J Am Soc Nephrol 9: 1798-1804, 1998

Effects of Smoking on Renal Hemodynamics in Healthy

Volunteers and in Patients with Glomerular Disease

EBERHARD RITZ,* URS BENCK,* EDUARD FRANEK,* CHRISTINE KELLER,*

MELCHIOR SEYFARTH,� and JOHN CLORIUSt*Depc,rt,ple,zt of internal Medicine, Ruperto Caro!a University Heidelberg and �Gerinan Cancer Research

Center, Heidelberg, Gertnany; and tTechnical Universir�’, Munich, Germany.

Abstract. Patients with renal disease who smoke have a poor

renal functional prognosis, but the mechanisms involved have

not been explored. In this controlled study, the effects of

smoking and sham smoking were compared in 15 healthy

normotensive volunteers. All were occasional smokers and

abstained from smoking for 48 h as documented by urinary

cotinine measurements. These data were compared with those

of seven patients with biopsy-confirmed IgA glomerulonephri-

tis, also occasional smokers. Renal clearance examinations

were obtained after hydration in the supine position before and

while smoking two cigarettes or sham cigarettes in random

order on 2 consecutive days. GFR and effective renal plasma

flow were determined using In” ‘-diethylenetniamine penta-

acetic acid and ‘311-hippurate with a dual tracer infusion clear-

ance technique. In an ancillary study with six volunteers, the

effect of smoking was compared with the effect of nicotine-

containing chewing gum. In healthy volunteers, sham smoking

caused a minor but significant increase of mean arterial pres-

sure (MAP) and OFR with no significant change of effective

renal plasma flow, filtration fraction (FF), or renovascular

resistance. Smoking caused a significant and more marked

increase of MAP (from baseline 92.8 ± 8.98 to 105 ± 7.78

mmHg) and heart rate (from 61.7 ± 7.52 to 86.4 ± 9.87

min’), accompanied by a significant increase in arginine

vasopressin (from 1 .27 ± 0.72 to 19.9 ± 27.2 pg/ml) and

epinephrine (from 37 ± 13 to 140 ± 129 pg/ml). During

smoking, OFR decreased in all but one volunteer (from 120 ±

17.7 to 102 ± 19.3 ml/min per 1.73 m2), and this was accom-

panied by a significant decrease of FF (from 2 1 .3 ± 4.24 to

17.4 ± 3.41%) and an increase in renovascular resistance

(from 97.6 ± 27.2 to 108 ± 30.4 mmHg . minlml per 1.73 m2).

These findings were reproduced with nicotine-containing

chewing gum. In contrast, when patients with IgA glomerulo-

nephritis smoked, a similar increment in MAP was noted, the

changes of FF were not uniform, and a small but consistent

increase of urinary albumin/creatinine ratio was observed. An

additional 20 volunteers were subjected to the smoking arm of

the study for statistical evaluation of the GFR change in pa-

tients. The difference between the change of OFR between all

volunteers (n = 35) and patients (n = 7) was significant (P <0.005). It is concluded that the known effects of smoking and

nicotine on the sympathetic nervous system and on systemic

hemodynamics are accompanied by significant acute changes

in renal hemodynamics and albuminuria. These findings are of

interest because of the known effects of smoking on progres-

sion of renal disease.

Smoking is a significant health hazard (1-3). It is amazing that,

in contrast to the well known cardiovascular and oncogenic

side effects, the adverse influence of smoking on the kidney

has not attracted more interest (4,5). In subjects with primary

hypertension, urinary albumin excretion, as an index of renal

damage, is highly correlated with smoking (6,7). In patients

with type I diabetes mellitus, the key observation of Chris-

tiansen of an increased renal risk of albuminuria in smokers (8)

was confirmed by numerous subsequent investigators (9-1 1).

The same holds true for patients with type II diabetes (12-15).

In established diabetic nephropathy, the rate of progression to

renal failure in smokers is higher by a factor of 2, and this is

Received April 28. 1997. Accepted March 18, 1998.

Correspondence to Dr. Eberhard Ritz, Department of Internal Medicine, Ru-

perto Carola University Heidelberg, Bergheimer Stra�e 58, D-69l 15 Heidel-

berg, Germany.

