Nephrology Dialysis Transplantation

Transcription

1 Nephrol Dial Transplant ( 1997) 12: Original Article Nephrology Dialysis Transplantation An investigation of the effect of advancing uraemia, renal replacement therapy and renal transplantation on blood pressure diurnal variability C. K. T. Farmer, D. J. A. Goldsmith, J. Cox, P. Dallyn, J. C. Kingswood and P. Sharpstone Trafford Department of Renal Medicine, Royal Sussex County Hospital, Brighton, East Sussex, UK Abstract phenomenon is not modulated by successful renal Background. Ambulatory blood pressure recordings transplantation. have been shown to correlate better with target organ damage than have isolated clinic blood pressure readhypertensives; Key words: ambulatory blood pressure; dipper; anti- ings. There have been some small studies demonstrating uraemia that abnormal blood pressure diurnal rhythm is common in uraemia and in patients on renal replacement therapy. Abnormal blood pressure diurnal rhythm itself may be a risk factor for accelerated target organ damage. Introduction Methods. We retrospectively studied 480 ambulatory blood pressure recordings in 380 patients with essential Ambulatory blood pressure monitoring (ABPM ) has hypertension, secondary hypertension, and on renal become an increasingly reliable and useful tool for the replacement therapy. We examined diurnal blood presblood accurate measurement of both daytime ambulatory sure rhythm in each group. pressure and night-time (sleeping) blood pres- Results. Abnormal blood pressure diurnal rhythm sure, when in normal individuals there is a decline in (non-dipping) is significantly more prevalent in patients blood pressure. Data from population studies have with underlying renal disease, even with normal excretof blood pressure in healthy individuals [1,2]. This given different values for the normal nocturnal decline ory renal function (53%) than in age-, sex-, and racechange in blood pressure is approximately a 15% matched controls with essential hypertension (( 30%), P<0.01). In patients with renal disease the prevalence reduction in systolic and 20% reduction in diastolic of non-dipping rose with worsening renal function, blood pressure with sleep [2]. There has been shown reaching statistical significance once plasma creatinine to be good correlation between clinic blood pressure was greater than 400 mmol/l. There was a direct correlarecordings [3]; however, ambulatory blood pressure recordings and daytime ambulatory blood pressure tion between plasma creatinine and percent decline in blood pressure at night for both systolic (r=0.23) and ( ABP) over a 24-h period has been shown to have diastolic (r=0.24) blood pressure in patients with a better correlation with end-organ damage than have underlying renal disease and impaired excretory renal isolated clinic readings [4,5]. This more accurate function. High prevalences of abnormal diurnal BP correlation may be first because ABPM can identify rhythm are seen in patients on haemodialysis (82%), those individuals with white coat hypertension who peritoneal dialysis (78%), patients with plasma creatthat ABPM provides data about total 24-h blood are less likely to have target-organ damage, and second inine >600 mmol/l (75%), and in renal transplant recipients (74%). pressure load, specifically allowing identification of Conclusions. Abnormal blood pressure diurnal rhythm subjects whose sleep-related fall in BP is blunted, or ( non-dipping ) is significantly more common in secrisk factor. reversed, itself representing an additional end-organ ondary than in primary hypertension, even with normal renal function. Abnormal blood pressure diurnal Blood pressure control is an important modifiable rhythm becomes increasingly common with advancing risk factor for cardiovascular disease which is signific- uraemia. Once the plasma creatinine is greater than antly more prevalent with uraemia, as well as an 600 mmol/l the prevalence of non-dipping is the same important factor determining the rate of renal funcas that seen with renal replacement therapy. This tional decline in renal impairment [6,7]. The aim of our study was to examine the prevalence of abnormal blood pressure diurnal rhythm in a large number of Correspondence and offprint requests to: Dr C. K. T. Farmer, Royal Sussex County Hospital, Eastern Road, Brighton, East Sussex BN2 patients with secondary hypertension with varying 5BE, UK. degrees of renal impairment, in renal transplantation, 1997 European Renal Association European Dialysis and Transplant Association

