Content uploaded by Lorenzo Giovanni Mantovani
Author content
All content in this area was uploaded by Lorenzo Giovanni Mantovani on Jan 29, 2016
Content may be subject to copyright.
Nephrol Dial Transplant (2009) 24: 3434–3440
doi: 10.1093/ndt/gfp281
Advance Access publication 16 June 2009
Setting dialysis start at 6.0 ml/min/1.73 m2eGFR—a study on safety,
quality of life and economic impact
Lucia Di Micco1, Serena Torraca2, Andrea Pota1, Daniela Di Giuseppe1, Antonio Pisani1,
Letizia Spinelli3, Simona De Portu4, Massimo Sabbatini1, Lorenzo Mantovani4and Bruno Cianciaruso1
1Division of Nephrology, School of Medicine, University ‘Federico II’ of Naples, 2Division of Nephrology, Lauria-Maratea,
3Division of Cardiology, School of Medicine, University ‘Federico II’ of Naples and 4CIRFF-Center of Pharmacoeconomics, School
of Pharmacy, University ‘Federico II’ of Naples, Italy
Correspondence and offprint requests to: Bruno Cianciaruso; E-mail: cianciar@unina.it
Abstract
Background. End-stage renal disease care requires enor-
mous economic resources. A timely dialysis start could
reduce the costs of the renal replacement therapy (RRT).
Our aim was to measure the time to dialysis in CKD pa-
tients, with an estimated glomerular filtration rate (eGFR)
≤11.0 ml/min/1.73 m2(MDRD derived), and to evaluate
the safety, economic impact and the quality of life (QoL).
Methods. In a prospective, observational study, 70 con-
secutive CKD patients, stage 5, were screened and 30 pa-
tients were selected and followed up monthly, for 24 months
or until the start of RRT, set at an eGFR =6.0 ml/min/
1.73 m2or at the occurrence of pre-defined urgent cri-
teria. The SF-36 questionnaire to evaluate the QoL was
performed at the first and the last visit.
Results. The median time to the start of dialysis was 11.8
(25th and 75th: 5.5–17.3) months. Only seven patients ur-
gently started dialysis, after 8 months (25th and 75th: 4.8–
20). The mean monthly cost of care was €1146 ±917
per patient. The QoL was similar to that of the general
population and did not change at the last assessment.
Discussion. This is the first study evaluating the economic
impact of intensive conservative management of CKD stage
5 to postpone start of dialysis in tertiary care. This strategy
allows us to safely gain a significant amount of time free
from dialysis, with good QoL and major savings in the costs
of nation’s dialysis budget. The present results, however,
are applicable only to low comorbidity patients referred to
nephrology care and may not be generalized to all patients
starting RRT.
Keywords: CKD conservative treatment; dialysis start; haemodialysis;
pre-dialysis cost; pre-dialysis QoL
Introduction
The number of patients with chronic kidney disease (CKD)
who reach end-stage renal disease (ESRD) is increasing
worldwide. The prevalence of CKD (stages 1–4) in the USA
increased from 10% to 13% in the decade 1994–2004, with
472 000 prevalent patients treated by dialysis and trans-
plantation (ESRD) in 2006 [1]. Also in Italy, CKD affects a
significant proportion of the general population, especially
among the elderly [2]. It is estimated that more than 300
patients per million population (pmp) have a diagnosis of
CKD each year and about 804 patients pmp are on chronic
renal replacement therapy (RRT) [3].
ESRD care requires a great amount of economic and
human resources, and its growing cost derives from the
greater number of patients rather than an increase in dialysis
cost [4]. ESRD expenditure represents, in the USA, 6.7%
of the total Medicare budget [1]. In Italy, chronic dialysis
consumes 2% of the total health expenditure every year [5].
In the near future, the number of patients with ESRD is
expected to rise, mainly because of the increased longevity
of the population. It is conceivable, therefore, that in the
next decade, the RRT’s costs will become unbearable and
will eventually limit access to dialysis, as already happens
in developing countries.
In this context, the issue of an appropriate start of RRT
is of paramount importance, considering the huge amount
of resources that can be saved by postponing start of dialy-
sis. It is our opinion that all the measures that are safe and
that may delay initiation of dialysis whilst ensuring that the
patients remain healthy and are not at any increased risk
should be taken. The optimum timing of dialysis treatment,
however, is still a matter of debate, and is mainly based on
clinical judgment and clinical experience. K/DOQI guide-
lines suggest that the estimation of glomerular filtration
rate (eGFR) in conjunction with the evaluation of the nutri-
tional and clinical status of the patient should guide deci-
sion making regarding dialysis initiation [6]. A prospective,
multicentre, randomized, controlled trial has been designed
to compare several outcomes in patients starting dialysis
therapy with a Cockcroft–Gault GFR of 10.0–14.0 ver-
sus 5.0–7.0 ml/min/1.73 m2and the results are awaited
[7].
CThe Author 2009. Published by Oxford University Press [on behalf of ERA-EDTA]. All rights reserved.
