Professional Documents
Culture Documents
2011 - Denicola (Jurnal)
2011 - Denicola (Jurnal)
2011 - Denicola (Jurnal)
DOI:10.5301/JN.2011.6335
Luca De Nicola 1, Silvio Borrelli 1, Paolo Chiodini 2, Division of Nephrology, Second University of Naples,
1
10%-25% of CKD patients followed in the real world of Standards for CKD care
clinical practice (14-23). Surprisingly, this picture does
not substantially change in nephrology clinics (16-20, In our clinic, each patient was seen by the same nephrolo-
23). Such a discouraging finding may either be the con- gist at all visits, the frequency of which was determined
sequence of undertreatment in the specialist setting or based on the K/DOQI guidelines and the clinical status of
evidence of resistance to antihypertensive intervention the patient. Each participating nephrologist is well versed
in the unselected patients commonly seen in the dai- in, and committed to, the goals of hypertension treat-
ly practice. This critical question has remained unan- ment recommended by the K/DOQI guidelines (10, 12). In
swered so far because cross-sectional studies do not particular, BP treatment is aimed at reaching the recom-
assess the effect of therapy intensification on achieve- mended goal (systolic BP <130 mm Hg and diastolic BP
ment of BP targets. <80 mm Hg).
We therefore designed this longitudinal study to evaluate Laboratory protocols were standardized with in-house
the relationship between BP management and BP control in analysis of blood and urinary samples, including measure-
an incident cohort of hypertensive nondialysis CKD patients ment of creatinine by modified kinetic Jaffé reaction, and
over their first year of follow-up in an academic renal clinic. proteinuria by the pyrogallol red-molybdate method. GFR
was estimated by the 4-variable Modification of Diet in
Subjects and methods Renal Disease (MDRD) Study equation. Twenty-four-hour
urine collection was obtained at each visit where protein-
This is a historical cohort study using a prospective da- uria was quantified. This urine collection was also used to
tabase that included all consecutive patients incident in evaluate the adherence to the routinely prescribed restric-
the outpatient clinic of the Department of Nephrology tion of dietary salt (<6 g NaCl/day) and protein (≤0.8 g/kg
at the Second University of Naples, Italy. The database body wt/day) by means of urinary excretion of Na (UNaV)
provided comprehensive and detailed information on de- and urea. The collection was considered inaccurate, and
mographic, clinical, laboratory, therapeutic features and repeated, if the creatinine excretion was outside of the
presence or absence of left ventricular hypertrophy (LVH) 60% to 140% range of the value calculated according to
and history of CV disease (hospitalization for coronary Dwyer and Kenler (24).
heart disease, cerebrovascular and peripheral vascular Body weight and BP were recorded at each visit. BP
disease). All patients were referred to the clinic by the measurement was performed by the nephrologist in a
general practitioners (GPs) working in the same area as quiet environment in the morning using a mercury sphyg-
our hospital. Clinical and laboratory data were collected momanometer with a cuff size of appropriate dimension
at baseline and month-12 visit. Baseline treatment was and with the patient in a sitting position after 10 minutes
that prescribed by the GP; from then on, treatment was of rest. The first and fifth Korotkoff sounds were used to
that prescribed by the nephrologist in clinic. The study define systolic and diastolic BP values, respectively; the
was approved by the institutional review board, and all mean of 3 consecutive readings taken 2 minutes apart
patients gave informed consent to use of their data. was considered for analysis.
The patients who did not achieve the BP goal were screened
Patients for secondary causes of hypertension other than CKD. Ad-
herence to the prescribed drug therapy was also checked at
We considered eligible for the study all consecutive inci- each visit by means of specific questions to the patient and
dent patients who were not on dialysis and without a kidney family members. A patient was considered poorly compliant
transplant, who were referred by a GP from June 1, 2004, when pharmacological therapy differed from prescription in
to May 31, 2007, because of diagnosis of CKD, defined as at least 2 follow-up visits. Efforts to minimize poor adher-
an estimated glomerular filtration rate (eGFR) <60 or ≥60 ence included dedicated time at each visit to remind the pa-
ml/min per 1.73 m2, plus proteinuria >0.3 g/24 hours in 2 tient and his/her family members of the high risk associated
consecutive visits with an interval ≥3 months, and who had with CKD and the benefits of compliance with therapy.