2 2302 C. K. T. Farmer et al. haemodialysis, and peritoneal dialysis compared to an Results age-, sex-, and race-matched group of controls with essential hypertension and normal excretory renal The distribution of blood pressure recordings is sumfunction. marised in Table 1. There was no statistical difference in age, gender or race distribution in each group. Relatively fewer patients were available for study with plasma creatinine over 600 mmol/l as this was the Subjects and methods default threshold for commencing renal replacement therapy in our unit. This study is a retrospective analysis of a total of 480 The prevalence of abnormal sleep-related BP rhythm ambulatory blood pressure recordings in 380 patients taken in those subjects with essential hypertension was 30% over the last 8 years in our regional renal unit. Patients came (15/50), while in those patients with underlying renal from nephrology outpatient clinics, dialysis units (haemodia- disease and normal plasma creatinine (<110 mmol/l) lysis and peritoneal dialysis), and from renal transplant it was 53% (39/74), x2=11.51, DF=1, P<0.01. The clinics. For each recording subjects age, gender, plasma patients with essential hypertension received 1.4 antihycreatinine, haemoglobin, antihypertensive therapy, and pertensives (mean), those with renal disease 1.1 antihypatient group (essential hypertension, secondary hypertenpertensives (mean) P=n.s. There was no statistical sion, and renal replacement therapy) were recorded. For difference in the number of antihypertensives received patients with renal disease the underlying renal diagnosis by dippers or non-dippers. These results are illustrated was coded according to EDTA codes. All subjects completed a sleep and activity diary at the in Table 2. time of ABPM and night-time was taken as self-reported Patients with underlying renal disease and hyper- time asleep. Subjects on haemodialysis were monitored for tension were subdivided according to plasma 48 h in the interdialytic period. Mean night-time blood creatinine(cr), Cr <110 mmol/l, Cr mmol/l, pressure was calculated manually from ABPM data in Cr mmol/l, Cr mmol/l and Cr these subjects. >600 mmol/l. The prevalence of nocturnal dipping and All patients were monitored with the SpacelabA percent decline of blood pressure during sleep was ambulatory monitor (Spacelabs Medical, Redmond, USA) calculated in each group. Antihypertensive medication, every 15 min from 0700 to 2200 and every 30 min from 2200 age, gender, and race were also compared. There was to Monitors were calibrated against a mercury sphygmomanometer on five occasions at the beginning of each Table 1. Breakdown of groups according to renal replacement session, and monitoring practice was in accordance with therapy, plasma creatinine, and diagnosis (secondary or essential recent guidelines [8]. Blood pressure monitoring recorded hypertension) mean daytime, mean night-time, and overall mean blood pressure. Subjects were defined as dippers and non-dippers according the results of ambulatory blood pressure recording. Group/Plasma Number of Mean age %Males creatinine blood pressure (years)+(sd) Dippers were defined as those subjects whose mean sleeping recordings blood pressure (both systolic and diastolic) was 10% less than mean waking blood pressure. Non-dippers were those subjects whose blood pressure did not fall (by 10%) or rose Essential (14.65) 54 hypertension during sleep. <110 mmol/l (15.72) 55 The patients with secondary hypertension were divided mmol/l (16.3) 56 into five groups, creatinine <110 mmol/l, creatinine mmol/l (13.73) mmol/l, creatinine mmol/l, creatinine mmol/l (15.7) mmol/l, creatinine >600 mmol/l. The prevalence of >600 mmol/l (13.1) 62 non-dipping was calculated in each group and statistical Haemodialysis (18.0) 62 differences were calculated using the chi-squared test, statist- Peritoneal dialysis (17.6) 57 ical