcium channel blockers; third line--add another second line drug or hydralazine or alpha-blocker. The guideline includes management of specific situations, i.e. hypertensive emergency and urgency, severe hypertension with target organ damage and refractory hypertension (BP > 160/95 mmHg on triple therapy), hypertension in diabetes mellitus, etc. VALIDITY: Developed by the Working Groups established by the Executive Committee of the Southern African Hypertension Society with broader consensus meeting endorsement. The 2001 version was endorsed by the South African Medical Association Guideline Committee. The 2003 revisions were endorsed by the Executive Committee and a wider Working Group.

Magnesium supplementation attenuates, but does not prevent, development of hypertension in spontaneously hypertensive rats.

This study evaluated the effects of magnesium supplementation on the development of hypertension in spontaneously hypertensive rats (SHR) and assesses the actions of magnesium on extracellular and intracellular Ca2+, Na+, and K+ status. Ten-week-old SHR (n = 72) and Wistar-Kyoto rats (WKY) (n = 60) were divided into four groups: WKY, Mg-WKY (WKY receiving 650 mg/L Mg(2+)supplementation), SHR, and Mg-SHR (SHR receiving Mg2+). Rats were studied for 17 weeks. Serum and erythrocyte Mg2+ and Ca2+ (measured by atomic absorption spectroscopy) and Na+ and K+ (measured by flame photometry) were determined every 3 or 4 weeks. From 13 weeks of age, blood pressure (BP) was significantly elevated in SHR compared with age-matched WKY. BP was reduced (P < .01) in SHR after 10 weeks of Mg2+ supplementation and at 27 weeks of age, BP in SHR was 218 (12 mm Hg v 184 +/- 9 mm Hg) in Mg-SHR. From 18 weeks of age, serum and intracellular Mg2+ levels were significantly lower (P < .05) and from 21 weeks of age, erythrocyte Ca2+ was significantly higher in SHR than in WKY. Mg2+ supplementation normalized intracellular Mg2+ and Ca2+ concentrations in SHR. BP was inversely correlated with erythrocyte Mg2+ (r = -0.74, P < .01) and positively correlated with erythrocyte Ca2+ (r = 0.78, P < .001). In conclusion, long-term Mg2+ supplementation significantly attenuates, but does not prevent, the development of hypertension in SHR. Furthermore, intracellular Mg2+ deficiency and Ca2+ overload in SHR are normalized by Mg2+ treatment.

Illicit intravenous drug use in Johannesburg--medical complications and prevalence of HIV infection.

OBJECTIVE: To describe the magnitude of the problem of abuse by self-injection of dipianone HCl/cyclizine HCl (Wellconal) and to document the associated morbidity, mortality and prevalence of HIV infection. DESIGN: We conducted a retrospective analysis of 121 admissions of 86 patients who were current intravenous Wellconal abusers and presented to Johannesburg and J G Strijdom Hospitals over an 18-month period. Case records were analysed in respect of age, sex, median hospital stay, complications, HIV antibody status and eventual outcome. MAIN OUTCOME MEASURES: Age, sex, median hospital stay, complications, HIV antibody status and eventual outcome. RESULTS: Complications of Wellconal abuse occurred in young adults (median age 24 years) with an approximately equal gender distribution. Opiate overdose was the most frequent presenting diagnosis (32%), followed by right-sided endocarditis (20%) and deep-vein thrombosis (12%). A wide variety of complications accounted for the remaining 36%. A 2% HIV antibody positivity rate was found, which is substantially lower than that encountered in intravenous drug abusers in other parts of the world. Seventy-eight per cent of patients completed therapy successfully, but 19% left hospital prematurely against medical advice. There was a mortality rate of 3%. CONCLUSIONS: While the prevalence of Wellconal abuse in the broader South African community is unknown, our study draws attention to the extent of the problem in Johannesburg.

Alterations in intracellular cations and cell membrane ATPase activity in patients with malignant hypertension.

