C Blackwell Munksgaard 2003 Copyright
American Journal of Transplantation 2003; 4: 244–247 Blackwell Munksgaard
doi: 10.1046/j.1600-6143.2003.00326.x
Systolic Blood Pressure Diurnal Variation is Not a Predictor of Renal Target Organ Damage in Kidney Transplant Recipients Ali A. Haydar, Adrian Covic, Mohsen Agharazii, Satish Jayawardene, John Taylor and David J. A. Goldsmith∗ Renal and Transplantation Unit, Guy’s Hospital, London, UK ∗ Corresponding author: Dr David J. A. Goldsmith, MA, FRCP,
[email protected] Elevated blood pressure and diurnal blood pressure variation detected by ambulatory blood pressure monitoring has been shown to be predictive of worse outcome in end-stage renal disease patients in small studies. What has been lacking is a large study to determine whether these ambulatory blood pressure monitoring (ABPM)-derived variables are predictors of worse outcome in renal transplant recipients. All the patients that underwent renal transplantation and follow up at this institution from January 1998 till October 2002 were involved in this study (n = 177). All patients were followed up for at least 48 weeks. Last creatinine correlated positively with duration of dialysis (p = 0.035, r = 0.158), kidney-donor age (p < 0.0001, r = 0.377), early kidney function (p < 0.0001, r = 0.610, r = 0.683), 24-h systolic blood pressure (SBP) load (p = 0.002, r = 0.228), and ABPM-derived pulse pressure (p < 0.0001, r = 0.269). However neither office blood pressure nor SBP diurnal variation were predictors of kidney outcome. Regression analysis showed that early kidney function was the only independent predictor of transplant outcome (p < 0.0001). Systolic blood pressure diurnal variation, though an important predictor of target organ damage in chronic kidney disease patients, was not a predictor of renal transplant function in renal transplant recipients. Only early kidney function was an independent predictor of later serum creatinine. Key words: Kidney transplantation, renal outcome, SBP diurnal variation Received 5 August 2003, revised 1 September 2003 and accepted for publication 29 September 2003
244
Introduction Ambulatory blood pressure monitoring (ABPM) has been growing in popularity as a maneuver that can more accurately diagnose hypertension and predict outcome in hypertensive and chronic kidney disease (CKD) patients (1,2). It has been shown that ABPM-derived blood pressure variables can predict left ventricular hypertrophy (LVH), mortality and morbidity, and progression toward end-stage renal disease better than office blood pressure in CKD patients (3,4). Over and above the increased accuracy of using ABPM uniquely, it can assess sleep-related BP behaviour – in particular whether there is the normal fall in BP during sleep. Diurnal BP normally falls with sleep in normotensive and hypertensive populations and is attenuated with age (5). Abnormality in the fall in diurnal BP with sleep is seen in CKD patients (1,6,7). Abnormal diurnal BP variability has been reported as having negative end-organ consequences in CKD patients (renal failure) (3). Arterial hypertension is known to play a major role in the progression of kidney graft failure (8), as well as in the cardiovascular morbidity and mortality (9), together with other risk factors frequently present in most renal transplant patients, like diabetes, hypercholesterolemia, obesity, and LVH (10). A large study including almost 30 000 renal transplant patients, with a 7-year follow-up period, clearly showed that the graft survival rate was significantly related to both systolic and diastolic blood pressure (DBP) (8). Data in the literature studying the role of ABPM-derived BP variables [e.g. systolic blood pressure (SBP) diurnal variation] in predicting the renal outcome in kidney transplant recipients comes if at all from small longitudinal studies. In this large cohort of patients (n = 177) with a long follow up, we examined the role of ABPM-derived blood pressure variables and other possible demographics and laboratory measurements in predicting worse renal outcome in kidney transplant recipients.
Systolic BP and Transplant Outcome
Methods Study subject All the patients who underwent kidney/kidney-pancreas transplant in the renal unit of Guy’s hospital in London from January 1998 till September 2002 were included in our study (n = 177). All the patients were followed up for at least 48 weeks (Census point August 1 2003). All the patients had their ABPM when their grafts were independently functioning and at least 2 weeks following their last hospital discharge after transplantation. There were 105 males and 72 females. All patients were aged between 19 and 74 years. Data about duration/mode of dialysis, nature of transplanted kidney (cadaveric/live/kidney-pancreas), transplant-to-graft function interval (delayed defined by the need for dialysis in the first two post transplant weeks), HLA mismatch, cold ischemia time, plasma creatinine at 3 and at 6 months (early kidney function) and within 1 month of the census point (August 1 2003) after transplantation (last creatinine), office BP, and antihypertensives in use were all monitored. Cyclosprin levels were monitored as trough values for 86 patients (before March 2002) and as 2-h post dose values for the remaining 31 patients. Tacrolimus trough values were measured in 60 patients. Only five patients were not taking cyclosporin or tacrolimus. From the ABPM tracings the patients were then categorized according to their asleep/awake SBP ratio into three groups: dippers ratio ≤0.9, nondippers 0.9 < ratio ≤1, and risers ratio >1 (1,2).