1046-6673/09010- 1798$03.00/0Journal of the American Society of Nephrology

Copyright U 1998 by the American Society of Nephrology

true for both type I and type II diabetes (16). Some information

is also available in nondiabetic renal disease. In autosomal

dominant polycystic kidney disease, Chapman er a!. (17) found

that patients with established proteinuria had a greater pack-

year smoking history than their nonproteinuric counterparts. A

retrospective cohort study of lupus nephritis patients docu-

mented that smoking at the time of onset of nephritis was an

independent risk factor for accelerated progression to end-stage

renal failure (18), and this effect was not explained by differ-

ences in hypertension or in immunosuppressive treatment. Fi-

nally, preliminary data from the MRFIT (multiple risk factor

intervention trial) study indicated that the relative risk of end-

stage renal failure was increased by a factor of I .7 in subjects

smoking 20 to 40 cigarettes per day (5).

The potential mechanisms underlying the deleterious effect

of smoking on progression of renal disease have not been

explored. To address this issue, we compared the acute effects

of smoking versus sham smoking in healthy volunteers and in

patients with IgA glomerulonephritis. All subjects were occa-

sional smokers, i.e. , less than five cigarettes per day, and had

J Am Soc Nephrol 9: 1798-1804. 1998 Effects of Smoking on Renal Hemodynamics 1799

abstained for 48 h, as documented by cotinine measurements.

Systemic hemodynamics (mean arterial pressure [MAP], heart

rate) and renal hemodynamics (OFR, effective renal plasma

flow [ERPF], filtration fraction [FF1, renovascular resistance

[RVRI), as well as albuminuria, were assessed in the two

groups to identify the changes that occur during smoking.

Materials and MethodsVolunteers

Thirty-five healthy normotensive volunteers with a history of reg-

ular but limited cigarette consumption (<lO!d) were recruited. There

were 25 male and 10 female volunteers, ranging ir age from 19 to 46

yr. none of whom used medication. Renal disease was excluded based

on the history, physical examination, serum creatinine determination,

urine analysis, and renal ultrasonography. Furthermore, seven patientswith biopsy-confirmed IgA glomerulonephritis (4 men, three women,

age 21 to 45 yr. serum creatinine 0.79 to 1 .68 mgfdl, urinary protein

0.04 to 2.5 g!24 h) were examined. There was nc evidence of acute

exacerbation of glomerulonephritis. Two of these seven patients were

hypertensive, but they were off medication 1 d before the study.

Medication had included bisoprolol, hydrochlorothiazide, and felo-

dipine. All patients were instructed to adhere to a dietary protein

intake of approximately 0.8 g!kg per d.

Protocol

The protocol was approved by the local ethics committee. All

subjects gave informed written consent. Both the consecutive 15

volunteers and all patients were exposed in random order on subse-

quent days to either sham smoking, i.e. , sucking a dummy cigarette,

or active smoking (Marlboro lOOs filter cigarettes containing I mg ofnicotine per cigarette). All individuals were instructed to stop smoking

for 2 d before the study, and this was verified by measuring urinary

cotinine concentrations (below 250 ng!ml in all). They were advised

to drink approximately 700 ml of liquids before and 400 ml during the

study. At 8 am., the subjects emptied their bladders and then rested in

a supine position in a quiet room for 40 mm. Subsequently, they

smoked a first cigarette (or sham cigarette) for 10 mm. After an

interval of 6 mm, they smoked a second cigarette (or sham cigarette)

for 10 mm. (The purpose of the second cigarette was to verify that

changes with respect to baseline were consistent and reproducible.

This was the case in all volunteers. All calculations were based on the

results of the first cigarette. Patients smoked one cigarette only.) After

the first 15 volunteers had been examined, 20 additional individuals

completed the smoking arm of the study only.

For hormonal measurements, blood was taken at the beginning ofthe baseline and at the end of the second smoking period fromindwelling catheters. Urine was collected at the beginning of the

baseline period and at the end of the smoking period.

Ancillary Studies

To determine whether possible effects of cigarette smoking are

provoked by nicotine, the effects of smoking and chewing a nicotine-containing chewing gum (Nicorette#{174}, Pharmacia Co., Erlangen, Ger-

many; 6 mg of nicotine) for 16 mm were compared in six of the 15healthy volunteers. This is known to cause a plasma nicotine level of

approximately 15 ng!ml (information provided by the manufacturer).

Hemodynamic measurements were made during steady-state condi-

tions, which were reached 16 mm after the volunteers had begun to

chew the gum.

Measurements of Renal Hemodynamics

Using a single compartment dual-tracer infusion clearance and the

tracers � � � I-hippurate and � ‘ � In-diethylentriamine penta-acetic acid

(DTPA), we measured both GFR and ERPF in the supine position

using the method of Pedersen et a!. ( 19). The clearance examinationof subjects began 40 mm after simultaneous intravenous injection of

120 j.�Ci ‘3’I-hippurate and 120 p�Ci ‘ ‘ ‘In-DTPA.