For Permissions, please e-mail: journals.permissions@oxfordjournals.org
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
Dialysis start at 6 ml/min/1.73 m2eGFR 3435
The aim of this study was to estimate the time to dialy-
sis in stable CKD patients, stage 5, followed up in tertiary
care. The patients satisfying the inclusion criteria for low
comorbidity entered the study at an eGFR ≤11.0 ml/
min/1.73 m2and were followed up until the start of RRT,
that was set at an eGFR =6.0 ml/min/1.73 m2or at the oc-
currence of pre-specified events that were representative of
uraemic complications and reflected the eventually declin-
ing health of the patients. The safety, the economic impact
and the quality of life (QoL) of patients were evaluated.
Subjects and methods
The present study had a prospective, observational design and was con-
ducted in the CKD out-patients clinic of the Nephrology Division at
the University ‘Federico II’ of Naples, Italy. The recruitment time was
18 months, starting from 1 January 2005, and patients were followed up
from 1 January 2005 to 1 August 2007. Consecutive patients with a cre-
atinine clearance <15.0 ml/min/1.73 m2were considered for the study.
Seventy patients were screened and 30 of them, 18 males and 12 females,
were enrolled in the study and followed up for a maximum of 24 months
(range: 2.5–24 months). The primary outcome of the study was the time
(in months) to RRT. The study was approved by the local medical ethics
committee.
The inclusion criteria were the following: a written informed consent,
age >18 years, eGFR ≤11.0 ml/min/1.73 m2. The exclusion criteria were
the inability to perform a 24-h urinary collection, malignancies, pregnancy,
breast feeding, systemic diseases treated with steroids or immunosuppres-
sive drugs, non-steroidal anti-inflammatory drugs given more than once
in a week, dilatative cardiomyopathy, congestive heart failure (NYHA
III–IV), type I and II diabetes mellitus and proteinuria >5g/24h.
GFR was estimated either with the MDRD-derived equation, including
demographic and serum variables [8], or with the arithmetic mean of
the 24-h urea and creatinine clearances, corrected for body surface area
[9]. MDRD eGFR has several limitations, including the variability of
non-standardized serum creatinine dosage; however, since it is easy to
perform and spread worldwide, we adopted this method for the purpose
of the study. The two GFR measurements (Table 1) were correlated in our
data series (Pearson’s coefficient =0.72, P<0.001). The values of eGFR
were confirmed twice, either before entering the study (during the run-in
period) or before starting dialysis. A variation of 10% above or below the
initial data was allowed. The two levels of eGFR (11.0 ml/min/1.73 m2and
6.0 ml/min/1.73 m2) were, respectively, chosen for entering the study and
for starting RRT, and were derived from the published guidelines [6,10].
Forty patients were excludedat the screening (Table 1) for the following
reasons: 5 for inability to perform a reliable 24-h urinary collection, 9 for an
eGFR ≤6 ml/min/1.73 m2, 8 for diabetes, 1 for malignancy, 5 for systemic
diseases treated with immunosuppressive drugs, 1 for congestive heart
failure, 1 for proteinuria >5 g/24 h, 2 died before completing the screening,
1 needed urgent HD treatment for uaremic haemolytic syndrome, 4 refused
the study and 3 were not compliant to the run-in scheduled visits.
The included patients were extensively informed of the dialysis modali-
ties [peritoneal dialysis (PD) and haemodialysis (HD)], and those choosing
HD were encouraged to undergo surgery for the vascular access, if not al-
ready present. After the enrolment, the patients were prescribed a low
protein diet [0.6 g/kg ideal body weight (IBW)/day] [11,12]. The details
of the diet have been previously reported [13] and included the prescrip-
tion of aproteic special products; three patients did not use such products.
Dietary protein intake (DPI) was estimated in stable, non-catabolic con-
dition, according to the Maroni formula [14]. Protein-caloric malnutrition
was defined as the occurrence of one of the following: loss of body
weight >5% in 1 month or 10% in 3 months or BMI <20 kg/m2with
serum albumin <3.2 g/dl and normal values of C-reactive protein (CRP)
<0.5 mg/dl. During 1-month run-in period, eGFR was confirmed, patients
were trained to comply with the diet and the achievement of therapeutic
targets was applied. Verif ication of the dietary adherence was accom-
plished by an expert renal dietitian every other month. After 1 month, the
patients underwent the first clinical visit, referred to as ‘baseline visit’.
Demographic and clinical data were obtained at baseline, for the included
patients, and at screening for those who were excluded.Primar y kidneydis-
eases were classified according to the European Renal Association codes.
Comorbidity was assessed using the Charlson comorbidity index score
[15]. At baseline and in each month, we performed a complete clinical
evaluation, and recorded the therapy, cardiovascular (CV) events and hos-
pitalizations. The following blood chemistries were measured monthly, in
the central laboratory of our institution: urea nitrogen, creatinine, bicarbon-
ate, sodium, potassium, phosphate, calcium, intact parathyroid hormone
(iPTH), total proteins, albumin (Bromo Cresol Purple method; normal
range: 3.5–4.8 g/dl), total cholesterol, HDL cholesterol, LDL choles-
terol, triglycerides, uric acid, haemoglobin, total iron binding capacity
(TIBC), iron, CRP, f ibrinogen and homocysteine. Urea nitrogen, creati-
nine, sodium, potassium, phosphate, and proteins were measured in the
24-h urinary collection. The patients were instructed about the study pro-
cedures and how to obtain a careful urinary collection. Standard laboratory
procedures were used for blood and urinary measurements. At the first
and last check-up, bioelectrical impedance analysis (B.I.A.) and SF-36
questionnaire were performed [16]. Chest x-ray, ECG and echocardiog-
raphy were performed every 6 months or before reaching the end point.