OBJECTIVE: To determine whether cellular cation concentrations and cell membrane ATPase activity are altered in patients with malignant hypertension. DESIGN: Sixteen black patients with malignant hypertension were studied and compared with age- and sex-matched essential hypertensive and normotensive subjects. Calcium, magnesium, sodium and potassium concentrations and cell membrane Na,K-ATPase, Ca-ATPase and Mg-ATPase activities were determined in platelets and erythrocytes. METHODS: Intracellular concentrations of total magnesium and calcium were measured by atomic absorption spectrophotometry, and those of sodium and potassium by flame photometry. Cell membrane ATPase activity was measured by a colorimetric method. RESULTS: The intracellular calcium level was significantly elevated and intracellular magnesium and potassium levels and cell membrane ATPase activity significantly decreased in the hypertensive group. These changes were more marked in patients with malignant hypertension than in patients with benign essential hypertension. In the malignant hypertensive group, mean arterial pressure was negatively correlated with intracellular magnesium and positively correlated with intracellular calcium and sodium levels. CONCLUSIONS: Cellular cation changes in malignant hypertension may be related to altered transmembrane transport mechanisms and activation of the renin-angiotensin system. These alterations may be more pronounced in the malignant than in the benign phase of hypertension.

Relations between magnesium, calcium, and plasma renin activity in black and white hypertensive patients.

The heterogeneous status of magnesium and calcium metabolism in the hypertensive population may be related to the plasma renin activity (PRA). This study investigates the relationships between serum and erythrocyte magnesium (Mg2+) and calcium (Ca2+) concentrations and PRA in black and white essential hypertensive patients. Thirty-nine normotensive (20 black, 19 white) and 47 hypertensive (25 black, 22 white) subjects were studied. The PRA was measured by radioimmunoassay, Mg2+ and Ca2+ by atomic absorption spectroscopy, and serum ionized Ca2+ by a specific electrode. PRA and ionized Ca2+ were significantly lower in the black hypertensive as compared with the white hypertensive group (1.99 +/- 0.33 vs. 5.96 +/- 1.02 ng/ml/h for PRA; 1.28 +/- 0.07 vs. 1.42 +/- 0.01 mmol/l for ionized Ca2+: black hypertensives vs. white hypertensives p < 0.05). Ionized Ca2+ was significantly increased (p < 0.05) in the white hypertensive patients as compared with the normotensive controls (1.42 +/- 0.01 vs. 1.29 +/- 0.04 mmol/l). In the black hypertensive group, serum and erythrocyte Mg2+ were significantly (p < 0.05) decreased as compared with the other groups. The erythrocyte Ca2+ concentration was significantly elevated in both black and white hypertensive patients. In the group as a whole, serum Mg2+ and PRA were negatively correlated and ionized Ca2+ and PRA and ionized Ca2+ and erythrocyte Ca2+ positively correlated. However, in the subgroups, these correlations were only significant in the white group: r = -0.67 and p < 0.05 serum Mg2+ vs. PRA; r = 0.64, and p < 0.05 ionized Ca2+ vs. PRA; r = 0.82 and p < 0.01 ionized [Ca2+]i vs. erythrocyte Ca2+. These data suggest a relationship between PRA, Mg2+, and Ca2+ which may be more important in white than in black hypertensive patients.

Partial recovery of renal function in black patients with apparent end-stage renal failure due to primary malignant hypertension.

We report the largest series in which 12 out of 54 patients with primary malignant hypertension requiring dialysis recovered sufficient renal function to allow withdrawal of dialysis. The patients were divided into recovery (RC; n = 12) and non-recovery (N-RC; n = 42) groups. The two groups were compared for variables which might predict RC. They were also assessed for survival. Nine of the RC and 6 of the N-RC patients presented with acute oliguria (p = 0.01). The initial mean arterial pressure was significantly higher in the RC than the N-RC group (178 +/- 17 vs. 160 +/- 27 mm Hg; p = 0.03). Although not statistically significant, more females recovered (8 of 12 vs. 16 of 42; p < 0.1). More patients presenting with serum creatinine concentrations < 1,000 mumol/l (11 mg/dl) recovered (p = 0.09), while the presence of microangiopathic-haemolytic anaemia occurred more frequently in the RC (7 of 10) than in the N-RC (15 of 35) group (p = 0.16)> Age, kidney size, and the presence of hypertensive retinopathy did not distinguish between the two groups. RC patients had a greater long-term survival (Mantel-Cox chi2 = 4.48; p = 0.03). The renal function RC may be related to the type of dialysis provided (intermittent peritoneal dialysis) and to the use of modern potent peripheral vasodilator antihypertensive agents. Potential renal function RC should always be considered in patients being dialyzed for primary malignant hypertension.