significance taken as P<0.05. Two further subgroups were analysed, dialysis patients ( haemodialysis and peritoneal dialysis) and renal transplantation. In subjects with secondary Transplant (15.1) 65 hypertension a second analysis was performed using plasma Table 2. Comparison between essential hypertensives and patients with primary renal disease and elevated blood pressure, both groups creatinine as a continuous variable, using the CSS Statsoft with normal excretory renal function program both linear regression analysis of 1/creatinine (mmol/l ) against per cent decline in blood pressure (systolic and diastolic). Multiple regression analysis was also perhypertension Essential Creatinine P formed using the CSS program. Fifty patients with essential hypertension, matched for <110 mmol/l age, gender and race with the cohort of patients with normal n renal function but proven underlying renal disease, under- Age (mean/sd) 50.6 (14.65) 41.9 ( 15.72) NS going ABPM in our hypertension clinic to establish degree Males/females 27/23 41/33 NS of control of BP (ie prior exclusion of white-coat effect) over Dippers <0.001 Non-dippers 15 ( 30%) 39 ( 53%) <0.001 the same period ( ), were compared to those with Antihypertensives NS normal renal function and underlying renal disease. The Daytime BP (mmhg) 143/89 144/89 NS prevalence of nocturnal dipping in each group was calculated Night-time BP (mmhg) 120/69 126/74 <0.05 and using the x2 test statistical significance was calculated.

3 BP diurnal rhythm in CRF 2303 no significant difference between age, gender and race not practiced for chronic renal failure patients with distribution between groups. plasma creatinine <400 mmol/l; and was only relevant Table 3 describes the characteristics of each study to 16% of patients with plasma creatinine >600 mmol/l. group. Figure 1 shows the prevalence of dipping in Its use was much higher in the dialysis cohorts. each group of subjects. The difference in prevalence of The effect of rising creatinine on renal function was nocturnal dipping reached statistical significance once also examined as a continuous variable. Linear regres- plasma creatinine rose above 400 mmol/l (P 0.028) sion analysis was performed on plots of percentage and was still more significant at plasma creatinine decline in blood pressure against log plasma creatinine. >600 mmol/l (P<0.002). This increase in abnormal For systolic blood pressure r=0.23, diastolic r=0.24 diurnal rhythm was present despite the increase in (P<0.001) Figures 3 and 4. numbers of antihypertensives used as plasma creatinine Multiple regression analysis was performed of increased. The mean number of antihypertensives prescribed plasma creatinine against blood pressure decline using to the patients in each group, and within each age, sex, race, haemoglobin, and daytime blood pres- group by diurnal rhythm, is shown in Figure 2. Both sure as confounding variables. This produced an values the absolute and percentage fall in blood pressure at of r2=0.54 for systolic blood pressure, r2=0.61 for night reduced with declining renal function which is diastolic blood pressure decline (P<0.001). compatible with the results from the diurnal rhythm Results from the ambulatory blood pressure recordings analysis, reaching statistical significance once plasma from those patients on the renal replacement creatinine reached 400 mmol/l (P<0.001). The mean therapy programme (haemodialysis, peritoneal dialysis haemoglobin fell with rising creatinine for this large and transplantation) were also investigated. A total of group of patients erythropoietin administration was 151 patients were monitored over the same study Table 3. Haemoglobin, distribution of males and females, and quantity of antihypertensive drugs against patient group Plasma creatinine Hb %Males Antihypertensives Daytime BP Night-time BP %Erythropoietin Essential hypertension /89 120/69 0 <100 mmol/l /89 126/ mmol/l /87 135/ mmol/l /85 140/ mmol/l /84 139/78 8 >600 mmol/l /88 138/84 16 Haemodialysis /85 141/81 64 Peritoneal dialysis /83 133/78 78 Transplantation /83 132/77 11 Fig. 1. Prevalence of nocturnal dipping of blood pressure against renal function in a population of patients with underlying renal disease.