ABPM protocol Ambulatory blood pressure monitors were calibrated against standard mercury sphygmomanometers, used in the other arm, five times at the beginning of each monitoring session by the same technician using the same protocol subject to regular audit over the study period 1998–2002. Spacelab® 90217 monitors and customized analytical software were used (Spacelabs Medical, Richmond, VA) throughout. Appropriate cuff sizes were chosen for each patient.
Routine ABPM was for 24 h, with readings every 15 min from 07:01 to 22:00 h and every 30 min from 22:01 to 07:00 h. Blood pressure during sleep was calculated using sleep times derived both from the results of event diaries completed by the patients and from a post-ABPM interview. Traces were rejected for analysis, and ABPM repeated, if BP values were missing for a continuous period of 2 h or more, day or night. From the blood pressure monitors, data mean BP while awake, mean BP during sleep, and mean 24 h BP and pulse rates were derived. Systolic blood pressure diurnal variation was defined as average asleep/awake SBP.
Statistical analysis All the values are displayed as mean ± SE unless stated in the text. Spearman’s rho correlation was used for correlation of nonparametric means. Kruskall-Wallis test was used to compare nonparametric means. Multiple stepwise regression analysis was used to predict independent variables affecting BP diurnal variation. p < 0.05 was considered statistically significant. SPSS V 11.0 was used for statistical analysis.
Results Patients’ characteristics Baseline patients’ characteristics are outlined in Table 1. We studied 177 patients, 105 males and 72 females. One hundred and thirty-nine patients had cadaveric and 38 had live kidney transplants. There were 26 kidney/pancreas transplant recipients and the remaining were kidney transplant recipients only. The mean age of patient was 44.2 ± 13.32 (SD; range 19–74 years). Mean duration of dialysis before transplantation in weeks was 114.6 ± 6.81 (median = 94.29, range 0–411). Mean duration of
Table 1: Patients’ characteristics Patients’ characteristics
Age (years) Dialysis duration (weeks) Donor age (years) Creatinine at 3 months (lmol/L) Creatinine at 6 months (lmol/L) Creatinine at census point (lmol/L) ABPM-transplant interval (weeks) cyclosporin trough (lg/L) tacrolimus level (lg/L) Cyclosporin peak (lg/L) Duration of follow up (weeks) Office SBP (mmHg) Office DBP (mmHg) Average 24-h SBP (mmHg) Average 24-h DBP (mmHg) Pulse pressure (mmHg) Average awake SBP (mmHg) Average awake DBP (mmHg) Average asleep SBP (mmHg) Average asleep DBP (mmHg) Asleep/awake SBP Asleep/awake DBP
n (valid)
Mean
SE mean
Median
SD
Minimum
Maximum
177 177 177 177 177 177 177 86 60 31 177 177 177 177 177 177 177 177 177 177 177 177
44.27 114.60 42.73 156.90 147.28 160.53 14.5 223.33 9.70 939.77 160.75 139.95 81.97 136.81 81.69 55.12 137.32 82.49 135.85 79.28 0.99 0.96
1.00 6.81 1.15 4.17 3.57 6.20 1.19 9.75 0.38 121.50 5.99 1.37 0.79 1.00 0.58 0.86 1.00 0.60 1.25 0.65 0.01 0.01
44.00 94.29 44.50 147.0 141.5 138.0 11 233.0 9.5 915.0 187.14 140 80 135 82 53 136 82 134 80 0.97 0.95
13.32 90.54 15.25 55.53 46.55 82.44 13.00 87.70 2.94 676.95 79.75 18.18 10.46 13.37 7.77 11.40 13.34 7.94 16.58 8.63 0.09 0.08
19 0 7 73 49 60 2 25 5 250 48 104 46 108 61 28 108 62 101 52 0.79 0.77
74 411 74 498 342 610 34 384 15 3090 287 214 118 179 102 92 174 109 192 101 1.49 1.25
SBP = systolic blood pressure, DBP = diastolic blood pressure.