A superficial vein of the right arm was used for a continuousinfusion clearance. Both tracers were monitored by two scintillation

probes placed over the right and left shoulder of each individual. This

site is selected because of the large vessels in the probe’s field of

view. Each detector monitored one of the two radioisotopes by means

of energy discrimination. The signals monitored by the probes acti-

vated an infusion pump system via feedback control. Two pumps were

used, one of which contained ‘31I-hippurate, the other ‘ ‘ ‘In-DTPA. A

separate step motor drove the pumps, whereby steady-state conditions

were reached. The data from the first 10 mm of the clearance exam-ination were discarded, because this time was required to equilibrate

the feedback control system. At the end of the baseline clearance

period, 10 ml of blood were drawn from the cubital vein of thecontralateral arm to obtain a plasma sample. A probe from the infused

saline containing each isotope served as standard. A microcomputer

registered the motor step rates, documented the serum activity level of

each isotope in the probe’s field of view, and carried out the clearance

calculations after the activity of the standard and the serum sample

had been registered. By using these data. the clearance was calculated

using the equation:

mx A�,C! =

A�1

where C! is the clearance (ml!min), I is the number of motor steps per

time (min’), A.,� is activity pumped per motor step (MCi), and A�, is

specific activity of plasma (MCi!ml).

The clearance was calculated for 10-mm (or for pauses 6-mm) time

intervals. These separate GFR and ERPF values were then used to

compute a mean clearance for any appropriate time interval studied,

such as resting and smoking or sham smoking.

Single compartment infusion clearance measurements are highly

reproducible. Phantom studies showed that clearance values variedless than 5% in consecutive measurements. Head-on comparison

between classical clearance based on urine collection and infusion

clearance in the same individual had shown that the measured clear-

ance values differ on average by 8.4% and no more than I 1% inindividuals with a GFR >50 ml/min (20). This was now reconfirmed.

Clearance values (GFR) based on urine collection (without catheter-ization relying on spontaneous voiding) and single compartment in-

fusion clearance were compared in five individuals studied for 75 mm.

Clearance based on urine collection was 105 ± 6.51 ml!min and

infusion clearance 96.2 ± 10.2 ml!min, the former being slightly, but

consistently, higher by an average of 10% (2 to 20%).

Ancillary Measurements

BP and pulse were monitored using Dinamap (R. Criticon, Inc.,

Tampa, FL). Other measurements included: urinary sodium and crc-atinine, by Autoanalyzer; cotinine in urine according to the method of

Langone et a!. (2 1 ); active renin by immunoradiometric assay using

Renin III generation assay of Diagnostics Pasteur (Paris, France) (22);

epinephrine according to a two-step extraction and separation with acation-exchange HPLC system (Nucleosil 10 SA, Latek Co., Heidel-

berg, Germany), and subsequent determination as performed by

Smedes et a!. (23) using electrochemical detection for quantification

1800 Journal of the American Society of Nephrology J Am Soc Nephrol 9: 1798-1804, 1998

(detection limit 0. 1 nmolIL): arginine vasopressin (AVP) with RIAusing ‘51-ARG8-vasopressrne (New England Nuclear, Dupont Co.,

Dreieich, Germany), according to Rascher et a!. (24); and urinary

albumin by laser nephelometry (13).

Statistical Analyses

Data are given as mean ± SD. To avoid the problem of multiple

testing (Bonferroni), only a few parameters were defined as primary

end points for statistical analysis (GFR, ERPF, and FF). In eachindividual, the mean value for each phase of the examination in the

smoking and sham smoking arms was compared using the paired t

test. Relative changes (in percentages) between two subsequent pen-ods were also compared, using the paired t test. Comparisons betweenpatients and volunteers were made for the respective periods of the

examination using the unpaired t test. For the beta-error analysis to

assess the power of the study to detect differences between all vol-

unteers (ti = 35) and patients (n = 7), normality of values wasassessed and confirmed in both groups by normal probability plot. For

comparison of the change in GFR, it was established that variances in

the two groups were approximately equal (F test). One-sided unpaired

I test was used.

ResultsEffects of Sham Smoking and Smoking in Healthy

Volunteers

Sham smoking caused a minor but significant increase in

MAP, but no significant change in heart rate (Table 1), non-

epinephrine, epinephrine, AVP, or plasma renin activity (PRA)

(Table 2). In contrast, smoking caused a major increase of

systolic BP and MAP (Table 1), which was accompanied by a

significant increase in heart rate, epinephrine, and AVP con-

centrations, a tendency for an increase in norepinephnne con-

centrations, and a decrease in PRA (Table 2). Marked intenin-

dividual differences in the epinephnine response to smoking

were noted: Some individuals exhibited a pronounced increase,

whereas many remained below the detection limit.