Echocardiography was done by the same operator using an ultrasound sys-
tem (Sonos 2000, Hewlett-Packard, Andover, MA, USA) equipped with
a 2.5-MHz transducer. Left ventricular mass (LVM) was calculated by
the Penn Convention formula [17]. The LVM index was obtained dividing
LVM by the body surface area. The SF-36 questionnaire, recently validated
in Italy [18], assesses eight dimensions of health-related QoL (HRQoL)
that refer to the physical and mental components of health perception.
Physical functioning (PF), role-physical (RP) and bodily pain (BP) are
more related to the physical component; social functioning (SF), role-
emotional (RE) and mental health (MH) are more related to the mental
component, and energy/vitality (VT) and general health (GH) are related
to both components [16]. These global measures were estimated using
standard algorithms [19].
A cost-of-treatment analysis was performed. Direct medical costs, fi-
nanced by the National Health Service (NHS), which in Italy provides
health care services to patients with advanced CKD, were calculated by
multiplying resources absorbed by their unit cost. They included the cost of
therapies, hospitalizations, laboratory, diagnostic examinations, surgery,
rehabilitation procedures, physicians’ visits and any other possible cost,
including the aproteic foodstuff. The cost of medical services [20] was
estimated using the Italian NHS tariffs, hospital costs were computed ap-
plying Diagnosis Related Group [21] and charges and drug therapies were
quantified using market prices reported by the Italian National Therapeu-
tic Formulary (Italian Directory of Medicines and Manufacturers 2008).
The cost of aproteic products (Plasmon Dietetici Alimentari S.r.l. Milano,
Italy) was calculated on the basis of the reimbursement that the local gov-
ernment (Regione Campania) recognized to the seller, at the time of the
study period. All costs were expressed in Euro format and were computed
as Euro per patient per month.
We used pharmacological and non-pharmacological therapies to
achieve the therapeutic targets that were in keeping with the cur-
rent clinical practice guidelines. The target blood pressure (BPr) was
<130/80 mmHg [22]. All classes of antihypertensive drugs were used. The
range of PTH levels was between 150 and 300 pg/ml; hyperparathyroidism
was treated by calcitriol (1.25 dihydroxycholecalciferol); the serum cal-
cium level ranged between 8.4 and 9.5 mg/dl and the serum phosphate
level between 3.5 and 5.5 mg/dl [23]. Hyperphosphataemia was treated
by calcium carbonate (0.2–0.5 g/meal up to a maximum of 2 g/day) or
sevelamer (3.2–10 g/day) if needed. The patients with low serum cal-
cium levels (<8.4 mg/dl) were treated by calcium supplements and/or
calcitriol. Serum bicarbonate levels were aimed at 22 mEq/l or above
[23], and oral supplements of sodium bicarbonate were prescribed [24].
Lipid levels were treated by dietary modification and/or medications to
achieve a target non-HDL cholesterol (total cholesterol minus HDL) <130
mg/dl, LDL cholesterol <100 mg/dl and triglycerides <200 mg/dl [25].
The haemoglobin target was between 11.0 and 12.0 g/dl for both men and
women according to the indications of K/DOQI [26].
The patients ended the study if they registered and confirmed an
eGFR =6.0 ml/min/1.73 m2or reached 24 months of follow-up or met
one of the following criteria for RRT: (1) wKt/V urea <2 associated with a
normalized protein nitrogen appearance (nPNA) <0.6 g/kg/day or reduc-
tion of body weight >5%/month or >10% in 3 months; (2) serum albumin
<3.3 g/dl; (3) serum urea ≥2.5 g/l (with wKt/V urea >2); (4) dyspnoea
on exertion with Hb >10.5 g/dl; (5) oedema resistant to diuretics (in the
absence of therapy with calcium channel blockers and with a pathologic
echocardiography); (6) serum potassium >6.5 mEq/l resistant to therapies;
(7) pericardial effusion and/or atrial or ventricular dilatation; (8) persistent
BPr values >180/110 mmHg [either systolic BPr (SBPr) >180 mmHg or
diastolic BPr (DBPr) >110 mmHg] resistant to antihypertensive therapy
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
3436 L. Di Micco et al.
Tab l e 1 . Clinical characteristics of the screened populationa
Included patientsbExcluded patientsc,d
N◦30 40
Age (years) 55.7 ±13.4 61.0 ±14.7
Gender (M/F) 18/12 19/21
Cause of nephropathy
Primary glomerulonephritis 4 (13) 8 (20)
Interstitial nephropathies 9 (30) 15 (37)
Multisystem disease 6 (20) 5 (12)
Diabetic nephropathy – 5 (12)
Not known and Other 11 (37) 6 (15)g
Urea wKt/V 1.76 ±0.40 1.57 ±0.46
eGFR, ml/min/1.73 m2e(range) 9.8 ±1.3 (7.0–11.0) 9.3 ±2.1 (5.0–13.8)
GFR, ml/min/1.73 m2f (range) 8.9 ±2.3 (5–15) 8.5 ±2.7 (2.5–13.2)
Smoking status
Current/former 4/8 (13/27) 14/10 (35/25)
Not smoker 18 (60) 11 (27)
Unknown 0 5 (12)
Charlson comorbidity index (range) 3.97 ±1.22 (2–7) 5.2 ±2.0 (2–9)g
Cardiovascular history
Hypertension 28 (93) 34 (85)
Dislipidaemia 2 (7) 5 (12)
arrhythmia 1 (3) 3 (7)
myocardial infarction 1 (3) 3 (7)
cerebral-vascular accident 1 (3) 3 (7)
coronary angioplasty 0 2 (5)
peripheral vascular disease 0 6 (15)g
aData are expressed as number (N◦), (%), or mean ±SD.