4 2304 C. K. T. Farmer et al. Fig. 2. Antihypertensive medication prescription within each patient group. Fig. 3. Scattergram of systolic blood pressure decline at night against renal function ( log transformation). period ( Tables 1 and 3). The prevalence of dipping at night was derived using the same methods as for patients not on RRT, those patients on haemodialysis were monitored for 48 h in the interdialytic period and sleeping blood pressures were calculated manually. There was no association between dialysis treatment time and diurnal BP levels or rhythm (but there were no patients receiving long-hours haemodialysis treatment). The prevalence of non dipping in patients on renal replacement therapy was very high, even in renal

5 BP diurnal rhythm in CRF 2305 Fig. 4. Scattergram of diastolic blood pressure decline at night against plasma creatinine ( log transformation). transplant recipients (74%); use of antihypertensive has been demonstrated that there is a relationship medication declined once patients started RRT (from between circadian blood pressure rhythm, ambulatory 2.7 meds/pt/day to 1.4 meds/pt/day on RRT ). The blood pressure levels and end organ damage one prevalence of non-dipping was significantly higher in study of hypertensives suggested that the duration of transplant recipients than in those patients with exposure to increased levels of BP and wall stress over impaired renal function even when both groups were 24 h can play an important role in the pathogenesis of matched for plasma creatinine, age and haemoglobin. LVH [10]. Loss of normal blood pressure diurnal (P<0.01), x2=32.23, DF=1. All transplant recipients rhythm may render patients more susceptible to cerebwere on triple immunosuppressive therapy (predniso- rovascular accidents [11]. Two studies of patients with lone, azathioprine, and cyclosporin A). chronic renal failure have suggested that blood pressure The underlying renal diagnosis was recorded in each diurnal rhythm affected the rate of decline of renal group. No trend or statistical difference could be found function in diabetic nephropathy, those patients who between diagnosis and prevalence of nocturnal dipping. were non-dippers had a renal functional decline of There were no patients with diabetic nephropathy in 7.9 ml/min/year against 2.9 ml/min/year both groups the group with plasma creatinine above 400 mmol/l being well matched for daytime blood pressure [12], which reflects our practice or starting patients with while Timio et al. [13] showed a similar effect in diabetes on renal replacement therapy rather earlier. patients with hypertensive renal disease and declining renal function. There is therefore mounting evidence Discussion that blood-pressure rhythm has an important role in target organ damage in hypertensive individuals. The underlying explanations for non-dipping are There are no clearly defined points at which blood unclear. In patients with diabetic nephropathy ( DN) pressure on ambulatory blood pressure monitoring there is an association with autonomic neuropathy and becomes abnormal, but useful cut off points can be abnormal blood-pressure diurnal rhythm [14,15] derived from the use of 95th percentile blood pressure Latent overhydration has also been cited as a possible from normal individuals [9]. Population data from the mechanism for the lack of nocturnal fall of blood Allied Irish bank study suggest that in normal indiextracellular pressure in those with DN, following a study of vidual blood pressure declines by 10 20% at night [2]; fluid estimation and ambulatory blood it is also accepted that this diurnal change in BP is pressure [16]. normally distributed in the general population [1]. It This is only the second, and is by far the larger and