During sham smoking, in parallel with higher BP, GFRincreased slightly with no significant change in ERPF and FF

(Table 3). In contrast, smoking caused a highly significant

decrease in GFR. The average reduction during the 10-mm

smoking period was - 15%. These changes were consistently

noted during the second smoking period as well. The decrease

in GFR was accompanied by a decrease in the FF and by an

increase in RVR. Urine albumin excretion was below the

detection threshold in these hydrated individuals, both before

and after smoking.

To obtain more statistical power for comparison between

volunteers and patients, an additional 20 volunteers were stud-

ied in the smoking arm of the protocol only. The results in the

first 15 volunteers were perfectly reproducible, e.g., in the

overall cohort of 35 volunteers (baseline versus smoking),

MAP increased from 93.5 ± 8.4 to 103 ± 8.3 mmHg; GFR

decreased from 1 15 ± 15.2 to 97.3 ± 16.9 ml/min per 1 .73 m2;

and FF decreased from 21 .4 ± 4.21 to 17.3 ± 33%. The power

analysis is given below.

The Effect of Chewing a Nicotine-Containing Gum on

Renal Hemodynamics in Healthy Volunteers

To examine whether the changes induced by smoking can be

reproduced solely by nicotine, six of the I 5 volunteers had a

third study period during which they chewed nicotine gum. In

principle, all findings seen with cigarette smoking were repro-

duced by the nicotine-containing gum (Table 4).

The Effects of Sham Smoking and Smoking on Systemic

and Renal Hemodynamics in Patients with IgA

Glomerulonephritis

Sham smoking caused no significant changes in renal or

systemic hemodynamics. The effects of smoking on MAP and

heart rate were comparable in patients with IgA glomerulone-

phritis and control subjects (Table 5). In contrast, the direction

of change of GFR, ERPF, and FF varied considerably between

individuals. As a result, the mean values were not significantly

different between the baseline and the smoking periods. Al-

though FF decreased rather consistently in healthy volunteers,

i.e., the changes within individuals, the direction of change of

FF was inconsistent in subjects with IgA glomerulonephnitis.

In contrast to the healthy volunteers, smoking failed to cause

a decrease of GFR in the seven patients with IgA glomerulo-

nephntis. Beta-error analysis comparing the GFR change in the

smoking arm of the study between all 35 volunteers and seven

patients showed that the study had 95% power to detect a 25%

difference in the change of GFR between the groups at a

significance level of P < 0.005. In the patients there was a

Table I. Effects of sham smokingand smoking on systemic circulation in 15 healthy volunteersa

Period Mm

Sham Smoking Smoking

MAP HR MAP HR(mmHg) (mm ‘ ) (mmHg) (mm

Basal 30 93.2 ± 10.9 64.2 ± 8.56 92.8 ± 8.98 61.7 ± 7.52

Cigarette 10 95.6 ± 106b 64.6 ± 6.94 105 ± 7.78c 86.4 ± 9.87’

Pause 6 94.9 ± 12.9 64.6 ± 6.97 102 ± 7.18 82.0 ± 9.05

Cigarette 10 95.1 ± 10.7 64.5 ± 8.92 104 ± 7.55 82.5 ± 8.49

a Mm, minutes; MAP, mean arterial pressure; HR. heart rate.

b p < 0.05, intraindividual differences basal versus sham smoking by paired I test.

C p < 0.0001 , intraindividual differences basal versus smoking by paired t test.d p < 0.0()l , intraindividual differences basal versus smoking by paired t test.

J Am Soc Nephrol 9: 1798-l8()4, 1998 Effects of Smoking on Renal Hemodynamics 1801

Table 2. Hormonal changes during smoking in healthy vo1unteers�’

Norepinephrine Epinephrine AVP Active Renin

Period (pg/mI) (pg!ml) (pg!ml) (ng!ml)

Sham Smoking Smoking Sham Smoking Smoking Sham Smoking Smoking Sham Smoking Smoking

Basal 325 ± 129 293 ± 124 29.0 ± 10.0 37.0 ± 13.0 2.05 ± 1.04 1.27 ± 0.72 6.83 ± 2.61 6.59 ± 3.35

Cigarette 332 ± 127 348 ± 128 27.0 ± 1 1.0 140 ± 129b 144 ± 0.60 19.9 ± 272c.d 5.46 ± 2.06 4.60 ± 180b

a AVP, arginine vasopressin.

b p < 0.01, intraindividual differences basal versus smoking by paired t test.C p < 0.001 by Wilcoxon test for paired differences.d Median, 4.00 pg!ml; range, 0.90 to 83.9 pg!rnl.