bData collected at baseline.
cData collected at screening.
dCauses of exclusion explained in the text.
eGFR estimated with MDRD equation [8].
fGFR estimated with creatinine +urea clearance/2 [9].
gP<0.05 vs included patients.
and (9) uraemic symptoms (anorexia, nausea and vomiting in the presence
of metabolic acidosis treated with sodium bicarbonate). Death and kidney
transplant were considered end-points of the study.
Statistical methods
All the laboratory and clinical data were checked for normal distribu-
tion (Shapiro–Wilk test), and mean ±SD or median with 25th and 75th
percentiles were used if data had normal or skewed distribution, respec-
tively. The paired samples t-test, Fisher’s exact test, Wilcoxon signed-rank
test and McNemar’s test were used as appropriate. The Kaplan–Meier
method was chosen to represent the time to the end of the follow-up, due
to event occurring (starting dialysis or death) or the reaching of the 24th
month. Analyses were considered significant at P-value <0.05. Analy-
ses were performed using STATA 9.2 SE (Stata Corp., College Station,
TX, USA).
Results
Table 1 shows the clinical characteristics of the screened
population. The patients who matched the inclusion crite-
ria, as compared to those who were excluded, were slightly
younger, had similar renal diseases, except for diabetes
and unknown causes, and comparable mean eGFR and urea
Kt/V. The included subjects had a significantly lower Charl-
son’s comorbidity index, with an average score of 3.9 ±
1.22 as compared to 5.2 ±2.0 (P<0.05) and a higher fre-
quency of peripheral vascular diseases (P<0.05). The renal
survival of the studied patients is shown in Figure 1. The
median time to dialysis start was 11.8 months (5.4–17.3).
Eight patients (27%) ended the study without starting dialy-
0.00
0.25
0.50
0.75
1.00
renal survival (%)
0 6 12 18 24
(25)
(16)
(10)
(8)
Months of follow-up
Fig. 1. Kaplan–Meier for renal survival. In parenthesis: number of patients
remaining in the follow-up. −−−, median renal survival.
sis after 21.8 months (16.6–24.5); 14 patients (47%) started
dialysis because their eGFR reached 6.0 ml/min/1.73 m2
after 8.8 months (4.9–11.8); 7 subjects (23%) started dial-
ysis due to the occurrence of one of the pre-defined crite-
ria after 8 months (4.8–20.0) and with an eGFR =7.9 ±
1.3 ml/min/1.73 m2; 1 patient withdrew his informed con-
sent after 8 months for difficulty in attending the monthly
visits. None of the patients died during the follow-up, and all
of them were still alive 12 months after the end of the follow-
up period (31 August 2008). Table 2 shows initial and final
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
Dialysis start at 6 ml/min/1.73 m2eGFR 3437
Tab l e 2 . Nutritional and biochemical parameters at baseline and at the end of follow-upa
Initial Final
N◦30 30
Weight (kg)
Males 72 ±871±8
Females 61 ±17 62 ±16
BMI (kg/m2) 24.4 (22.0–28.0) 24.7 (22.8–27.0)
Body fat (%) 21.6 ±9.9 21.8 ±9.8
FFM (kg) 52.5 ±10.2 52.9 ±10.4
BCM (kg) 29.2 ±6.6 29.3 ±7.4
Serum urea nitrogen (mg/dl) 62.0 ±18.0 74.0 ±24.0b
Protein intake (g/kg IBW/day) 1.0 (0.97–1.08) 0.67 (0.60–0.90)b
Protein intake (g/day) 47.3 ±14.3 48.1 ±15.3
C-reactive protein (mg/dl) 0.31 (0.31–0.31) 0.31(0.31–0.31)
Serum albumin (g/dl) 4.24 ±0.42 4.15 ±0.49
Total iron binding capacity (mcg/dl) 248 ±54 260 ±36
Serum bicarbonate (mEq/l) 24.1 ±3.3 23.3 ±3.8
Urinary creatinine (mg/24 h) 983 ±325 1111 ±355
Urinary urea nitrogen (g/24 h) 5.2 (4.1–7.0) 5.2 (4.0–7.0)
Urinary sodium (mEq/24 h) 134 ±58 139 ±52
Initial: data collected at baseline; Final: data collected at the last visit of follow-up; N◦: number of patients.
aData are expressed as mean ±SD for normally distributed variables and as median (25th–75th percentiles) for variables with skewed distribution.
bP<0.05 versus initial.
BMI, body mass index; FFM, free fat mass at B.I.A.; BCM, body cellular mass at B.I.A.; IBW, ideal body weight.
