EURURO Vol. 73 No. 1

arms was detected (adjusted difference:

–

2.6, 95% CI:

–

5.8

to 0.7,

= 0.12).

No significant differences were observed on 6-mo

renal unit % split function radionuclear scintigraphy

scans (adjusted difference:

–

1.7, 95% CI:

–

3.8 to 0.4,

= 0.11). Overall, 15 (7.5%) grade 3

–

5 complications were

observed in the trial, and no difference was detected

between the study arms (difference: 3.2%, 95% CI:

–

4.1% to

11%,

= 0.4). Estimated blood loss and renal ischemia

time were also similar (

= 0.2 and

= 0.7, respectively;

Table 2

We did not find that the effect of mannitol on post-

operative eGFR differed by baseline eGFR (6-mo eGFR

= 0.9; 6-wk eGFR,

= 1) or by type of surgery (6-mo eGFR

= 0.2; 6-wk eGFR,

= 0.5).

Post-hoc analysis was completed to determine if there

was interaction between mannitol and clamp time and it

was not significant (

= 0.09).

Discussion

The primary aim of this study was to assess the impact on

renal function of intravenous mannitol to determine its

role as an agent that protects against the effects of

transient renal ischemia during NSS. The results of our

study show that mannitol has no clinically measurable

effect on renal function recovery after NSS in patients

with adequate preoperative renal function. Indeed, the

upper bound of the 95% CI for the difference between

study groups excluded the prespecified minimum clini-

cally significant difference. We can thus conclude that

mannitol should not be used in NSS in this clinical

context.

Earlier publications of several well-conducted animal

studies have been used to support the use of mannitol

[1]

. Green et al

[9]

used acute and chronic models to test the

administration of medications on the recovery of rabbit

kidneys from the effects unilateral normothermic renal

ischemia after 1 h of renal vascular occlusion. Only

propranolol and the diuretics mannitol and furosemide

showed beneficial effect. A follow-up study determined that

0.25 g/kg of mannitol administered 15 min before warm

renal ischemia was beneficial

[10]

. The standard dose of

12.5 g of intravenous mannitol administered during NSS

resulted from these studies

[9,10] .

Considering higher doses

of mannitol, a prospective randomized trial on 50 recipients

of cadaveric donor renal transplant comparing the admin-

istration of either 50 g of intravenous mannitol or saline

infusion prior to graft revascularization showed a signifi-

cantly decreased incidence of postoperative ATN in the

mannitol group, no impact on the 3-mo eGFR was detected

[11]

To date, the use of intravenous mannitol during NSS has

been part of surgical practice largely due to its legacy effect,

and despite the potential risks. Detrimental renal effects

investigated by Gelman

[6]

identified disproportionate

shifts of the renal blood flow received by the renal cortex

(94%) compared with the renal medulla (6%). By selectively

[(Fig._2)TD$FIG]

Fig. 2

–

Estimated glomerular filtration rate (eGFR) over the study

period by treatment group. Small gold dots represent the observed eGFR

in patients enrolled in the mannitol arm; small blue dots represent the

observed eGFR in patients randomized to placebo. Large dots represent

the treatment group means.

Table 2

–

Primary and secondary endpoint outcomes

Placebo (

= 98)

Mannitol (

= 101)

Difference

95% CI

value

6-mo postoperative eGFR (

= 178)

0.9

eGFR ml/min/1.73m

2 a

80 (18)

78 (20)

0.2

–

3.1 to 3.5

% change from preoperative eGFR

–

8.4 (12)

–

8.5 (14)

0.1

–

3.8 to 4.0

eGFR

60 ml/min/1.73m

12 (13%)

18 (21%)

–

18% to 3.5%

6-wk postoperative eGFR (

= 180)

0.12

eGFR ml/min/1.73m

2 a

80 (19)

79 (20)

–

2.6

–

5.8 to 0.7

% change from preoperative eGFR

–

9.4 (14)

–

6.2 (13)

–

3.2

–

7.1 to 0.7

eGFR

60 ml/min/1.73m

13 (14%)

19 (21%)

–

18% to 4.1%

6-mo renal scan % split functio

n a ( n = 161)

42 (8)

44 (8)

–

1.7

–

3.8 to 0.4

0.11

Grade 3

–

5 complications (

= 199)

9 (9.2%)

6 (5.9%)

3.2%

–

4.1% to 11%

0.4

Estimated blood loss (ml)

235 (218)

284 (297)

–

119, 26

0.2

Clamping time (min)

27 (13)

27 (12)

–

0.7

–

3.9, 2.6

0.7

Statistics presented in the

rst two columns indicate the mean (standard deviation).

CI = con

dence interval; eGFR = estimated glomerular

ltration rate.

Individual renal unit where nephron sparing surgery was completed. Difference adjusted for preoperative eGFR and surgical approach. The 6-mo renal scan

model was also adjusted for baseline renal scan (% split function for individual renal units).

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

–

5 9