EURURO Vol. 73 No. 1

2.2.

Interventions and follow-up

All patients underwent routine history and physical examination, as well

as baseline renal scintigraphy scan within 6 mo of surgery, an abdominal

and pelvic computed tomography scan and/or a renal ultrasound within

2 mo of surgery, chest X-ray or computed tomography within 30 d of

surgery, and preoperative laboratory investigations, including baseline

eGFR.

The choice of approach and use of renal hypothermia during NSS was

at the discretion of the patient and treating physician. Selective occlusion

of multiple primary renal arteries, when present, or nonclamping of the

artery, was performed at the surgeon's discretion and recorded. Routine

clamping of the renal vein(s) was not performed (

= 2, both in the

mannitol arm).

The intraoperative

uid management was standardized tomaintain a

minimum systolic blood pressure of 90

–

100 mmHg and a minimum

urine output of 0.5 ml/kg/h. This included intravenous hydration at a

target infusion rate of 10 1 ml/kg over the 1st h of surgery and of

6 1 ml/kg thereafter. Intraoperative hypotension (systolic blood

pressure

90 mmHg) or urine output

0.5 ml/kg/h were treated with

250-ml intravenous boluses of Normosol. Hypotension refractory to

intravenous

uid boluses was treated with small doses of intravenous

phenylephrine or ephedrine or neostigmine or combinations thereof.

The protocol allowed for replacement of surgical blood loss on a volume

basis ml per ml using colloids, such as albumin or 6% hetastarch solution,

or packed red blood cell units transfused.

The treatment arm received a standard dose of 12.5 g of mannitol

(200 ml of a 6.25% mannitol solution), completely infused intrave-

nously over 5

–

10 min within 30 min prior to vascular occlusion of

the renal artery. The placebo arm received 200 ml of normal saline

solution.

Follow-up assessments included serum creatinine and eGFR

measurements on postoperative d 1 and 2, at 6 wk, and 6 mo; and a

renal scintigraphy scan at 6 mo.

Secondary endpoints included the determination of differences

between the arms in eGFR at 6 wk ( 2 wk), a postoperative radionuclear

scintigraphy renal scan at 6 mo ( 1 mo), rates of grade 3

–

5 com-

plications within 30 d of surgery, estimated blood loss, and ischemia

time.

2.3.

Adverse events

Perioperative complications were de

ned as any complication grade

2 in the Memorial Sloan Kettering-modi

ed Clavien-Dindo grading

system that occurred intraoperatively and within 30 d postoperative-

ly. Intraoperative complications were de

ned as any complication

of grade 2 related to either surgical or clinical factors during

the procedure. Immediate perioperative morbidity was de

ned as

any complication grade 2 that occurred during the initial hospital

stay.

2.4.

Sample size

The primary objective was to determine the difference in eGFR between

the mannitol group and the placebo group 6 mo after NSS. Based on

historical data

[7]

, the expected mean eGFR at 6 mo after NSS in patients

who did not receive mannitol was 60 ml/min/1.73 m

. The observed

correlation between preoperative eGFR and 6-mo postoperative eGFR

was 0.6. The standard deviation was 14.4 ml/min/1.73 m

, and the 75th

percentile of the standard deviation estimated using bootstrap methods

was 14.9 ml/min/1.73 m

. This in

ated standard deviation was used for

the sample size calculation as a conservative measure to reduce the

likelihood of underpowering. To detect a minimal clinically signi

cant

difference of six units of eGFR, a trial with an

of 5%, a power of 90%, and

a planned interimanalysis for ef

cacy required 88 patients per treatment

arm. Sample size was expanded to 105 patients per treatment arm to

account for loss to follow-up.

2.5.

Statistical analysis

The difference in eGFR at 6 wk and 6 mo between the two groups was

assessed by analysis of covariance (ANCOVA) with eGFR after surgery

as the outcome, and preoperative eGFR, randomization treatment

group, and surgical approach as covariates. A threshold of six units

of eGFR was used to de

ne clinical signi

cance. Two-tailed

value and

a 95% con

dence interval (CI) for the difference between groups

were determined. In accordance with the study protocol, a linearly

interpolated 6-mo eGFR was calculated for patients who missed an

eGFR measurement between 4 mo and 8 mo after surgery, but who

had eGFR measured at 3 mo and between 8 mo and 12 mo (

= 15).

The ANCOVA model was also used to compare baseline and 6-mo

nuclear renal scintigraphy differences, estimated blood loss, and

ischemia time. The ANCOVA model included treatment, baseline eGFR,

and type of surgery as covariates. The renal scintigraphy model also

included baseline renal scintigraphy measurements. Owing to the

paucity of 90-d grade 3

–

5 complications occurring during the study

(

= 15 grade 3

–

5 complications), the Fisher's exact test was used to

assess the differences between the treatment groups.

The primary analysis was conducted according to intention to treat.

Analyses of the main and secondary endpoints were also repeated

according to the actual treatment received. O

’

Brien-Fleming stopping

boundaries and a signi

cance level of 0.0031 were utilized for the

interim analysis

[8] .

All analyses were conducted using Stata 13.0

(StataCorp LP, College Station, TX, USA).

Results

Overall, 210 patients were enrolled in the trial. Of these,

105 were assigned to each treatment arm. Five patients in

the placebo arm and four patients in the mannitol arm

withdrew consent prior to surgery and were excluded from

all analyses. Two patients in the placebo arm had their

surgery cancelled and they were excluded as well. NSS was

converted to radical nephrectomy in one patient in each

arm, and one patient in the mannitol arm never received the

study drug during NSS. These three patients were included

in the analyses according to the intent-to-treat principle;

thus, 98 and 101 patients were included in the placebo and

mannitol arms, respectively

( Fig. 1 )

. Baseline characteristics

of the two groups were similar

( Table 1 )

. None of the

patients received intraoperative intravenous diuretics (ie,

furosemide).

3.1.

Renal function and adverse event outcomes

Renal function outcomes are shown in

Table 2

and

Figure 2

. Of the 199 patients studied, 178 (91 in the placebo

arm and 87 in the mannitol arm) had complete 6-mo eGFR

data and were included in the primary analysis. After

adjusting for preoperative eGFR and surgical technique, the

difference in 6-mo eGFR was 0.2 (95% CI:

–

3.1 to 3.5,

= 0.9).

A total of 180 patients (91 in the placebo arm and 89 in the

mannitol arm) had complete 6-wk eGFR data and were

included in the 6-wk comparison. Similar to the 6-mo

analysis, no significant difference in eGFR between the two

E U R O P E A N U R O L O GY 7 3 ( 2 0 18 ) 5 3

–

5 9