completed at least 1 year of follow-up in our clinic. Patients
were excluded if at the first visit, BP was <130/80 mm Hg Statistical analysis
without antihypertensive therapy, there was evidence of
acute kidney injury, active malignancy, advanced liver dis- Continuous variables are reported as either means ± SD or
ease or active steroid or immunosuppressive therapy use. median and interquartile (IQR) range on the basis of their
hours in only 23% of the patients by the end of the study. and diastolic BP at baseline (Tab. II). Conversely, age, sex,
Poor compliance with drug therapy was infrequent in the smoking habit, previous CV disease, presence of LVH and
not-at-goal patients (4.4%). Conversely, significant dif- distribution of underlying kidney diseases were similar in
ferences were observed in the baseline characteristics the 2 subgroups (data not shown). After adjustment for
of the patients reaching and not reaching the BP goal at potential confounders by logistic regression analysis, dia-
month-12 visit. The not-at-goal group was in fact charac- betes and systolic BP emerged as the only factors inde-
terized by greater body mass index (calculated as kg/m2: pendently associated with the lack of achievement of BP
30.2 ± 6.2 vs. 28.7 ± 4.5; p=0.04), higher prevalence of targets at month 12 (Tab. III).
diabetes (44.5% vs. 30.1%, p=0.02) and higher systolic In not-at-goal patients, the decrease in systolic/diastolic BP
TABLE I
SUBGROUP ANALYSIS OF BLOOD PRESSURE CONTROL (<130/80 mm Hg) AT REFERRAL AND LAST VISIT IN THE
RENAL CLINIC
Baseline Month 12
% (95% CI) % (95% CI)
Age
≤65 years (n=98) 12.2 (5.8-18.7) 32.7 (23.4-41.9)
>65 years (n=177) 14.7 (9.5-19.9) 34.5 (27.5-41.5)
Sex
Female (n=109) 9.2 (3.8-14.6) 30.3 (21.6-38.9)
Male (n=166) 16.9 (11.2-22.6) 36.1 (28.8-43.5)
Obesity*
No (n=167) 17.4 (11.6-23.1) 35.3 (28.1-42.6)
Yes (n=108) 8.3 (3.1-13.6) 31.5 (22.7-40.2)
Diabetes
No (n=166) 14.5 (9.1-19.8) 39.2 (31.7-46.6)
Yes (n=109) 12.8 (6.6-19.1) 25.7 (17.5-33.9)
Cardiovascular disease
No (n=161) 10.6 (5.8-15.3) 35.4 (28.0-42.8)
Yes (n=114) 18.4 (11.3-25.5) 31.6 (23.0-40.1)
CKD stage
Stage 2 (n=28) 3.6 (3.3-10.4) 35.7 (18.0-53.5)
Stage 3a (n=80) 16.3 (8.2-24.3) 35.0 (24.5-45.5)
Stage 3b (n=108) 13.0 (6.6-19.3) 34.3 (25.3-43.2)
Stage 4-5 (n=59) 16.9 (7.4-26.5) 30.5 (18.8-42.3)
TABLE II
CHANGES IN MAIN CLINICAL, LABORATORY AND THERAPEUTIC CHARACTERISTICS FROM BASAL TO MONTH-12
VISIT IN PATIENTS AT-GOAL AND NOT-AT-GOAL
At-goal Not-at-goal
(n=93) (n=182)
eGFR, ml/min per 1.73 m2 42.3 ± 15.6 42.7 ± 17.2 41.9 ± 15.5 40.7 ± 17.1
Proteinuria, g/24 hours 0.23 (0.10-0.89) 0.15 (0.04-0.52)* 0.20 (0.10-0.85) 0.17 (0.03-0.55)*
BP drugs, no. 2.3 ± 1.3 2.8 ± 1.2* 2.5 ± 1.2 3.2 ± 1.3*‡
Data are expressed as means ± SD, and as % or median (IQR). Goal defined as month-12 blood pressure <130/80 mm Hg.
ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor antagonist; BP = blood pressure; CCB =
calcium channel blockers; eGFR = GFR value by the 4-variable MDRD equation; Na = sodium.
*p<0.001, vs. baseline.
†
p<0.005, vs. at-goal.
‡
p<0.05, vs. at-goal.
§
p<0.05, vs. baseline.
Discussion
We found that exposure to nephrology care increased the
prevalence of BP goal attainment, from 14% obtained in GP
offices, to 34% registered at the end of follow-up in our renal
clinic. The observed rates of BP control were definitely bet-
Fig. 2 - Reduction of systolic (dotted columns) and diastolic ter than those described in previous cross-sectional surveys
blood pressure (white columns), and proteinuria (black col-
umns), from baseline to month-12 visit in patients at-goal and in CKD patients (14-23). In particular, the control rate ob-
not-at-goal (goal defined as month-12 blood pressure <130/80 tained in our clinic was almost threefold higher when com-
mm Hg). BP = blood pressure; Uprot = urinary protein.
pared with data from the multicenter cross-sectional survey
we recently conducted in CKD patients prevalent in Italian
renal clinics (17). The observed improvement in BP control
values was smaller compared with that in the at-goal group, also held true when examining another previous cross-sec-
but it was associated with a similar reduction of proteinu tional analysis comparing BP management in our clinic and
In not-at-goal patients, the decrease in systolic/diastolic BP that in the GP offices in our area (18). In that study, optimal
values was smaller compared with that in the at-goal group, BP control was registered in 20% of our patients and 6% of
but it was associated with a similar reduction of proteinuria patients exclusively seen by a GP.
(Fig. 2). The difference in BP response was observed in spite In the GP offices, achievement of the BP goal, even though
TABLE III
LOGISTIC REGRESSION ANALYSIS OF BASAL PREDICTORS OF NOT-AT-GOAL STATUS (MONTH-12 BLOOD PRES-
SURE ≥130/80 mm Hg)
BP = blood pressure; CVD = cardiovascular disease; eGFR = GFR value by 4-variable MDRD equation.
improved with respect to the past (18), was largely insuffi- as the main features associated with refractory hyperten-
cient. Similar results have recently been reported in a large sion in CKD. Under these conditions, BP elevations may
sample of UK patients (22), where primary care providers be more difficult to control due to severe vascular damage
failed to achieve the BP goal (<130/80 mm Hg) in stage and resulting vasoconstriction (11, 13, 30, 31).
3-5 CKD patients about 90% of the time, in spite of ac- An important finding of our study is the small but significant
cess to computerized assessment of CKD. The problems reduction in BP of 7/3 mm Hg in those individuals with BP
with management of BP in the GP setting underscore the not-at-goal when such reduction is coupled with a simulta-
need for integrated care from the early stages of CKD. As neous reduction in proteinuria (Fig. 2). Indeed, reductions
recently pointed out by Hallan and Stevens (25), high-risk in BP of similar magnitude are significant to prevent CV
groups such as patients with diabetes mellitus or hyper- events in high-risk patients like those in our cohort (11, 13,
tension and those above age 60 should have their GFR 32). More important, both CV and renal prognosis further
estimated and be tested for albuminuria in GP offices and, improve when proteinuria reduction is coupled with a de-
if correctly identified as having CKD on the basis of these crease in BP (6, 33), even in patients with low-range protei-
2 parameters, they should become the first candidates for nuria (34), as in the case of our cohort.
integrated management. The need for an early integrated In conclusion, the transition of management from primary to
approach is further supported by the observation that our tertiary nephrology care consistently decreased BP levels,
intervention consistently improved the limited and hetero- and allowed the achievement of the recommended BP goal
geneous BP control detected at referral. Indeed, we over- of less than 130/80 mm Hg in more than one third of our
came the failure of GPs in controlling BP possibly due to CKD patients. Baseline high systolic BP levels and diabetic
underestimation of the hypertension problem, as in the status emerged as the main barriers to further amelioration
case of mild CKD, female sex, absence of CV disease, of BP control. Prospective studies in larger samples of un-
or in the presence of hard-to-treat hypertension as in the selected regular CKD patients are required to verify whether
case of obesity (Tab. I). the mild but significant reductions in BP and proteinuria in
Although nephrology care markedly reduced BP levels the patients who remained above goal herald better cardio-
as compared with that in GP offices, about 66% of the renal outcomes.