Tab l e 3 . Clinical and metabolic parameters with achievement of K/DOQI targets at baseline and at the end of follow-upa
Initial Final
N◦30 30
Systolic blood pressure (mmHg) 132 ±17 (57) 130 ±17 (63)
Diastolic blood pressure (mmHg) 79 ±9 (77) 76 ±10 (67)
Transferrin saturation (%) 30.5 (21–48) (80) 28 (17–35) (73)
Haemoglobin (g/dl) 11.4 ±1.0 (83) 11.1 ±1.1 (77)
Ferritin (ng/dl) 110 (67–189) 113 (49–174)
iPTH (pg/ml) 213.5 (160–316) (73) 302 (261–426)b(60)
Serum calcium (mg/dl) 9.6 ±0.6 (93) 9.4 ±0.6 (97)
Serum phosphate (mg/dl) 4.5 ±0.7(93) 5.0 ±0.9b(80)
Calcium ×phosphate (mg2/dl2)43±6 (97) 47 ±8 (93)
Non-HDL cholesterol (mg/dl) 116 ±32.9 (50) 117 ±32.6 (77)
HDL cholesterol (mg/dl) 53 ±14 51 ±16
LDL cholesterol (mg/dl) 101 ±33 (40) 86 ±25b(67)b
Triglycerides (mg/dl) 115 ±42 (90) 116 ±41 (100)b
Ventricular mass index (g/m2)
Male (n.v. <130) 107.3 ±15.8 (94) 112.2 ±16.6 (88)
Female (n.v. <110) 122.6 ±20.9 (25) 117.6 ±10.4 (33)
Initial: data collected at the baseline; Final: data collected at the last visit of follow up; N◦: number of patients.
aData are expressed as mean ±SD for normally distributed variables and as median (25th–75th percentiles) for variables with skewed distribution
(percentage of patients achieving K/DOQI targets).
bP<0.05 versus initial.
n.v.: normal values.
nutritional data. These indexes were generally within nor-
mal limits in all patients at the end of the study and similar
to the initial values, such as the mean body weight, serum
albumin, bicarbonate levels, TIBC, CRP, urinary urea ni-
trogen and urinary sodium. There was a significant differ-
ence between initial and final values for median protein
intake that decreased from 1.0 g/kg IBW/day (0.97–1.08)
to 0.67 (0.60–0.90) (P<0.05) and for serum urea nitro-
gen that increased on the average of 15 mg/dl. Nineteen
patients had an average protein intake between 0.50 and
0.80 g/kg IBW/day, and the remaining 11 patients had an
intake between 0.81 and 1.50 g/kg IBW/day. None of the
patients registered a reduction of albumin or a weight loss
that reached the pre-defined malnutrition criteria. Table 3
shows the initial and final clinical and metabolic parameters
with the LVM data. At baseline, 57% and 77% of patients
achieved targets for SBPr and DBPr, respectively. At the
end of the follow-up, 63% and 67% of patients achieved the
same targets. The LVM index did not significantly increase
during the observation period. Among males, 1/18 patients
at the start and 2/18 at the end had a LVM >130 g/m2.
Among females, 9/12 at the start and 8/12 at the end of the
follow-up had a LVM >110 g/m2. Fifty percent of patients
were on target for non-HDL cholesterol at baseline and
77% at the end, with 40% of patients achieving LDL target
at the start and 67% at the end of the study (P<0.05);
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
3438 L. Di Micco et al.
Tab l e 4 . CKD conservative treatment costs expressed in Euro 2008 per
patient-montha
Therapy 300.4 ±529.8 (26.2%)
E.S.A. 223.6 ±531.7
Hospitalizations 484.2 ±589.2 (42.3%)
Diagnostics and laboratory exams 290.9 ±130.5 (25.4%)
Low protein food 70.1 ±37.4 (6.1%)
Total direct cost 1145.7 ±917.5
aData are expressed as mean ±SD (percentage of total cost).
ESA, erythropoiesis-stimulating agents.
90% were on target at baseline for triglycerides, and
all the patients achieved the target at the end of the
follow-up (P<0.05). Eighty-three percent of patients had
haemoglobin levels >11 g/dl at baseline and 77% at the
end of the study; among these patients, 7 (28%) at baseline
and 14 (61%) at the end were treated with erythropoiesis-
stimulating agents (ESA). Eighty percent of patients at
baseline and 73% at the end of study had a percentage
of transferrin saturation >20%. Seventy-three percent of
patients at baseline and 60% at the end of follow-up were
on target for iPTH levels (<300 pg/ml); seven patients at the
end of follow-up had an iPTH between 450 and 900 pg/ml,
two of whom had a baseline iPTH >300 pg/ml; 93% of
patients at baseline and 97% at the end of the study were on
target for serum calcium levels; 93% at baseline and 80%
at the end of the study achieved serum phosphate target,
with final levels significantly higher than initial values. At
baseline, 27% of patients achieved all the targets for PTH,
calcium, phosphorus and calcium–phosphorus product at
the same time. At the end of the study, 17% of patients
achieved the same goal.
The average monthly cost of care was €1146 ±917
per patient, mainly because of hospitalization that repre-
sented 42% of the expenses (Table 4). During the study, a
total of 27 hospitalizations occurred (three patients needed
2 hospitalizations each), with a median length of 5 days
(2.0–9.0). In 22 cases, the main reason for hospitalization
was vascular access surgery and/or the first 23 dialysis ses-
sions; in one patient a catheter for PD was inserted; two
patients needed a central venous catheter for urgent start
of dialysis after 5 and 8 months of follow-up, respectively.