Racial differences in cell membrane ATPases and cellular cation content in urban South African normotensive and hypertensive subjects.

Platelet and erythrocyte membrane Na+,K(+)-ATPase, Ca(2+)-ATPase and Mg(2+)-ATPase activities and platelet, erythrocyte and serum magnesium, calcium, sodium and potassium concentrations were determined in black (n = 52) and white (n = 24) essential hypertensive patients from the city of Johannesburg in South Africa. The hypertensive groups were matched for age and body mass with black (n = 52) and white (n = 26) normotensive controls. In the black group, platelet and erythrocyte membrane ATPase activities were significantly depressed in the hypertensive subjects. In the white group, there were no significant differences for any of the ATPases studied between the normotensive and hypertensive subjects. Platelet sodium and calcium were significantly increased and serum magnesium, serum potassium, platelet magnesium and erythrocyte magnesium significantly decreased in the black hypertensive group compared to the black normotensive group. In the white hypertensive patients, platelet sodium and calcium were significantly raised and platelet magnesium significantly decreased compared to the normotensive controls. In blacks, platelet magnesium and ATPase activity were negatively correlated with mean arterial pressure. Unlike whites, black hypertensives have widespread magnesium changes with associated cell membrane ATPase depression and cytosolic sodium and calcium accumulation. These results suggest possible racial differences in cellular cation regulation in essential hypertension.

Altered calcium binding to erythrocyte membranes in essential hypertension: relation to magnesium.

Abnormal cellular ion transport resulting in altered membrane control over intracellular calcium may be aetiologically related to essential hypertension. Cell membrane calcium binding, which is influenced by magnesium, is an important membrane system involved in cellular calcium regulation. In this study the relationships between erythrocyte calcium binding and extra- and intracellular calcium and magnesium concentrations were determined in essential hypertensive patients (52 black, 24 white) and normotensive controls (52 black, 26 white). Calcium depletion of the erythrocytes by MgCl2 and EDTA resulted in the removal of more calcium ions from the outer erythrocyte membrane in the hypertensive group compared with the normotensive group. This may be considered as evidence of increased calcium binding to the outer cell membrane in essential hypertension. Calcium binding was greater in the hypertensive males compared with the females. In black hypertensives, serum and erythrocyte magnesium concentrations were significantly increased. In white hypertensives, serum calcium was significantly decreased and erythrocyte calcium significantly raised. Serum and erythrocyte magnesium correlated inversely with the amount of calcium released from the membranes in the black hypertensive group. The results of this study suggest that changes in magnesium levels may contribute to altered cell membrane calcium binding in essential hypertension.

Platelet and erythrocyte Mg2+, Ca2+, Na+, K+ and cell membrane adenosine triphosphatase activity in essential hypertension in blacks.