Table 3. Renal hemodynamics during sham smoking and smoking in 1 5 healthy volunteersa

Sham Smoking Smoking

Period Mm GFR ERPF FF RVR GFR ERPF � RVR(ml/min�per (ml!minper ((7 ) (mmHg . min/ (ml/min,per (ml!min�er (137 ) (mmHg . min/

I .73m) I .73m) C ml � I .73m2) I .73m) I .73m) #{176} ml . I .73m2)

Basal 30 116± 14.9 603± 114 20.0±2.94 100±39.1 120± 17.7 587±91.7 21.3±4.24 97.6± 27.2

Cigarette 10 130 ± 2l.8’� 627 ± 105 22.3 ± 6.40 84.8 ± 25.0 102 ± 19.3c 603 ± 84.9 17.4 ± 3.4lc 108 ± 3#{216}#{149}4(�

Pause 6 109±34.3 604± 135 18.7±5.46 95.2±40.3 111 ±23.7 587± 103 19.2±2.81 101 ± 25.8

Cigarette 10 111 ± 18.3 618± 113 19.1 ±4.01 95.4±34.8 100±20.0 577±98.0 17.8±2.49 111 ±44.3

a ERPF, effective renal plasma flow; FF, filtration fraction; RVR, renovascular resistance.h p < 0.007, intraindividual differences basal versus sham smoking by paired I test.C p < 0.001, intraindividual differences basal versus smoking by paired t test.

d p < 0.006, intraindividual differences basal versus smoking by paired t test.

significant increase in norepinephrine concentration (from

276 ± 1 84 to 392 ± 2 1 5 pg/ml; P < 0.05) and a tendency for

an increase in epinephrine (from 20.9 ± 26.8 to 34.0 ± 42.9

pg/mi).

Smoking was accompanied by a minor, but consistent and

significant (P < 0.05) increase in urinary albumin/creatinine

ratio in all six patients investigated (one female patient re-

fused), i.e. , from a median of 55.5 (range, 2.7 to 547) to 74.3

(range, 2.8 to 570) [albumin (mg/L):creatinine (mg/L) X 100].

The finding was replicated and confirmed in the individual

patients on two separate occasions.

DiscussionIn view of the known increase of renal risk conferred by

smoking (4,5,8-17), it is of interest to analyze the mechanisms

by which smoking affects the kidney. Smoking is known to

activate the sympathetic nervous system (25) and to induce

secretion of AVP (26). Although the ensuing antidiuresis is

well known (27) and has been investigated in detail (28),

potential renal hemodynamic effects of smoking in humans

have remained largely unexplored.

The main result of the present study is the observation that

in healthy individuals, smoking causes an acute decrease in

GFR and FF in parallel with an increase in HP and heart rate

associated with sympathetic activation. The effect appears to

be mediated by nicotine per se, because it could be reproduced

by chewing a nicotine-containing gum. This observation ex-

eludes a number of other mechanisms that have been discussed

to explain effects of smoking, e.g., damage to endothelial cells

or microcirculatory disturbances (29-33). We acknowledge,

however, that in the genesis of chronic renal effects of smok-

ing, such additional pathomechanisms may definitely play a

role.

Smoking accelerates progression of renal disease, as docu-

mented by several clinical observations (4,5,8 - 17). Conse-

quently, it seemed of interest to investigate whether in patients

with primary glomerular disease the renal hemodynamic

changes induced by smoking were similar compared with

volunteers. To address this issue, we studied seven patients

with IgA glomerulonephritis, occasional smokers with normal

or slightly reduced renal function who had no acute glomeru-

lonephritic episodes at the time of the study. In contrast to the

volunteers, no consistent reduction of GFR and FF was seen

despite a similar significant increase in systemic BP. We

acknowledge that the group sizes were unbalanced, but beta-

error analysis showed that the design had sufficient power to

detect meaningful differences. It is obvious, then, that the renal

hemodynamic effect of smoking is different in healthy mdi-

viduals and in renal patients, but the precise hemodynamic

details will have to be worked out in animal studies. One

finding that would be in line with an injurious effect of smok-

ing on glomerular microcirculation and permselectivity is the

1802 Journal of the American Society of Nephrology J Am Soc Nephrol 9: 1798-1804, 1998

-� � �1�CC � -

P�’.� �) �l.) � +1+1o_ EU �C

�

�

� � H +i+i�.c- �

�I

� �E NoO-r��

>�ar-� +1+1

a �o�- �

� �a5) C � C�C� � !�� � �‘c�

.� C,)

� .C �E +1+1 �C‘6L) �E �c �� - .