patients remained with BP ≥130/80 mm Hg by the end
of nephrology follow-up in spite of the strict monitor- A portion of the study was presented and published as an abstract
ing and intensification of drug therapy. Of interest, BP (J Am Soc Nephrol. 2008;19:307A), at the 2008 Renal Week of the
control was more difficult in patients carrying greater CV American Society of Nephrology.
and renal risk, such as diabetics and those with higher
systolic BP at baseline (Tab. III). This finding stresses Financial support: This work was partially supported by an Italian
the fact that the paradoxical association between higher government grant from the Health Minister (Ricerca Sanitaria
risk and worse BP control, previously reported by cross- 2006: attività di ricerca finalizzata-articolo 12), Rome, Italy, to
sectional studies in renal and nonrenal patients followed L.D.N. in 2006.
in specialist practices (17, 26), persists after sustained
exposure to therapy intensification. Conflict of interest statement: There is no conflict of interest to
The retrospective nature of our study design precludes declare.
us from generating definitive conclusions regarding the
mechanism for why BP goal is so difficult to achieve in
high-risk patients treated in a renal clinic. Nevertheless,
we can reasonably exclude the potential role of poor com-
pliance to therapy in most patients, which constitutes a
major weakness in the management of BP in primary care Address for correspondence:
(27) but is less common in the specialist setting (28), espe- Prof. Luca De Nicola
cially when the frequency of control visits is as high as in Cattedra di Nefrologia
our cohort. Similarly, underuse of diuretics, and in particu- Dip. Gerontologia, Geriatria
Mal. Metabolismo
lar of furosemide, which are the first-choice antihyperten-
Seconda Università di Napoli
sive drugs in CKD especially in the absence of adequate Piazza Miraglia
compliance to a low salt diet (29), may also be excluded IT-80131 Napoli, Italy
(Tab. II). Hence, diabetes and severe hypertension emerge luca.denicola@unina2.it
pertensives managed in specialist practice. J Hypertens. 32. Ogden LG, Hè J, Lydick E, Whelton PK. Long-term absolute
2004;22:2387-2396. benefit of lowering blood pressure in hypertensive patients
27. Mazzaglia G, Ambrosioni E, Alacqua M, et al. Adherence to according to the JNC VI risk stratification. Hypertension.
antihypertensive medications and cardiovascular morbidity 2000;35:539-545.
among newly diagnosed hypertensive patients. Circulation. 33. Eijkelkamp WB, Zhang Z, Remuzzi G, et al. Albuminuria is
2009;120:1598-1605. a target for renoprotective therapy independent from blood
28. Garg JP, Elliott WJ, Folker A, Izhar M, Black HR; RUSH pressure in patients with type 2 diabetic nephropathy: post
Hypertension Service. Resistant hypertension revisited: a hoc analysis from the Reduction of Endpoints in NIDDM with
comparison of 2 university-based cohorts. Am J Hypertens. the Angiotensin II Antagonist Losartan (RENAAL) trial. J Am
2005;18:619-626. Soc Nephrol. 2007;18:1540-1546.
29. Ernst ME, Gordon JA. Diuretic therapy: key aspects in hyper- 34. Lea J, Greene T, Hebert L, et al. The relationship between
tension and renal disease. J Nephrol. 2010;23:487-493. magnitude of proteinuria reduction and risk of end-stage
30. Benetos A, Waeber B, Izzo J, et al. Influence of age, risk renal disease: results of the African American study
factors, and cardiovascular and renal disease on ar- of kidney disease and hypertension. Arch Intern Med.
terial stiffness: clinical applications. Am J Hypertens. 2005;165:947-953.
2002;15:1101-1108.
31. Merjanian R, Budoff M, Adler S, Berman N, Mehrotra R.
Coronary artery, aortic wall, and valvular calcification in non-
dialyzed individuals with type 2 diabetes and renal disease.
Kidney Int. 2003;64:263-271. Accepted: November 12, 2010