OBJECTIVE: To assess the relationship between intracellular Mg2+, Ca2+, Na+ and K+ and cell membrane adenosine triphosphatase (ATPase) activity in normotensive and hypertensive blacks. DESIGN: Intracellular cations and cell membrane ATPase activity were studied in black patients with untreated essential hypertension and age-, weight- and height-matched normotensive controls. Platelet, erythrocyte and serum Mg2+, Ca2+, Na+ and K+ levels as well as platelet and erythrocyte membrane Na+,K(+)-ATPase, Ca(2+)-ATPase and Mg(2+)-ATPase activities were measured in all subjects. METHODS: Intracellular Na+ and K+ were measured by flame photometry and Mg+ and Ca+ by atomic absorption spectrophotometry. Cell membrane ATPase activity was determined by a colorimetric method. RESULTS: The hypertensive group consistently demonstrated depressed activity of each ATPase studied, with significantly lower serum Mg2+, serum K+, erythrocyte Mg2+ and platelet Mg2+ levels compared with the normotensive group. Platelet Na+ and Ca2+ and erythrocyte Ca2+ were significantly elevated in the hypertensive group. In the hypertensive group, mean arterial pressure (MAP) was inversely correlated with platelet and erythrocyte membrane Na+,K(+)-ATPase, Ca(2+)-ATPase and Mg(2+)-ATPase. Serum Mg2+, serum Ca2+ and platelet Mg2+ were negatively correlated with MAP in the hypertensive group whilst erythrocyte and platelet Ca2+ were positively correlated. In the normotensive group, platelet Mg2+ and MAP were negatively, and erythrocyte Ca2+ and MAP, positively correlated. CONCLUSIONS: Black patients with essential hypertension have widespread depression of cell membrane Na+,K(+)-ATPase, Ca(2+)-ATPase and Mg(2+)-ATPase activities with serum and intracellular Mg2+ depletion and cytosolic Na+ and Ca2+ overload, which may reflect an underlying membrane abnormality in essential hypertension. These cellular abnormalities may be related to the defective transport mechanisms that in turn may be aggravated by Mg2+ depletion.

Intracellular Mg2+, Ca2+, Na2+ and K+ in platelets and erythrocytes of essential hypertension patients: relation to blood pressure.

Alterations in intracellular cation metabolism have been implicated in the pathophysiology of essential hypertension. Total magnesium, calcium, sodium and potassium levels were studied in serum erythrocytes and platelets, from 154 subjects (76 hypertensive and 78 normotensives; 104 blacks and 50 whites). In the combined black and white hypertensive group, platelet sodium and calcium and erythrocyte calcium were elevated and serum potassium, serum magnesium and platelet magnesium decreased. In the black hypertensive patients, platelet sodium and calcium and erythrocyte calcium were increased, whereas serum magnesium, serum potassium, platelet magnesium and erythrocyte magnesium were decreased. In the white hypertensive group, platelet sodium and erythrocyte calcium were raised and platelet magnesium was decreased. In the black hypertensive patients, serum and platelet magnesium and serum calcium were negatively and erythrocyte and platelet calcium positively correlated with mean arterial pressure. In the white hypertensive patients platelet sodium was directly related to mean arterial pressure. These results suggest that intracellular sodium and calcium overload and magnesium depletion may be important in the pathophysiology of hypertension. Magnesium disturbances are more consistent and widespread in black hypertensive patients than in white hypertensive patients.

The renal histopathology of essential malignant hypertension in black South Africans.

The renal histopathology of essential malignant phase hypertension in two groups of black South Africans was studied. The first, an autopsy series, consisted of material obtained in 1956-1961, a period when adequate antihypertensive therapy had not yet become generally available. The second group, a renal biopsy series, was from an era when effective antihypertensive therapy was available (1979-1989). The study showed that the clinical and histopathological changes of malignant hypertension were similar in both the pre-treatment and treatment eras. Fibrinoid necrosis was found in 92% of the autopsy sections and 44% of the biopsies. While mucinous, onion-skin and fibrotic changes of the blood vessels were commonly found, they were not invariably present and could not be considered the histopathological hallmark of malignant nephrosclerosis. It was often impossible on histological examination to distinguish malignant nephrosclerosis arising de novo from that superimposed on long-standing previous benign hypertension. Replication of internal elastic lamina, which has been considered a marker of long-standing previous benign hypertension, was a nonspecific finding and was frequently observed in young hypertensive subjects with a short clinical history. When fibrinoid necrosis is present, diagnosis of malignant nephrosclerosis can be made with confidence in black South Africans.