-� C +1+1� � 0 0oc.�

-� ,�C-

-� I) --1) .0

.C � -;‘< �Cr� .�0

.� tI_’� ‘�‘ u�.CE oc�

.� LI� +1 +1 < � +1 +1

E r�o� -Cc�c - -�� c-I- .�-.�

�a ‘-C .0 � �‘.� � � C

.� r� �E 22 � .� o� +1 +1

C LL�� 0� �

� c�R� +1+1 Cs� �- � I)

r-i- � _.� �- -- Q) -

c,� I� .� 00-C.) U .� �g +1+1.�

C �c� � C � 00�

� �T N- .-�C

I � +1+1 E ��1005) ‘- .. � � �

�I �‘�‘�C -- �Q �.C00 Q� C,� �C � � - Ncl��c� U C)

� ��o c�.o �

r-1r�1 � C � �“�: �‘�‘:� � � �0. � �rE r��i

�� � >�>�E +1+1 .0.0 � +1+1c,� ! “fr- �il00 �

CC.C � �T �. 00C � #{149}�C �C) 00� �

C) E.� � !‘� o�’� �u C � N �� � � .�

�ir� � u � r--c� �o.-.� � -� .� r-�i�

�U C � �lC) c#� +1+1 �.� I � nn � -- .0.0

� 0 --�cc.�J nn

+1+1 � .� a�C�

>.� C.1) � .0.0 C

.� -- �CI�Cl� >� C- tr1_ �0 C)C)� #{231}�� C u� � rl�� � rn C CE .c .�CC �. E �

C) C22 C) �Ern +1+1.C r-�� n�,� C� -� -C u� �

C -- � .�C) I � � (r�Cr�C- +1+

� � C- N-C� IflN n.�.u c,� � � �0 � E-.�C � C,� � -“1� �CcI� � r� +� +�

nn C � -I C � c� � 5� �

�‘ � � c� 0CCC - .-, --C � C) .� �c

C-cc � �

L) � � -

.� C) � � .� � .0VV�: �

-� C) �a<0� .� C #{149}�-� -� �

�2 �tj � E2

J Am Soc Nephrol 9: 1798-1804, 1998 Effects of Smoking on Renal Hemodynamics 1803

consistent and reproducible, although modest, increase of the

urinary albumin/creatinine ratio in patients with glomerular

disease.

As for the exact intrarenal hemodynamic changes taking

place, we wish to refrain from speculation. It is of interest,

however, that in agreement with previous observations we

noted an acute increase in BP (34,35) and of sympathetic tone

(25). (The increase in BP and in catecholarnine levels was

similar in patients with glomerulonephritis.) We ascribe the

decrease in PRA to the increase in BP. More detailed measure-

ments of systemic hemodynamics have previously shown that

cigarette smoking causes a dose-dependent concomitant in-

crease of cardiac output and of peripheral resistance (36-38).

The increase in cardiac output is mainly due to tachycardia

with only modest changes in stroke volume (38). Whether the

increase in vascular resistance is of similar magnitude in all

vascular beds, or whether the renal circulation is particularly

sensitive to the effects of smoking, remains to be established.

We considered a number of potential artefacts when designing

the present study. We were afraid that smoking would induce a

number of untoward autonomic responses in nonsmokers, which

could confound the results. On the other hand, it appeared difficult

to draw adequate conclusions from studies of heavy smokers in

whom persistent effects of preceding smoking might not have

been adequately washed out. For this report, we decided to study

casual smokers with limited cigarette consumption (< lO/d) who

had abstained from smoking for at least 48 h. The latter was

documented by urinary cotinine measurements.

We also considered that conditioned responses might be in-

volved during smoking. To exclude such nonpharmacologic ef-

fects of smoking, we included a sham smoking arm in the study.

This cautionary measure was well justified: The observation of

minor, but consistent, changes in BP and GFR show that sham

smoking is not neutral with respect to hemodyr�amic responses.

Only a few animal experiments on the renal effects of

smoking are available (4). In the study of Pawlick et al. (39),

anesthetized dogs were examined under three sets of condi-

tions. Nicotine was infused into the renal artery of untreated

animals, of animals after pretreatment with propranolol, and of

animals subjected to bilateral adrenalectomy. Nicotine in-

creased urine flow rate, electrolyte excretion (Na, Cl), and

GFR of control dogs, whereas it decreased urine flow rate,

electrolyte excretion, and GFR in dogs preti-eated with pro-

pranolol or having undergone adrenalectomy. The authors con-

cluded that nicotine was saluretic and diuretic by releasing

catecholamines acting on the kidney via beta-adrenergic recep-

tors. Additional studies using adrenoreceptor blockers in hu-

mans will be necessary to prove or refute this hypothesis.