They also underwent surgery for the artero-venous fistula
that they had previously refused; five patients were hospi-
talized for other causes (hyponatraemia, hypotension and
cranial trauma, hydrocephalus, worsening of renal function
and pulmonary pre-oedema).
SF-36 was administered at baseline and at the end of the
study, evaluating eight dimensions of QoL. All the compo-
nents were similar at the start as well as at the end of the
follow-up, thus SF-36 results showed QoL stable during the
observation period and quite similar to that of the general
population [18] (Figure 2).
Discussion
A wide range of GFR values are used in clinical practice to
support dialysis, and this interval may correspond to sev-
eral months, if not years, free of RRT. The timing of dialysis
Fig. 2. CKD patients at baseline, CKD patients at
the last visit, General population (18), Hemodial-
ysis patients (35); PF, Physical Functioning; RP, Role-Physical; BP, Bodily
Pain; GH, General Health; VT, energy/Vitality; SF, Social Functioning;
RE, Role-Emotional; MH, Mental Health.
start represents a complex clinical decision, that beyond the
estimate of GFR, has to take into account several compo-
nents, such as the patient’s well-being and QoL, the risks
related to the delaying of the therapy and the patients’ expo-
sure to the unpleasant experience of dialysis. A substantial
costs reduction could be attainable by starting dialysis at an
eGFR of 6.0 ml/min/1.73 m2; nevertheless, it is fundamen-
tal to guarantee an optimal metabolic and clinical status to
the patients waiting a timely start of RRT.
ThemediantimetodialysisfromaneGFR≤11.0 ml/
min/1.73 m2was 11.8 months. During this time, the ma-
jority of patients did not present relevant clinical problems
or any other conditions that are traditionally considered as
indicators for the initiation of dialysis. Only 23% [7] of sub-
jects needed to start RRT because of the occurrence of at
least one of the pre-defined criteria. Comparing these data
with those obtained from the excluded patients, it is evident
that the higher comorbidity was the major determinant for
the start of dialysis. In fact, the median time to dialysis of
this group of patients was of 4.0 months (2.0–6.0), with a
mean eGFR of 7.0 ±2.4 ml/min/1.73 m2. At 31 August
2008 (12 months after the end of the follow-up), 31 patients
were alive after 37.5 months (28.0–40.0) of dialysis, 3 of
them received a kidney transplant and 1 did not start RRT.
There were no signs of malnutrition in any of the pa-
tients studied, even in those who started dialysis because
of uraemic symptoms, and none of the patients complained
anorexia or loss of appetite. Moreover, albumin levels were
always >3.5 g/dl. Only two patients had albumin levels be-
tween 3.2 and 3.5 g/dl during the observation; both had a
proteinuria of 4–5 g/24 h, without signs of inflammation.
The majority (63%) of patients studied were able to com-
ply with the dietary recommendations of KDOQI guidelines
[11] that suggest a protein content of the diet between 0.6
and 0.8 g/kg IBW/day. The compliance with such a diet is
not hard to obtain, especially with the help of a renal dieti-
tian, and can be reached even without the use of aproteic
special products. A low protein diet, even if does not play a
major role in retarding the progression of renal diseases, is
able to guarantee a good metabolic control of renal failure
[13] and, therefore, may be useful in delaying the start of
RRT.
Hypertension control is one of the main aims of delaying
the progression of renal disease as well as reducing cardio-
vascular risk of CKD. We obtained a good control for the
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
Dialysis start at 6 ml/min/1.73 m2eGFR 3439
SBPr (∼60% of patients) as well as for the DBPr (∼70%
of patients). A high prevalence of hypertension in CKD as
well as in dialysis patients has been described. The TABLE
study, that focuses on the prevalence of CV determinant
risks in pre-dialysis patients in Italy, described a poor hy-
pertension control with only 12% of patients, CKD stage
5, reaching BPr <130/80 mmHg [27]. In HD patients, a
prevalence of hypertension of 70–80% and a proportion
of adequately treated patients of 30% have been described
[28]. A recent retrospective study showed that hyperten-
sion in CKD patients, before starting HD, is a major risk
factor for CV mortality during HD [29], emphasizing the
importance of an early and adequate control of BPr in this
population. This strategy is important also for the preven-
tion of left ventricular hypertrophy (LVH) that has been
found in 65% of CKD patients [27] and in 70–80% of the
patients starting HD [30]. LVM is an independent predictor
of CV disease and a mortality risk factor in the general
population as well as in dialysis patients [31]. In our study,
LVM did not significantly increase during the follow-up;
94% of male patients at the start and 88% at the end of the
observation had a normal LVM value, while the majority
of females (∼70%) were hypertrophic from the beginning.
A sex dimorphism in the prevalence of LVH has been sug-
gested in the general population, and also in dialysis pa-
tients [32]. Another modifiable CV risk factor in CKD is
anaemia. Eighty percent of our patients had haemoglobin
levels >11 g/dl at the end of the follow-up, while the preva-
lence of anaemia has been reported to be of 50% in CKD
patients, stage 5, treated with ESA [27]. Concerning min-
eral metabolism, only a small minority of patients meet all
guidelines for phosphorus, calcium and PTH [33]. In our
study, only seven patients had iPTH levels >450 pg/ml at
the end of the observation. From the clinical point of view,
our patients started RRT and/or ended the study in quite
optimal clinical conditions, even though some of these pa-
tients did not have a nephrology follow-up prior to the entry
of the study.