American Journal of Transplantation 2003; 4: 244–247
245
Haydar et al.
follow up in weeks was 160.75 ± 5.99 (median = 187.14, range 48–287). Mean creatinine concentration at 3 months was 156.9 ± 4.17, at 6 months was 147.28 ± 3.57 and at the census point was 160.53 ± 6.2 lmol/L. Twentythree percent of the patients had delayed graft function (defined by the need for dialysis in the first 2 weeks post-transplantation). Percentage of acute graft rejections (within the first 3 months following transplantation) were as follows: 50% had no rejections, 30% had one rejection, 10% had two rejections, 5% had three rejections, and the remaining 5% had more than three acute graft rejections. Mean of count of antihypertensive medications was 1.4 ± 0.08 (median = 1, range 0–5). The majority of patients had abnormal ABPM values: 51% of patients had an average awake SBP of more than 135 mmHg and 75% had an average asleep SBP of more than 125 mmHg (1,2). Over the time course of the study 24% of the patients had a greater than 15% increase in their last creatinine over baseline. Only one patient died and five returned to dialysis dependence during the follow-up period. Correlations with last creatinine (creatinine at census point) Last creatinine correlated positively with duration of dialysis (p = 0.035, r = 0.158), kidney-donor age (p < 0.0001, r = 0.377), creatinine at 3 and 6 months after transplantation (p < 0.0001, r = 0.610, r = 0.683), average 24-h systolic blood pressure (p = 0.002, r = 0.228) and ABPM-derived pulse pressure (p < 0.0001, r = 0.269). However, last creatinine did not correlate with age, cyclosporin/tacrolimus level, HLA mismatch, nature of transplanted kidney, number of rejections, office blood pressure or average 24-h diastolic blood pressure. Last creatinine did not correlate also with systolic blood pressure diurnal variation (asleep/awake blood pressure ratio). After adjusting for confounders, only early kidney function (creatinine at 3 and at 6 months) correlated with last creatinine (p < 0.0001, r = 0.298). However after adjusting for creatinine all other correlates lost their significance. Multiple stepwise regression analysis of last creatinine After fitting a predictive model of last creatinine, only early kidney function (creatinine at 3 and 6 months) predicted independently a worse renal outcome, p < 0.0001. Addition of average 24-h SBP and ABPM-derived pulse pressure did not enhance the prediction of the model. SBP diurnal variation subgroups We divided SBP diurnal variation into three groups: Dippers (D) (n = 21, ratio ≤0.9), nondippers (ND) (n = 134, 0.9 < ratio ≤1.0), and risers (R) (n = 22, ratio > 1.0). The last creatinine means according to the groups were 161.7, 155.7, and 167.4 lmol/L, respectively. Running a Kruskall-Wallis test, we found that the three means were not statistically different (Figure 1). 246
Figure 1: Bar chart of last creatinine vs. asleep/awake systolic blood pressure.
Discussion It has been shown that hypertension is a powerful predictor of worse renal outcome in kidney transplant recipients (8). Several previous studies in renal transplant patients have shown abnormal ABPM–diurnal rhythm (6,11). This abnormal diurnal variation has been reported as having negative end-organ consequences in chronic kidney disease (CKD) patients (renal failure) (3). What has been lacking is a study to look at the implications of this abnormal diurnal BP rhythm in kidney transplant patients, as these patients differ from CKD patients in two important ways. First by usually having near-normal renal function, and second by taking many vasoactive drugs including immunosuppressive agents (1,2). So we decided to study whether diurnal BP rhythm had an impact on the renal outcome in kidney transplant recipients as well as to explore other possible predictors of renal outcome. Our data from this largest published prospective database (177 patients) showed that using univariate analyses there were six correlates of transplanted kidney function: Age of kidney donor, previous duration of dialysis, early kidney function post-transplantation (creatinine at 3 and 6 months), average 24-h SBP, and ABPM-derived pulse pressure. However, using partial correlations correcting for the confounding effect of early kidney function the four remaining positive correlates lost their correlation with last creatinine at follow up. This was not true for the correlation of last creatinine with early kidney function after correcting for dialysis duration, donor age, and ABPM-derived BP variables (p < 0.0001, r = 0.298). The same conclusion was reached after fitting a regression model where early kidney function was the only independent predictor of last creatinine (p < 0.0001). All other variables did not enhance the prediction models. The American Journal of Transplantation 2003; 4: 244–247
Systolic BP and Transplant Outcome