We conclude that in healthy individuals, smoking exerts potent

and consistent effects on glomerular function, as reflected by a

decrease of GFR and FF. The renal hemodynamic response in

subjects with glomerular disease is more variable and is accom-

panied by an increase in albuminuria. In both groups, smoking

increased systemic BP (34,35,40), a known promoter of renal

injury.

AcknowledgmentsThis work was conducted with financial support from Dr. Klara

Haeffner (Heidelberg). We thank Professor Rascher and Dr. H#{228}nze

(Giessen) for measurements of plasma AVP concentrations, Professor

Schmidt-Gayk (Heidelberg) for urinary cotinine measurements, and

Dr. Zuna (German Cancer Research Center, Heidelberg) for statisticaladvice.

References1 . Report of the Surgeon General: Cardiovascular disease. Wash-

ington DC, U.S. Department of Heath and Human Services

Office on Smoking and Health, DHHS(PHS) 84-50204, DHHS1904, 1983

2. Bartecchi CE, Mackenzie TD, Schrier RW: The human costs of

tobacco use. N Eng! J Med 330: 907-912, 1994

3. Bartecchi CE, Mackenzie TD, Schrier RW: The global tobacco

epidemic. Scientific American May 1995, pp 44-51

4. Orth SR, Ritz E, Schrier RW: The renal risks of smoking [Edi-

tonal]. Kidney mt 51: 1669-1677, 1997

5. Whelton PK, Randall B, Neaton J, Stamler J, Brancati FL, Klag

MJ: Cigarette smoking and ESRD incidence in men screened forthe MRFIT. J Am Soc Nephro! 6: 408a, 1994

6. Mimran A, Ribstein J, Ducailar G, Halimi JM: Albuminuria in

normals and essential hypertension. J Diabetes Complications 8:

150-156, 1994

7. H#{246}rnerD, Fliser D, Klimm HP, Ritz E: Albuminuria in normo-tensive and hypertensive individuals attending offices of general

practitioners. J Hypertens 14: 655-660, 1996

8. Christiansen JS: Cigarette smoking and prevalence of microan-

giopathy in juvenile-onset insulin-dependent diabetes mellitus.

Diabetes Care 1: 146-149, 1978

9. M#{252}hlhauserI, Sawicki P. Berger M: Cigarette smoking as a riskfactor for macroproteinuria and proliferative retinopathy in type

1 (insulin-dependent) diabetes. Diabeto!ogia 29: 500-502, 1986

10. Couper JJ, Staples AJ, Cocciolone R, Nairn J, Badcock N,

Henning P: Relationship of smoking and albuminuria in children

with insulin-dependent diabetes. Diabetic Med 1 1: 666-669,

1994

1 1. Telmer 5, Christiansen JS, Andersen AR, Nerup J, Deckert T:

Smoking habits and prevalence of clinical diabetic microangi-

opathy in insulin-dependent diabetics. Acta Med Scand 215:

63-68, 1984

12. Olivarius N de F, Andreasen AH, Keiding N, Mogensen CE:Epidemiology of renal involvement in newly diagnosed middle-

aged and elderly diabetic patients: Cross-sectional data from the

population-based study “Diabetes Care in General Practice,”

Denmark. Diabeto!ogia 36: 1007-1016, 1993

13. Keller CK, Bergis KH, Fliser D, Ritz E: Renal findings in

patients with short term type 2 diabetes. J Am Soc Nephro! 7:

2627-2635, 1996

14. Savage S, Nagel MJ, Estacio RO, Lukken N, Schrier RW: Risk

factors associated with urinary albumin excretion in type II

diabetes. Am J Kidney Dis 25: 836-844, 1995

15. Mattock M, Barnes D, Viberti GC, Keen H, Fitzgerald A, Burt D,

Jackson PG: Predictors of the development of microalbuminuria

in type 2 diabetic patients. Diabetologia 38[Suppl 1]: A63, 199516. Biesenbach G, Janko 0, Zazgornik J: Similar rate of progression

in the predialysis phase in type I and type II diabetes mellitus.