To our knowledge, this is the first study, in Italy, evalu-
ating the economic impact of intensive conservative man-
agement of CKD stage 5, to postpone start of dialysis. This
treatment strategy, with an average monthly cost of care of
∼1146 Euro per patient, seems to be less costly than dial-
ysis treatment, which amounts to more than 3000 Euro per
patient per month [34]. Dialysis, in fact, exposes patients to
several diagnostic and therapeutic procedures that represent
an additional burden and also an additional cost for the pa-
tients and the National Health Service, thereby influencing
lifestyle and possibly affecting HRQoL. Considering also
indirect costs, delaying HD will reduce the burden of mov-
ing frequently to and from the hospital, as well as the time
consumed during the dialysis treatment. The present data
provide evidence that conservative strategy has a positive
effect on HRQoL, which remained stable during the follow-
up with no statistical differences in any of the domains after
2 years of the treatment period. The QoL of these patients,
in fact, is comparable to that of the general population [18]
and better than the dialysis population [35]. In general, the
patients studied seem to perceive a better health state than
patients on RRT, especially in the BP domain. Dialysis, as
an invasive intervention, in fact reduces individual physical
performance, capability to work and social integration and
it often requires family engagement.
It is important to underscore that this is an observational
study, with all the limitations of that study design; fur-
thermore, the present results are applicable to nondiabetic
low comorbidity patients and may not be generalized to
all patients starting RRT, and even with this limitation,
some patients need to anticipate dialysis start. Our data,
however, are in agreement with recent findings showing
that, in a large cohort of CKD patients followed up in
a nephrology care unit, 37% of patients with an eGFR
<15.0 ml/min/1.73 m2after 2 years of follow-up did not
reach the outcomes of either dialysis therapy or death [36].
These results, taken together, are of great relevance in plan-
ning health care resources and for the allocation of funding
for CKD programmes.
In conclusion, our data show that for a consistent num-
ber of patients waiting to initiate dialysis, under nephrology
care, it is safe to use 6.0 ml/min/1.73 m2eGFR as a dialysis
start. This means that patients, within CKD stage 5, may
gain years free from dialysis, with good QoL and major sav-
ings for the costs of the nation’s dialysis budget. However,
patients need a close follow-up, with monthly checks, and
it is low comorbidity patients who benefit the most from
this therapeutic strategy.
Acknowledgements. This study was partially funded by a grant from
University of Naples “Federico II”. The results of the present study have
not been previously published. Part of this study has been presented at
the following meeting: Italian Society of Nephrology, Bari 2007. The
authors wish to thank Dr. Attilio Di Benedetto, Giulia Pacchiano, Roberta
D’Amato, from NephroCare-Italia, for the recruitment of three patients.
They also thank for the excellent work done: Dr. Patrizia Lombardi as
renal dietitian, the head nurse Wanda Marchese and patients who made
this trial possible.
Conflict of interest statement. The authors declare that no involvements that
might raise the question of bias in the work reported or in the conclusions
implications or opinion stated.
References
1. US Renal Data Systems, USRDS 2006 Annual Data Report:Atlas
of End-Stage Renal Disease in the United States. National Institutes
of Health, National Institute of Diabetes and Digestive and Kidney
Diseases, Bethesda, MD, 2007
2. Minutolo R, De Nicola L, Mazzaglia G et al. Detection and aware-
ness of moderate to advanced CKD by primary care practitioners: a
cross-sectional study from Italy. AmJKidneyDis2008; 52: 444–
453
3. www.sin-ridt.org/sin-ridt/sin-ridt.org.htm
4. US Renal Data System, USRDS 2001 Annual Data Report:Atlas
of End-Stage Renal Disease in the United States. National Institutes
of Health, National Institute of Diabetes and Digestive and Kidney
Diseases, Bethesda, MD, 2001
5. Rombol`
a G. Dialysis for everybody? At any cost? J Nephrol 2002;
15(Suppl 6): S33–S42
6. Hemodialysis Adequacy 2006 Work Group. Clinical practice guide-
lines for hemodialysis adequacy, update 2006. AmJKidneyDis2006;
48(Suppl 1): S2–S90
7. Cooper BA, Branley P, Bulfone L et al. The initiating dialysis early
and late (IDEAL) study: study rationale and design. PeritDialInt
2004; 24: 176–181
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
3440 L. Di Micco et al.
8. Levey AS, Bosch JP, Lewis JB et al. Modification of Diet in Renal
Disease Study Group. A more accurate method to estimate glomerular
filtration rate from serum creatinine: a new prediction equation. Ann
Int Med 1999; 130: 461–470
9. Bauer JH, Brooks CS, Burch RN. Renal function studies in man with
advanced renal insufficiency. Am J Kidney Dis 1982; 2: 30–35
10. European Best Practice Guidelines for Hemodialysis. Section I: Mea-
surement of renal function, when to refer and when to start dialysis.