same finding was shown in a larger study by Hariharan et al. (12). In our large cohort we found, using BP data derived from ABPM performed in the early post-transplant period, that neither BP levels nor diurnal BP rhythm had any impact on our outcome measures over a 3-year follow-up period. Systolic blood pressure diurnal variation did not correlate with last creatinine values in this cohort even after dividing SBP diurnal variation into the three categories of dippers, nondippers, and risers (according to asleep/awake SBP ratio). Only early plasma creatinine predicted later renal outcome. Our data are important because they comprise the only large-scale longitudinal trial regarding ABPM-derived variables in predicting renal outcome in renal transplant recipients. If diurnal BP per se were to be an independent target for intervention then a different BP treatment strategy would be needed, targeting specific drugs at specific periods of the 24-h period. There exist three much smaller studies for comparison. The first was in pediatric renal transplant patients by Lingens et al., who found that decreased diurnal BP variation beyond the first year post-transplantation was always associated with a renal pathology (13). In another study, including 27 children who had been transplanted 1.5–8.4 years earlier, decreased SBP diurnal variation was found in eight patients, all of which had renal pathology:renal artery stenosis (3), chronic rejection (3), recurrent focal segmental glomerulosclerosis (FSGS) (1), and past acute rejection (1,14). The third study was by Kooman et al., where 36 renal transplant recipients with chronic transplant nephropathy were studied and a significant relation between the nightly decline in mean BP and the creatinine clearance, regardless of the time after transplantation and of immunosuppressive therapy (cyclosporin or tacrolimus), was shown (15). Finally, our study was not powered to consider predictors of cardiovascular and all-cause mortality of renal transplant recipients because of the very small number of events in this carefully selected low-cardiac risk, transplanted population. In summary, we could find no evidence that diurnal BP rhythm was associated with adverse renal transplant outcome in 177 renal transplant recipients studied prospectively. These findings give no support to the concept of targeted interventions to manipulate diurnal BP rhythm in renal transplant patients.
American Journal of Transplantation 2003; 4: 244–247
References 1. Covic AC, Haydar AA, Goldsmith DJ. Ambulatory blood pressure monitoring in hemodialysis patients: a critique and review of the literature. Semin Dial 2003. 2. Covic AC, Segall L, Goldsmith DJA. Ambulatory blood pressure monitoring in renal transplantation, should ABPM be routinely performed in renal transplant patients? Transplantation 2003, in press. 3. Peixoto AJ, White WB. Ambulatory blood pressure monitoring in chronic renal disease: technical aspects and clinical relevance. Curr Opin Nephrol Hypertens 2002; 11: 507–516. 4. White WB, Schulman P, McCabe EJ, Dey HM. Average daily blood pressure, not office blood pressure, determines cardiac function in patients with hypertension. JAMA 1989; 261: 873– 877. 5. Staessen JA, Bieniaszewski L, O’Brien E et al. Nocturnal blood pressure fall on ambulatory monitoring in a large international database. The ‘Ad Hoc’ Working Group. Hypertension 1997; 29: 30–39. 6. Goldsmith DJ, Covic AC, Venning MC, Ackrill P. Ambulatory blood pressure monitoring in renal dialysis and transplant patients. Am J Kidney Dis 1997; 29: 593–600. 7. Farmer CK, Goldsmith DJ, Quin JD et al. Progression of diabetic nephropathy – is diurnal blood pressure rhythm as important as absolute blood pressure level? Nephrol Dial Transplant 1998; 13: 635–639. 8. Opelz G, Wujciak T, Ritz E. Association of chronic kidney graft failure with recipient blood pressure. Collaborative Transplant Study. Kidney Int 1998; 53: 217–222. 9. Fernandez-Vega F, Tejada F, Baltar J, Laures A, Gomez E, Alvarez J. Ambulatory blood pressure after renal transplantation. Nephrol Dial Transplant 2001; 16 (Suppl. 1): 110–113. 10. Olyaei AJ, de Mattos AM, Bennett WM. A practical guide to the management of hypertension in renal transplant recipients. Drugs 1999; 58: 1011–1027. 11. Baumgart P, Walger P, Gerke M, Dorst KG, Vetter H, Rahn KH. Nocturnal hypertension in renal failure, haemodialysis and after renal transplantation. J Hypertens Suppl 1989; 7: S70– S71. 12. Hariharan S, McBride MA, Cherikh WS, Tolleris CB, Bresnahan BA, Johnson CP. Post-transplant renal function in the first year predicts long-term kidney transplant survival. Kidney Int 2002; 62: 311–318. 13. Lingens N, Dobos E, Lemmer B, Scharer K. Nocturnal blood pressure elevation in transplanted pediatric patients. Kidney Int Suppl 1996; 55: S175–S176. 14. Lingens N, Dobos E, Witte K et al. Twenty-four-hour ambulatory blood pressure profiles in pediatric patients after renal transplantation. Pediatr Nephrol 1997; 11: 23–26. 15. Kooman JP, Christiaans MH, Boots JM, van Der Sande FM, Leunissen KM, van Hooff JP. A comparison between office and ambulatory blood pressure measurements in renal transplant patients with chronic transplant nephropathy. Am J Kidney Dis 2001; 37: 1170–1176.
247