Nephrol Dial Transplant 9: 1097-1 102, 199417. Chapman AB, Johnson AM, Gabow PA, Schrier RW: Overt

proteinuria and microalbuminuria in autosomal dominant poly-

cystic kidney disease. J Am Soc Nephro! 5: 1349-1354, 1994

1804 Journal of the American Society of Nephrology J Am Soc Nephrol 9: 1798-1804. 1998

18. Ward MM, Studenski S: Clinical prognostic factors in lupus

nephritis: The importance of hypertension and smoking. Arch

Intern Med 152: 2082-2088, 1992

19. Pedersen iF, Pedersen MF, Madsen P0: Clinical renal testing byexternal double isotope monitoring technique. Nuc! Med 10:

16-24, 1971

20. Mohring K: Vergleichende Prufung und klinische Anwendung

radiomarkierter Substanzen zur katheterlosen, auch seitenge-trennten, simultanen Bestimmung des Glomerulusfiltrates und

des Nierenplasmaflusses. Habi!itationsschrift, Heidelberg, 1973

2 1 . Langone JJ, Gjika HM, Van Vunakis H: Nicotine and its metab-

olites: Radioimmunoassays for nicotine and cotinine. Biochemn-i5tP)� 12: 5025, 1973

22. Roth Hi: Validation of the immunoradiometric assay “Renin IIIGeneration” for the determination of active renin concentration.

Clin Lab 40: 1007-1015, 1994

23. Smedes F, Kraak JC, Poppe H: Simple and fast solvent extraction

system for selective and quantitative isolation of adrenaline,

noradrenaline and dopamine from plasma and urine. J Chro-

mnatogr23l: 25-39, 1982

24. Rascher W, Rauh W, Brandeis WE, Huber KH, Scharer K:Determinants of plasma arginine and vasopressine in children.

Ada Paediatrica Scand 75: 1 1 1-1 17, 1986

25. Cryer PE, Haymond MW, Santiago JV, Shah SD: Norepineph-rine and epinephrine release and adrenergic mediation of smok-

ing-associated hemodynamic and metabolic events. N Eng!

J Med 295: 573-577, 1976

26. Rowe 1W, Kilgore A, Robertson GL: Evidence in man that cigarette

smoking induces vasopressin release via an airway-specific mech-

anism. J Cliii Endocrino! Metab 51 : 170-172, 1980

27. Burn iH, Truelove LH, Burn I: The antidiuretic action of nicotine

and of smoking. Br Med J 1 : 403-406, 1945

28. Cadnapaphornchai P. Boykin IL, Berl T, McDonald KM, SchrierRW: Mechanism of effect of nicotine on renal water excretion.Amn J Phvsiol 227: 1216-1220, 1974

29. Aronow WS, Dendinger J, Rokaw SN: Heart rate and carbon

monoxide level after smoking high-, low-, and non-nicotine

cigarettes: A study in male patients with angina pectoris. Ann

intern Med 74: 697-702, 1971

30. Davis JW, Arnold J, Wiegmann T: Cigarette-smoking affects

platelets and endothelium in chronic hemodialysis patients.

Nephron 64: 359-364, 1993

31. Nadler IL, Velasco IS, Horton R: Cigarette smoking inhibits

prostacyclin formation. Lancet 4: 1248 -1250, 1983

32. Uedelhofen WM, RUtzel A, Meese CO. Weber PC: Smoking

alters thromboxane metabolism in man. Biochim Biophys Acta

1081: 197-201, 1991

33. Noronha-Dutra AA, Epperlein MM, Woolf N: Effect of cigarette

smoking on cultured human endothelial cells. Cardiovasc Res

27: 774-778, 1993

34. Hesse E: Der Einflu� des Rauchens auf den Kreislauf. Dtsch

Arch K!in Med 89: 565-575, 1907

35. Mann IS, James GD, Wang RS, Pickering TG: Elevation of

ambulatory systolic blood pressure in hypertensive smokers: A

case control study. JAMA 265: 2226-2228, 1991

36. Sato T, Kunishi K, Kameyama A, Takano T, Saito D: Acute

hemodynamic effect of cigarette smoking and its relationship

with nicotine content. Kokvu-To-Junkan 39: 15 1-155, 1991

37. Omvik P: How smoking affects blood pressure. Blood Press 5:

71-77, 1996

38. Tachmes L, Fernandez Ri, Sackner MA: Hemodynamic effects

of smoking cigarettes of high and low nicotine content. C/test 74:

243-246, 1978

39. Pawlick WW, Jacobson ED, Banks RO: Actions of nicotine on

renal function in dogs. Proc Soc Exp Biol Med I 78: 585-590,

1985

40. Groppelli A, Giorgi DMA, Omboni S, Parati G, Manchia G:Persistent blood pressure increase induced by heavy smoking.

J Hvpertens 10: 495-499, 1992