Altered cations and muscle membrane ATPase activity in deoxycorticosterone acetate-salt spontaneously hypertensive rats.

The role of ions and cell membrane function in the pathogenesis of benign and malignant hypertension was investigated in spontaneously hypertensive rats (SHR). Ten-week-old male SHR (n = 50) and SHR treated with deoxycorticosterone acetate (DOCA; n = 70) and 1% NaCl drinking water were studied weekly for 14 weeks. Malignant hypertension developed only in DOCA-salt SHR and was characterised by severe hypertension, failure to thrive and renal fibrinoid necrosis. Fourteen DOCA-salt SHR and one SHR died. Extracellular (serum) and intracellular (erythrocyte and muscle) Na+, K+, Mg2+, Ca2+ and muscle membrane Na+,K(+)-adenosine triphosphatase (ATPase), Ca(2+)-ATPase and Mg(2+)-ATPase were measured at various stages in the development of malignant hypertension. Three developmental phases were defined: benign, premalignant and malignant. DOCA-salt SHR showed persistent hypokalaemia. In the benign phase, there were no differences in Na+, Mg2+ and Ca2+ between SHR and DOCA-salt SHR. In the premalignant phase, serum and erythrocyte Mg2+ and ATPase activity were significantly lower in DOCA-salt SHR compared with SHR. During the late premalignant and malignant phases, intracellular Ca2+ and Na+ were significantly higher in the DOCA-salt SHR compared with SHR. In view of these findings, the abnormalities in DOCA-salt SHR during the early phases of blood pressure elevation could be contributory factors to the development of malignant hypertension.

Skeletal manifestations of primary oxalosis.

We describe the clinical, radiographic and histological features of skeletal involvement in four patients with end-stage renal failure due to primary oxalosis. The clinical features were unrelenting bone pain, and in two patients multiple fractures. Radiographic features were, in chronological order: (1) radiodense metaphyses and other red marrow bone; (2) cortical defects in metaphyses; (3) spontaneous fracture-separations of epiphyses of long limb bones which healed poorly. The fractures occurred through crystal deposits, and fracture displacement was associated with extrusion of crystalline material from bone. On histological examination crystals were found to replace metaphyseal bone. Pericrystalline giant cell granulomata replaced bone marrow. Erosion surfaces near granulomas were increased. Subperiosteal and intra-osseous tophi of calcium oxalate were seen. Calcium oxalate appears to precipitate with greater facility than does physiological mineral. Bone showed the features of mixed uraemic osteodystrophy in all four patients. We conclude that: (1) the fractures occurred through heavy crystal deposits; (2) ununited fractures and intra-osseous and subperiosteal tophi contributed to the pain; (3) spontaneous fractures are of poor prognostic significance. We recommend that unstable fractures be internally fixed.

A method for determining the total magnesium, calcium, sodium and potassium contents of human platelets.

Platelet levels of magnesium, calcium, sodium and potassium may provide a better index of intracellular cations or total body ion status than serum, erythrocyte or leukocyte cation concentrations. A method is described for the determination of the total magnesium, calcium, sodium and potassium contents of human platelets. The method involves platelet isolation, repeated washing, lysis and measurement of the ions in the lysate. Magnesium and calcium were determined by atomic absorption spectroscopy and sodium and potassium by flame photometry. Platelet total potassium, magnesium, calcium and sodium were 4.5 +/- 1.5, 2.3 +/- 0.6, 1.8 +/- 0.5, 1.5 +/- 0.6 mumol/1 x 10(8) cells, respectively (means +/- SD). The method is simple, repeatable and accurate.

Physical properties of the red blood cells in chronic renal failure.

The biophysical properties of red blood cells in three groups of chronic renal failure (CRF) patients were compared to those of a non-renal control group. A limited number of biochemical and other parameters influencing the physical characteristics of red cells were also studied. Physical characteristics varied from normal to strikingly abnormal depending on the presence of factors other than CRF. Uraemia per se, in contrast to general belief, would appear not to be the major determinant of red cell survival in CRF.