6 2306 C. K. T. Farmer et al. more comprehensive, study to specifically to examine ABPM is repeated. In studies of the general population, the issue of the change in renal function and blood a figure of 60 70% reproducibility for diurnal BP pressure diurnal variability. One much smaller study rhythm is typically found where subjects have a second has been reported where 53 male veteran hypertensives ABPM [1]. This matter is less easy to establish in with chronic renal failure underwent ABPM, no con- uraemia, and no substantive data exist, but our previous sistent relationship was seen between GFR and diurnal studies suggest that reproducibility in patients on rhythm, but so small were the numbers of patients, dialysis is at least 80% (data not shown), while one and so broad were the renal functional categories, that recent study prospectively tracked a cohort of 60 only a very large effect of GFR on diurnal rhythm haemodialysis patients with five consecutive ABPM could be expected to be discovered [17], also, patients sessions over 1 year, and found that 46/60 ( 76%) had their antihypertensive medication withdrawn four preserved their diurnal rhythm pattern on each of the weeks prior to ABPM unlike our study where preexisting five ABPM recordings [25]. medication was continued. Portaluppi s study It is likely that abnormal blood pressure diurnal [18] examined ABPM in chronic renal failure patients rhythm is a further risk factor for accelerated renal in hospital with creatinine clearances ranging from 13 decline and atherosclerosis in patients with renal to 39, and did not find any association between degree disease; systolic blood pressure control is one of the of renal failure and diurnal BP rhythm. Thus our study key determinants of survival on dialysis [26,27]. is the first to do this, and while it may lack the Accurate measurement of true 24-h BP burden is an precision of others with respect to quantifying degree essential adjunct to properly targeted antihypertensive of renal impairment ( by reliance on plasma creatinine) therapy, and as patients approach end-stage renal this is more than compensated for by the large numbers failure and enter renal replacement therapy programmes, of patients reported, and by the range of renal function the results of this study suggest that the studied. information derived from ABPM about sleep-related The patients reported here included all patients who falls in BP becomes progressively more relevant. underwent ambulatory blood pressure over the previous 8 years with underlying renal pathology. All patients had ABPM performed by one of two trained References technicians and were monitored to the same protocol 1. Mancia G, Sega R, Bravi C et al. Ambulatory blood pressure throughout the period. The exact cause for the normality: results from the Pamela Study. J Hypertens 1995; 9: increased prevalence of abnormal blood pressure diurnal rhythm in patients with underlying renal disambulatory blood pressure in men and women aged O Brien E, Murphy J, Tyndall A et al. Twenty-four hour ease and normal GFR is unclear. In those patients years, The Allied Irish Bank Study. J Hypertens 1991; 9: with impaired renal function, or on renal replacement 3. Covic A, Goldsmith DJA, Venning MC, Ackrill P. Value of therapy it is likely that this abnormal rhythm can be office and blood pressure monitoring in CRF patients. Nephrol attributed to a combination of reduced vascular com- Dial Transplant 1993; 8 (7): Devereux RB, Pickering TG. Relationship between the level, pliance [19], autonomic neuropathy [20], latent pattern and variability of ambulatory blood pressure and target overhydration, use of erythropoietin [21] and, in trans- organ damage in hypertension. J Hypertens 1991; 9 [Suppl plant recipients, pressor drugs (steroids and cyclosporin 8]: S34 38 A, [22]). Renal anaemia increased as GFR fell in the 5. Covic A, Goldsmith DJA, Georgescu G, Venning MC, Ackrill chronic renal failure cohorts; this relationship was only P. Echocardiographic findings in long-term, long hours haemo- dialysis patients. Clin Nephrol 1996; 452: slightly blunted by modest use of erythropoietin in 6. Parving H-H, Jacobsen P, Rossing K, Smidt UM, Hommel E, these patients (reflecting the long time period that this Rossing P. Benefits of long-term antihypertensive treatment on study covers, and the evolution of use of erythropoietin prognosis in diabetic nephropathy. Kidney Int 1996; 49: from dialysis to predialysis settings). We have no data Wee PM, Epstein M. Angiotensin-converting enzyme inhibitors on parathyroid status for the predialysis patients, but and progression of non-diabetic renal disease. Arch Intern Med it is likely here that renal bone disease would have 