I.3: when to start dialysis. Nephrol Dial Transpl 2002; 17(Suppl 7):
10–11
11. Kopple JD. National kidney foundation K/DOQI clinical practice
guidelines for nutrition in chronic renal failure. Am J Kidney Dis
2001; 37(Suppl 2): S66–S70
12. Cianciaruso B. Italian Society of Nephrology. Conservative therapy
guidelines for chronic renal failure. Giornale Italiano di Nefrologia
2003; (S-24): S48–S60
13. Cianciaruso B, Pota A, Pisani A et al. Metabolic effects of two low
protein diets in chronic kidney disease stage 4–5; a randomized con-
trolled trial. Nephrol Dial Tansplant 2008; 23: 636–644
14. Maroni BJ, Steinman TI, Mitch WE. A method for estimating nitrogen
intake of patients with chronic renal failure. Kidney Int 1985; 27: 58–
65
15. Charlson M, Szatrowski TP, Peterson J et al. Validation of a combined
comorbidity index. J Clin Epidemiol 1994; 47: 1245–1251
16. Ware JE Jr, Sherbourne CD. The MOS 36-item short form health
survey (SF-36).I. Conceptual frameworkand item selection. Med Care
1992; 30: 473–483
17. Devereux B, Alonso DR, Lutas EM. Echocardiographic assessment
of left ventricular hypertrophy: comparison to necropsy findings. Am
J Cardiol 1986; 57: 450–458
18. Apolone G, Mosconi P. The Italian SF-36 Health Survey: translation,
validation, and norming. J Clin Epidemiol 1998; 51: 1025–1036
19. Ware JH Jr, Grandek B, Kosinski M et al. The equivalence of SF-36
summary health scores estimated using standard and country-specific
algorithms in 10 countries: results from the IQOLA Project: Inter-
national Quality of Life Assessment. J Clin Epidemiol 1998; 51:
1167–1170
20. Ministry of Health Degree—22 July 1996. Prestazioni di Assis-
tenza specialistica ambulatoriale erogabili nell’ambito del Servizio
Sanitario Nazionale e relative tariffe. Suppl. to ‘Gazzetta Ufficiale’
no. 216, 14 September 1996
21. Ministry of Health Degree—12 September 2006 ‘Tariffe delle
prestazioni di assistenza ospedaliera’. Supplemento ordinario alla
Gazzetta Ufficiale no. 289, 13 December 2006
22. K/DOQI. Clinical practice guidelines on hypertension and antihy-
pertensive agents in chronic kidney disease. AmJKidneyDis2004;
43(Suppl 1): S1–S290
23. K/DOQI. Clinical practice guidelines for bone metabolism and disease
in chronic kidney disease. AmJKidneyDis2003; 42(Suppl 3): S1–
S201
24. Bailey JL, Wang X, England BK et al. The acidosis of chronic re-
nal failure activates muscle proteolysis in rats by augmenting tran-
scription of genes encoding proteins of the ATP-dependent ubiquitin-
proteasome pathway. J Clin Invest 1996; 97: 1447–1453
25. K/DOQI. Clinical practice guidelines for managing dyslipidemias in
chronic kidney disease. Am J Kidney Dis 2003; 41(Suppl 3): S39–
S56.
26. KDOQI. Clinical practice guideline and clinical practice recom-
mendations for anemia in chronic kidney disease: 2007 update of
hemoglobin target. Am J Kidney Dis 2007; 50: 471–530
27. De Nicola L, Minutolo R, Chiodini P et al. on behalf of the TArget
Blood Pressure LEvels in Chronic Kidney Disease (TABLE in CKD)
Study Group. Global approach to cardiovascular risk in chronic kidney
disease: reality and opportunities for intervention. Kidney Int 2006;
69: 538–545
28. Agarwal R, Nissenson AR, Batlle D et al. Prevalence, treatment, and
control of hypertension in chronic hemodialysis patients in the United
States. Am J Med 2003; 115: 291–297
29. Lucas MF, Quereda C, Teruel JL et al. Effect of hypertension before
beginning dialysis on survival of hemodialysis patients. Am J Kidney
Dis 2003; 41: 814–821
30. Foley RN, Parfrey PS, Harnett JD et al. Clinical and echocardiographic
disease in patients starting end-stage renal disease therapy. Kidney Int
1995; 47: 186–192
31. Parfrey PS, Foley RN, Harnett JD et al. Outcome and risk factors for
left ventricular disorders in chronic uremia. Nephrol Dial Transplant
1996; 11: 1277–1285
32. Cheng LT, Gao YL, Tian JP et al. Sex difference in the prevalence of
left ventricular hypertrophy in dialysis patients. Am J Nephrol 2008;
29: 398–405
33. Young EW, Akiba T, Albert JM et al. Magnitude and impact of
abnormal mineral metabolism in hemodialysis patients in the Dialysis
Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis
2004; 44(Suppl 2): 34–38
34. Borghetti F, Scalone L, Brancati B et al. Cost-benefits analysis of
diet versus dialysis in CKD 5 patients. Value in Health 2008; 11(6):
A651
35. Fukuhara S, Akizawa T, Morita S et al. KRN321 A08 Study Group.
Quality of life improvements in dialysispatients receiving darbepoetin
alfa. Ther Apher Dial 2008; 12: 72–77
36. Levin A, Djurdjev O, Beaulieu M et al. Variability and risk fac-
tors for kidney disease progression and death following attainment of
Stage 4 CKD in a referred cohort. AmJKidneyDis2008; 52: 661–
671
Received for publication: 3.3.09; Accepted in revised form: 20.5.09
by guest on January 13, 2016http://ndt.oxfordjournals.org/Downloaded from