1993; 153: become more severe as renal function declined. Reports 8. White WB, Berson AS, Robbins C et al. National standard for of an association between PTH and BP are in the measurement of resting and ambulatory blood pressures with literature, but only in the context of haemodialysis automated sphygmomanometers. Hypertension 1993; 21: patients [23,24]. Yet other explanations, including 9. Staessen JA, Bieniaszewski L, O Brien ET, Fagard R. What is obstructive sleep apnoea, are likely to be relevant, but normal blood pressure on ambulatory monitoring? Nephrol Dial a comprehensive analysis has never been undertaken. Transplant 1996; 11: A frequently cited objection to this type of BP 10. Verdecchia P, Schillaci G, Guerrieri M et al. Circadian blood pressure changes and left ventricular hypertrophy in essential analysis is the arbitrary division by pre-chosen criteria hypertension. Circulation 1990; 81(2 ): between normal and abnormal diurnal BP rhythm. 11. O Brien E, Sheridan J, O Malley K. Dippers and non dippers. To reflect this sensible caveat we chose also to analyse Lancet 1988; 2: 397 the fall in sleep-related BP as a continuous variable 12. Farmer C, Cox J, Dallyn PE et al. Is diurnal BP rhythm more against renal function. This second analysis confirmed important than absolute BP level in the progression of renal failure in diabetics? Kidney Int J 1997; 2(1): (Abstract) the association seen by binary analysis. Another objec- tion is the observation that the same subject may display different sleep-related BP behaviour when 13. Timio M, Venanzi S, Lolli S et al. Non-dipper hypertensive patients and progressive renal insufficiency: a three-year longitudinal study. Clin Nephrol 1995; 43( 6):

7 BP diurnal rhythm in CRF Mann S, Altman DG, Raftery EB, Bannister R. Circadian patients with renal failure who develop erythropoietin-induced variation of blood pressure in autonomic failure. Circulation hypertension. Clin Nephrol 1995; 44( 3): ; 68(3): Dorpel A van der, Meiracker AH van der, Lameris TW et al. 15. Nielsen FS, Rossing P, Bang LE et al. On the Mechanisms of Cyclosporin A impairs the nocturnal blood pressure fall in renal blunted nocturnal decline in arterial blood pressure in IDDM transplant patients. Hypertension 1996; 28: patients with diabetic nephropathy: a 10-year prospective study. 23. Goldsmith DJA, Covic A, Venning MC, Ackrill P. Ambulatory Diabetic Med 1995; 12: blood pressure monitoring in hemodialysis, CAPD and renal 16. Mulec H, Blohme G, Kullenberg K, Nyberg G, Bjorck S. Latent transplantation. Am J Kidney Dis 1997; 29(4 ): overhydration and nocturnal hypertension in diabetic nephro- 24. Goldsmith DJA, Covic A, Venning MC, Ackrill P. Blood pathy. Diabetologia 1995; 38( 2): pressure reduction after parathyroidectomy for secondary hyper- 17. Rosansky SJ, Menachery BS, Wagner CM, Jackson K. Circadian parathyroidism further evidence implicating calcium homeostasis blood pressure versus renal function. Am J Kidney Dis 1995; in blood pressure control. Am J Kidney Dis 1996; 27(6): 26: Portaluppi F. Montanari L, Massari M, Di Chiara V, Capanna 25. Covic A, Goldsmith DJA, Mititiuc C, Georgcscu G, Covic M. M.Loss of nocturnal decline of blood pressure in hypertension Abnormal circadian BP profile, and not BP levels, are associated due to chronic renal failure. Am J Hypertens 1991; 4: with left ventricular dilatation and systolic dysfunction. Abstract. 19. London GM, Guerin AP, Marchais SJ et al. Cardiac and arterial XIVth International Congress of Nephrology, Sydney, Australia, interactions in end-stage renal disease. Kidney Int 1996; 50: May 1997; Charra B, Calemard E, Laurent G. Importance of treatment 20. Takahashi H, Matsuo S, Toriyama T, Kawahara H, Hayano time and blood pressure control in achieving long-term survival J. Autonomic dysfunction in haemodialysis patients with persist- on dialysis. Am J Nephrol 1996; 16: ent hypotension. Nephron 1996; 72: Tomita J, Kimura G, Inoue T et al. Role of systolic blood 21. Jones MA, Kingswood JC, Dallyn PE et al. Changes in diurnal pressure in determining prognosis of hemodialysed patients. Am blood pressure variation and red cell and plasma volumes in J Kidney Dis 1995; 25(3 ): Received for publication: Accepted in revised form: