Smoking, oxidative stress and inflammation: Impact on resting energy expenditure in diabetic nephropathy

BioMed CentralBMC Nephrology

Open AcceResearch articleSmoking oxidative stress and inflammation Impact on resting energy expenditure in diabetic nephropathyRajiv Agarwal

Address Department of Medicine Indiana University and RLR VA Medical Center Indianapolis Indiana USA

Email Rajiv Agarwal - ragarwaliupuiedu

Corresponding author

AbstractBackground Inflammation is associated with increased resting energy expenditure (REE) inpatients with chronic kidney disease Oxidative stress on the other hand appears not to increaseREE Smoking is a common mechanism for generating oxidative stress and inflammation Whethersmokers have increased REE and if so whether it is accounted for by the pro-oxidant andinflammatory state is not known

Methods A case control study of 11 smokers and 24 non-smokers with overt diabeticnephropathy was performed to evaluate the chronic effect of smoking on REE REE (indirectcalorimetry) glomerular filtration rate (iothalamate clearance) markers of oxidative stress (urinaryand plasma malondialdehyde (MDA) and protein carbonyls) and inflammation (C-reactive proteintumor necrosis factor-alpha interleukin-6) were measured on two occasions four months apart

Results Biomarkers of inflammation (C-reactive protein) and oxidative stress (urinary and plasmaMDA) were increased in smokers REE was increased in smokers 243 kcalkgday compared to 21kcalkgday (p = 0009) in non-smokers After adjusting for age GFR MDA C-reactive protein andhemoglobin A1C the difference in REE between the two groups persisted (adjusted difference 351kcalkgd 95 confidence interval 059 ndash 645 p = 0020)

Conclusion Patients with overt diabetic nephropathy who smoke have a higher REE oxidativeand inflammatory state Elevated REE is not attributable to heightened oxidative stress andinflammatory state Smoking is an independent risk factor for elevated REE in patients with diabeticnephropathy and provides an additional mechanism by which it may lead to poor outcomes

BackgroundDiabetic nephropathy is the commonest cause of end-stage renal disease (ESRD) Malnutrition in patients withESRD is associated with increased morbidity and mortal-ity Nutritional decline begins long before patients withprogressive CKD become dialysis dependent [12] whichin part is related to the spontaneous reduction in dietary

protein and caloric intake with reduction in glomerularfiltration rate (GFR) [3-8]

Recent studies point out that chronic subclinical inflam-mation in patients with CKD is associated with anincrease in REE [9] Other studies done in patients with-out CKD show that oxidative stress is not associated withincrease in REE [10] Smoking is associated with both an

Published 22 November 2005

BMC Nephrology 2005 613 doi1011861471-2369-6-13

Received 01 June 2005Accepted 22 November 2005

This article is available from httpwwwbiomedcentralcom1471-2369613

copy 2005 Agarwal licensee BioMed Central Ltd This is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby20) which permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

of 9(page number not for citation purposes)

BMC Nephrology 2005 613 httpwwwbiomedcentralcom1471-2369613

increase in oxidative stress and subclinical inflammationthus serves as a model to study the effect of oxidative-inflammatory state on REE [11] Smoking may thereforecompound the oxidative stress and the inflammatory statein patients with CKD [212-14] Whether smoking-induced oxidative-inflammatory state is sufficient toaccount for elevated REE is not known

Although smoking-induced increase REE is thought to beacute and transient ndash in part due to nicotine-induced sym-pathoadrenal activation ndash more recent studies point to achronic elevation in REE [15] In epidemiological studiesincreased REE is seen in young women who smoke com-pared to women who do not even after controlling fordifferences in body size and overnight abstinence fromsmoking [16] Smoking cessation often leads to weightgain [17] Thus smoking may accelerate malnutrition viaaccelerating attrition in renal function in addition to ele-vating REE

In this study we asked the question whether smokers withdiabetic nephropathy have an increased REE compared tonon-smokers with equally severe diabetic nephropathy Ifso is the increased REE mediated by the pro-oxidant andinflammatory state

MethodsSubjectsThirty-five clinically stable Veterans with diabetic neph-ropathy were recruited from the Renal Clinic of theRoudebush VA Medical Center Indianapolis IN afterreview and approval of protocol by the InstitutionalReview Board of Indiana University and VA Research andDevelopment Committee Signed written informed con-sent was obtained from each participant

Patients with type 2 diabetes requiring treatment with oralhypoglycemic drugs or insulin were required to have aurine proteincreatinine ratio of gt10 gg on a singlevoided specimen and a creatinine clearance of gt20 mLmin by Cockcroft-Gault formula All patients were partic-ipating in a randomized clinical trial examining the influ-ence of pioglitazone compared to glipizide on reductionof proteinuria Exclusion criteria included the presence ofliver disease NYHA Class III or IV heart failure unstableangina myocardial infarction or stroke in the previous 3months NSAID use or body mass index of ge40 kgm2

Patients underwent a standardized history and physicalexamination that included assessment of medications andcontrol of diabetes mellitus by hemoglobin A1C Currentsmoking status was ascertained by a yes or no responseThe following measurements were made in 32 of 35 sub-jects on two occasions four months apart in the remain-ing three only one measurement was made

Glomerular filtration rateGFR was measured by iothalamate clearance using a con-tinuous subcutaneous infusion at a rate of 125 microLhourthat was started after a bolus intravenous injection 24hour prior to actual measurements [18] The followingday six samples of plasma were collected at times 0 1015 20 25 and 30 hours while the subjects were waterloaded Plasma iothalamate concentrations were analyzedusing a previously reported high-performance liquidchromatography technique [19] The ratio of infusion rateto steady-state plasma concentration yielded the iothala-mate clearance The average of these six collections wasexpressed as the GFR

Resting energy expenditureSeated measurements were made in the morning after anovernight fast Subjects were asked not to drink or eat any-thing and refrain from smoking after midnight They wereasked only take half the usual dose of insulin if they wereon insulin Two indirect calorimetric measurements weremade 4 months apart in 32 patients and one measure-ment in the remaining 3 patients

REE was calculated using open circuit indirect calorimetryin a computerized metabolic system the gas concentra-tions of O2 and CO2 were measured in real time using asgas mass-spectrophotometer (Marquette Electronics IncMilwaukee WI) connected to a computer The flow rate ofthe inhaled and exhaled air was simultaneously measuredusing a flow meter interfaced to an analog to digital con-version board with data sampled at 20 Hz and stored to amemory file The data was analyzed using First BreathSoftware (First Breath Inc Ontario Canada) to calculateoxygen consumption and carbon dioxide production inmLmin The modified Weirs equation (REE (Kcalday) =[VO2 mLmin (394) + VCO2 mLmin (111)] times 144) [20]was used to calculate the REE Data were collected for 30minutes and for the purposes of analysis the first 5 min-utes of the collection were discarded since they did notrepresent steady state

Lean body massTo relate REE to differences in body composition betweensmokers and non-smokers lean body mass was measuredby anthropometric and body impedance spectroscopy

Anthropometric methodSkinfold thickness measurements were taken in triplicatefor biceps triceps subscapular and supra-iliac areas by asingle observer using Lange skin-fold calipers (Beta Tech-nology Inc Cambridge MD USA) and lean body mass(FFM SFT) was calculated by Durnin and Womersley for-mula [21]

Body impedance spectroscopyBody composition was measured using bioelectric imped-ance analysis (BIA) A BIA-106 Spectrum II analyzer (RJLSystems Inc Clinton Twp MI) was used and data wascollected using Weight Manager Version 20 software (RJLSystems Inc Clinton Twp MI) The technique of BIAinvolved introducing an imperceptible 800 microA alternatingcurrent at 50 kHz through signal introduction cablesplaced with pre-gelled electrodes on the ulnar head andmedial malleolus Sensing electrodes placed at the ipsilat-eral 3rd metacarpal and 2nd metatarsal determined theelectrical resistance and reactance Lean body mass wascalculated from estimates of total body potassium

Protein nitrogen appearanceTwo 24-hour urine samples were collected for the meas-urement of creatinine urea sodium potassium chlorideand protein The average of the two 24-hour excretionrates was used for analyses Dietary protein intake wasestimated from the urea nitrogen appearance rate (UNA)calculated as follows [22] Protein intake (gday) =625[urine urea nitrogen (UUN) (gday) + 0031 times bodyweight(kg)]

Oxidative stress markersMalondialdehyde (MDA)Malondialdehyde (MDA) a lipid hydroperoxide isformed by β-scission of peroxidized polyunsaturated fattyacids and was measured by high performance liquid chro-

matography following derivatization with thiobarbituricacid (TBA) as reported previously [23] Urinary MDA wasanalyzed in three spontaneously voided specimens over aperiod of 3 hours Each specimen was analyzed for MDAand creatinine The MDAcreatinine ratio was averagedand used further for statistical analysis Blood was col-lected in tubes containing EDTA at first and last visitsPlasma was separated by centrifugation and stored at -84degC until analysis

Carbonyl measurementsOxidation of plasma and urine proteins were measured byanalysis of Western blots with slight modifications of aprevious report [24] Total protein was determined usingbicinchoninic acid (BCA-1 Protein Assay Kit Sigma StLouis) Urine protein was measured by the dye bindingmethod using a complex of pyrogallol red and molybde-num acid (QuanTtest Red Total Protein Assay SystemQuantimetrix Redondo Beach CA) Plasma was diluted125 (vv) with phosphate-buffered saline (PBS) one aliq-uot of the diluted sample was derivatized and another pre-pared as an underivatized control using the OxyBlotprotein oxidation detection kit (Intergen Purchase NYUSA) Urine samples were derivatized with dinitro phenylhydrazine (DNP) or control reagent similarly except thatsamples were not diluted prior Derivatized and underiva-tized plasma or urine samples were loaded on electro-phoresis gels in volumes calculated to give 5 microg proteinper sample and electrophoresed according to the method

Table 1 Clinical and Laboratory Characteristics of Patients with Type 2 Diabetic Nephropathy

Parameter Non-smokers Mean plusmn SD or N () Smokers Mean plusmn SD or N ()

N 24 11Age (y) 691 plusmn 70 615 plusmn 93 MalesFemales 240 110Race WhiteBlack 195 92Weight (kg) 978 plusmn 168 1025 plusmn 234BMI (kgm2) 326 plusmn 49 326 plusmn 66Serum albumin (gdL) 38 plusmn 033 363 plusmn 030Serum Creatinine (mgdL) 275 plusmn 121 180 plusmn 074 BUN (mgdL) 465 plusmn 256 383 plusmn 218Hemoglobin (gdL) 133 plusmn 16 135 plusmn 2424 hour urine protein (gd) 3393 plusmn 2522 4423 plusmn 4385Iothalamate clearance (GFR) (mLmin173 m2) 289 plusmn 138 472 plusmn 348 Duration of diabetes (yr) 157 plusmn 91 133 plusmn 90Insulin use 15 (63) 6 (55)Hemoglobin A1C () 757 plusmn 202 898 plusmn 293WaistHip Ratio 100 plusmn 007 104 plusmn 007Antihypertensive medications 43 plusmn 16 39 plusmn 19ACE inhibitor or Angiotensin receptor blocker use 21 (88) 8 (73)Total Cholesterol (mgdL) 184 plusmn 51 223 plusmn 87LDL Cholesterol (mgdL) 100 plusmn 44 116 plusmn 24HDL Cholesterol (mgdL) 42 plusmn 111 46 plusmn 114Vascular Disease (coronary cerebrovascular or peripheral vascular disease) 17 (71) 7 (64)

p lt 005

of Laemmli on 4 to 20gradient SDS-PAGEgels (Bio-RadHercules CA USA) for 60 minutes at 200 V Followingelectroblotting to 02 micro nitrocellulose for 60 volt hoursthe membrane was blocked with subsequent immunob-lotting using OxyBlot Kit methods and reagents Bandswere visualized with chemiluminescence and captured onfilm and analyzed densitometrically using dedicated soft-ware using a GelLogic 100 System (Kodak RochesterNY)

Samples for individual patients before and after therapy(glipizide or pioglitazone) including derivatized andunderivatized control were analyzed on a single Westernblot to ensure that responses to therapy were comparedunder the same analytical conditions The blot wasstained for protein with amido black and the ratio of thedensity of the carbonyl band vs the corresponding proteinband was used to calculate the percent carbonylation oftotal protein and comparisons before and after therapymade The carbonyl band corresponding to the molecularweight of albumin was also compared to the correspond-ing amido black band to generate a ratio to assess the pro-portion of albumin oxidized

Inflammation markersTotal leukocyte count (WBC count)Blood was collected in EDTA containing vacutainers(Beckton-Dickinson) and total leukocyte count calculatedby the Coulter method (Coulter STK-S Coulter Electron-ics Inc Hialeah FL) by our clinical laboratory

C-reactive Protein (CRP)CRP was measured by Cobas Integra 400 autoanalyzerusing a particle enhanced turubidimetric assay (CobasIntegra C-Reactive Protein Latex Roche Diagnostics Indi-

anapolis IN) The intra-assay coefficient of variation was18 and the inter-assay coefficient of variation was 29at a mean level of 062 mgdL CRP

Interleukin 6 (IL-6)Interleukin 6 (IL-6) was assayed in plasma using a sand-wich ELISA (Quantikinereg kit for Human IL-6 Immu-noassay RampD Systems Minneapolis MN) A standardcurve was generated using a linear curve-fit The correla-tion coefficient for standards was greater than 099 andthe lowest detectable limit 0039 pgml in undilutedplasma The intra-assay coefficient of variation was 78and the inter-assay coefficient of variation was 72

Tumor Necrosis Factor-α (TNF-α)Tumor Necrosis Factor-α (TNF-α) was assayed in plasmausing a sandwich ELISA (Quantikinereg kit for Human TNF-α Immunoassay RampD Systems Minneapolis MN) Astandard curve was generated using a linear curve-fit Thecorrelation coefficient for standards was greater than 099and the lowest detectable limit 012 pgml in undilutedplasma The intra-assay coefficient of variation was 59and the inter-assay coefficient of variation was 126

Urinary Monocyte Chemotactic Protein-1 (MCP-1)MCP1 was assayed in urine using a sandwich ELISA(Quantikinereg kit for Human MCP1 Immunoassay RampDSystems Minneapolis MN) Corrections were made forconcentration and values expressed as pg MCP1 per mgcreatinine A standard curve was generated using a fourparameter logistic curve-fit The correlation coefficient forstandards was greater than 099 and the lowest detectablelimit 07 pgml in 12 diluted urine The intra-assay coef-ficient of variation was 25 plusmn 30 and the inter-assaycoefficient of variation was 56 plusmn 42

Table 2 Body Composition Analysis by Body Impedance Spectroscopy and Skin Fold Thickness

Parameter Non-Smokers N = 24 Smokers N = 11

Body Weight (kg) 992 (915 ndash 1069) 101 (896 ndash 1124)Phase angle 570 (446 ndash 693) 512 (330 ndash 694)Vector length 227 (210 ndash 244) 202 (177 ndash 227)Fat free mass (kg) 744 (682 ndash 806) 815 (724 ndash 906)Total body water (L) 553 (509 ndash 598) 606 (541 ndash 671)Fat mass (kg) 254 (201 ndash 307) 213 (135 ndash 290)Total body potassium (mEq) 3981 (3650 ndash 4313) 4130 (3641 ndash 4618)Body cell mass (kg) 332 (304 ndash 359) 344 (303 ndash 385)Extracellular mass (kg) 413 (364 ndash 461) 469 (398 ndash 541)Intracellular water (L) 303 (277 ndash 328) 314 (277 ndash 351)Extracellular water (L) 251 (216 ndash 285) 290 (239 ndash 341)Fat free mass from Skin fold thickness (kg) 534 (502 ndash 566) 564 (516 ndash 612)

Least square mean values with 95 confidence intervals are calculated from the mixed model ANOVA model as described in methods None of the comparisons were significantly different between the groups

Statistical analysisDifferences in baseline characteristics were compared by aChi-squared or unpaired t-test To determine the relation-ship between smoking status and REE a mixed modelANOVA was used to account for correlated observationswithin subject The dependent variable was REEkg bodyweight smoking status was used as the fixed effect varia-ble and the subject number as the random effect variableClinical oxidative stress and inflammation markers wereused as covariates as noted The adjusted least squaremeans and 95 confidence intervals obtained form thisANOVA model are reported SPSS for Windows (version13 SPSS Chicago IL) was used for all analyses Two sidedp-value of less than 005 was taken as significant

ResultsThe clinical and laboratory characteristics of the studypopulation are presented in Table 1 In keeping with adiagnosis of diabetic nephropathy the population wasobese required multiple drugs for treatment of hyperten-sion and had a high prevalence of vascular disease Smok-ers and non-smokers were well matched except in thatsmokers were younger and had higher baseline glomeru-lar filtration rate

Body composition as measured by body impedance spec-troscopy and skin-fold thickness assessment was no dif-ferent between smokers and non-smokers (Table 2)

Table 3 shows markers of oxidative stress inflammationand other metabolic parameters Plasma and urinaryMDA were significantly increased in smokers reflectingthe heightened state of oxidative stress However proteinoxidation markers were similar in the two groups C-reac-tive protein was also significantly increased in smokersreflecting the inflammatory state Plasma interleukin-6was marginally higher in smokers however tumor necro-sis factor-α did not reach statistical significance UrineMCP-1 excretion was nearly equal in smokers and non-smokers Thyroid stimulating hormone reflecting theoverall metabolic state was also no different between thesmokers and non-smokers GFR was significantly higherin smokers at 467 mLmin173 m2 compared to non-smokers at 291 mLmin173 m2

Body weight is an important determinant of REE in smok-ers and non-smokers (Figure 1) For any given weightsmokers had a higher REE compared to non-smokersUnadjusted increase in REE was 326 kcalkgd in smokers

Table 3 Comparison of REE Inflammation and Oxidative Stress Markers in Smokers and Non-smokers

Parameter Non-smokers Smokers Difference (Smokers ndash non-smokers)

REEkg body wt 210 (197 ndash 223) 243 (223 ndash 262) 326 (087 ndash 56) 0009Respiratory Quotient 085 (083 ndash 087) 081 (078 ndash 085) NSOxidative Stress MakersPlasma malondialdehyde (microML) 094 (079 ndash 108) 120 (098 ndash 141) 026 (00 ndash 052) 005Urine malondialdehydeurine creatinine ratio (microMg creatinine)

336 (244 ndash 428) 633 (495 ndash 772) 297 (131 ndash 464) 0001

Plasma total carbonyl (DU) 0818 (0728 ndash 0908) 0791 (0658 ndash 0924) NSUrine total carbonyl (DU) 0454 (0336 ndash 0571) 0444 (0270 ndash 0617) NSPlasma albumin carbonyl (DU) 1425 (1278 ndash 1572) 1441 (1225 ndash 1658) NSUrine albumin carbonyl (DU) 0598 (0438 ndash 0758) 0591 (0356 ndash 0827) NSInflammation MarkersWhite blood cell count (numbermicroL) 7897 (6885 ndash 8908) 9139 (7647 ndash 10631) NSC-reactive protein (mgdL) 080 (037 ndash 123) 177 (113 ndash 240) 097 (021 ndash 174) 0014Interleukin-6 (pgmL) 213 (148 ndash 278) 328 (231 ndash 425) 115 (-002 ndash 232) 0053Tumor necrosis factor-α (pgmL) 405 (305 ndash 505) 389 (240 ndash 538) NSMCP-1Creatinine (pgmg) 398 (288 ndash 508) 399 (228 ndash 569) NSOther variablesHbA1C () 746 (664 ndash 827) 873 (752 ndash 994) NSPlasma glucose (mgdL) 132 (110 ndash 153) 168 (136 ndash 200) NSProtein nitrogen appearance rate (gkgd) 076 (068 ndash 083) 081 (071 ndash 092) NSIothalamate clearance (mLmin173 m2) 291 (196 ndash 386) 467 (327 ndash 606) 176 (07 ndash 346) 004224 hour urine protein (mgd) 3384 (2180 ndash 4588) 4189 (2410 ndash 5967) NSThyroid stimulating hormone 232 (187 ndash 278) 204 (137 ndash 271) NS

Values reported are least square mean values with 95 confidence intervals are calculated from the mixed model ANOVA model as described in methods DU stands for densitometric units REE for resting energy expenditure MCP-1 for monocyte chemotactic protein-1

compared to non-smokers (p = 0009) When adjusted forbaseline differences in age GFR the oxidative stress mark-ers MDA in urine and plasma and C-reactive protein didnot diminish the elevated REE attributable to smoking(Table 4)

There was no independent relationship between MDAcreatinine ratio and REE (partial correlation coefficient =0163 (p = 019)) or between CRP and REE (partial corre-lation coefficient = 0177 (p = 016)) after accounting forcorrelated measurements None of the other markers ofinflammation or oxidative stress achieved statistical sig-nificance

DiscussionThe major finding of this study is that patients with dia-betic nephropathy who smoke have an elevated REE afteraccounting for body weight ndash the strongest single predic-tor of REE [25-27] Smokers had a heightened state of oxi-dative stress and inflammation However neitherbiomarkers of oxidative stress nor inflammation had anindependent relationship with REE Furthermore theassociation of smoking and elevated REE was notremoved when markers of oxidative stress and inflamma-tion were accounted for strengthening the possibility that

oxidative stress and inflammation markers by themselvesare insufficient to account for elevated REE in smokerswith diabetic nephropathy

That oxidative stress by itself has little role in elevatingREE has been demonstrated in the elderly without kidneydisease [10] Our study supports these observations andextends this to patients with CKD

The relationships between inflammation creatinine clear-ance and REE have been analyzed by Avesani et al in across-sectional study of in 91 non-dialyzed patients withCKD [9] Consideration of kidney function is importantfor evaluating REE because the oxygen consumption ofthe kidneys is high ndash estimated to be between 56 and20 of the total metabolic rate [2829] ndash and failing kid-neys may have reduced oxygen consumption contributingto reduced REE [2930] REE may also decrease as an adap-tation to the decreased energy intake or accumulation ofuremic toxins that depresses metabolic activity Howeverstudies are not conclusive regarding this observationWhereas some studies have shown decreased [3132] REEothers have shown no change [33] or even increased [34]REE in CKD Avesani et al did not find any relationshipbetween quartiles of creatinine clearance and REE but they

Resting energy expenditure is plotted against body weightFigure 1Resting energy expenditure is plotted against body weight A linear relationship is seen for smokers and non-smokers Inset shows that the mean unadjusted REE in smokers is higher by 326 kcalkgday (p = 0009)

50 70 90 110 130

Non Smokers

Smokers

Body Weight (kg)

Resting E

xpenditure

(kcalday)

Non-smoker Smoker

p = 0009

calkgd

were amongst the first to propose inflammation as a causeof elevated REE in CKD [35] Those in the highest tertileof CRP ndash and accordingly most inflamed ndash had an unad-justed REE that was higher compared to the lower tertilesof CRP Consideration of smoking status as a cause ofraised REE has not been reported in any of the above stud-ies It is possible that the association of increased CRPwith REE may simply be due to overrepresentation bysmokers in the highest tertile of CRP Our study suggestsconsideration of smoking as a covariate in analyses of REEin patients with CKD

Given that GFR can have a variable influence on REE asnoted above it is important to account for the effect ofGFR in studies dealing with REE in patients with CKDSmokers in this study had a statistically increased GFRcompared to non-smokers which is similar to what hasbeen reported for a large population based study fromFrance [36] However adjustment for GFR did not removethe effect of smoking to elevate REE Other adjustmentssuch as that for age CRP urine and plasma MDA also didnot remove the association of increased REE with smok-ing Thus other unmeasured factors may account for theelevated REE in smokers

The effect of cigarette smoking on REE are thought to beacute and transient and related to sympathetic activation[15] It should be noted that this study did not examinethe acute effect of smoking rather the chronic effect ofsmoking The large effect of smoking on REE was some-what unexpected In one epidemiologic study womenwho smoked had a statistically significant increase in REE[16] The magnitude of the effect of smoking was 68 kcald which was approximately one-third of the effect sizereported in the present study [16] In another study blackwomen who smoked were found to have a greater base-line REE compared to white women [37] Specifically

amongst black women heavy smoking was associatedwith 218 kcald higher REE and moderate smoking with177 kcald higher REE compared to non-smokersAlthough studies on the effect of smoking in diabeticnephropathy or those with CKD have not been reportedthe results of this study are in line with those reported byothers in populations without CKD [1637] and extendthese observations to patients with CKD Exactly howchronic smoking leads to increased REE is not clear frompresent investigations [38]

There are some limitations of this study First the samplesize was relatively small and limited to men Neverthelessrepeated observations in the same patient improved theprecision of the measurements and demonstrated a highlysignificant difference between smokers and non-smokersSecond diabetes control was less than optimal in somepatients that can elevate resting energy expenditure byitself However adjustment for diabetes control did notmitigate the significance of the results Third we did notquantify the amount and duration of smoking that mayhave an independent effect on REE [37] Finally thesedata may not be applicable to women

In summary this study demonstrates higher REE in smok-ers with diabetic nephropathy after adjusting for bodyweight These findings generate the hypothesis that smok-ers with diabetic nephropathy would be most prone toprotein-energy malnutrition at the onset of ESRD and addto other health-risks of smoking in this popula-tion[31333439] Aside from increasing REE smoking isknown to suppress appetite which given the anorexicstate of uremia can produce malnutrition by two ways ndashlowering intake and increasing energy expenditure Smok-ing should be considered as a cause of increased REE inpatients with CKD when analyzing the effect of other var-iables such as inflammation and GFR Longitudinal stud-

Table 4 Increase in resting energy expenditure in unadjusted and adjusted models associated with smoking

Adjusted variable Resting Energy Expenditure (kcalkgd)

95 confidence interval of the estimate

P value

None (Unadjusted model) 326 087 ndash 56 0009Age 370 105 ndash 635 0008Age + GFR 367 094 ndash 640 001Age + GFR + Urine MDACr 395 106 ndash 684 0009Age + GFR + Urine MDACr + CRP 371 074 ndash 669 0016Age + GFR + Urine MDACr + CRP + Plasma MDA 348 057 ndash 639 0021Age + GFR + Urine MDACr + CRP + Plasma MDA + HbA1C 351 059 ndash 645 0020

Adjustment for covariates are evaluated at the means which were age = 667 years GFR = 356 mLmin173 m2 Urine MDACr = 43 micromolg Plasma MDA 103 micromolL CRP = 100 mgdL HbA1C = 778

ies are required to better define the complex relationshipbetween REE falling renal function oxidative stress andinflammation in patients with CKD

Competing interestsThe author(s) declare that they have no competing inter-ests

Authors contributionsThe author designed and secured funding for the studytrained the study personnel in the collection of the datarecruited the patients and finally analyzed the data andwrote the manuscript

AcknowledgementsThe grant support of Takeda Pharma North America and assistance of Roudebush VA Clinical Chemistry and Microbiology Lab Indianapolis for CRP and TSH assay is gratefully acknowledged The author thanks Nadine G Sachs for the nursing assistance Shawn D Chase and Nicole L Knipe for technical assistance Raji Bala Clinical Chemistry supervisor for facilitat-ing the clinical assays and all the patients who gratefully agreed to partici-pate in the research study

References1 Group PDS Adequacy of dialysis and nutrition in continuous perito-

neal dialysis association with clinical outcomes Canada-USA(CANUSA) J Am Soc Nephrol 1996 7198-207 [httpPM8785388]

2 Stenvinkel P Heimburger O Paultre F Diczfalusy U Wang T Ber-glund L Jogestrand T Strong association between malnutri-tion inflammation and atherosclerosis in chronic renalfailure Kidney Int 1999 551899-1911

3 Kopple JD Greene T Chumlea WC Hollinger D Maroni BJ MerrillD Scherch LK Schulman G Wang SR Zimmer GS Relationshipbetween nutritional status and the glomerular filtrationrate results from the MDRD study Kidney Int 2000571688-1703

4 Lusvarghi E Fantuzzi AL Medici G Barbi L DAmelio A Natural his-tory of nutrition in chronic renal failure Nephrol Dial Transplant1996 11 Suppl 975-84

5 Caravaca F Arrobas M Pizarro JL Sanchez-Casado E Uraemicsymptoms nutritional status and renal function in pre-dialy-sis end-stage renal failure patients Nephrol Dial Transplant 200116776-782

6 Pollock CA Ibels LS Zhu FY Warnant M Caterson RJ Waugh DAMahony JF Protein intake in renal disease J Am Soc Nephrol1997 8777-783

7 Heimburger O Qureshi AR Blaner WS Berglund L Stenvinkel PHand-grip muscle strength lean body mass and plasma pro-teins as markers of nutritional status in patients with chronicrenal failure close to start of dialysis therapy Am J Kidney Dis2000 361213-1225

8 Ikizler TA Greene JH Wingard RL Parker RA Hakim RM Sponta-neous dietary protein intake during progression of chronicrenal failure J Am Soc Nephrol 1995 61386-1391

9 Avesani CM Draibe SA Kamimura MA Colugnati FA Cuppari LResting energy expenditure of chronic kidney diseasepatients influence of renal function and subclinical inflam-mation Am J Kidney Dis 2004 441008-1016

10 Bell C Jones PP Seals DR Oxidative stress does not modulatemetabolic rate or skeletal muscle sympathetic activity withprimary aging in adult humans J Clin Endocrinol Metab 2003884950-4954

11 Cai H Harrison DG Endothelial dysfunction in cardiovasculardiseases the role of oxidant stress Circ Res 2000 87840-844

12 Raj DS Lim G Levi M Qualls C Jain SK Advanced glycation endproducts and oxidative stress are increased in chronic allo-graft nephropathy Am J Kidney Dis 2004 43154-160

13 Himmelfarb J Stenvinkel P Ikizler TA Hakim RM The elephant inuremia oxidant stress as a unifying concept of cardiovascu-lar disease in uremia Kidney Int 2002 621524-1538

14 Agarwal R Chronic kidney disease is associated with oxidativestress independent of hypertension Clin Nephrol 200461377-383

15 Perkins KA Metabolic effects of cigarette smoking J Appl Phys-iol 1992 72401-409

16 Kimm SY Glynn NW Aston CE Poehlman ET Daniels SR Effectsof race cigarette smoking and use of contraceptive medica-tions on resting energy expenditure in young women Am JEpidemiol 2001 154718-724

17 Smoking cessation and severity of weight gain N Engl J Med1991 325517-518

18 Agarwal R Ambulatory GFR measurement with cold iothala-mate in adults with chronic kidney disease Am J Kidney Dis2003 41752-759

19 Agarwal R Vasavada N Chase SD Liquid chromatography foriothalamate in biological samples J Chromatogr B Analyt TechnolBiomed Life Sci 2003 785345-352

20 Weir JB New methods for calculating metabolic rate withspecial reference to protein metabolism J Physiol 19491091-9

21 Durnin JV Womersley J Body fat assessed from total body den-sity and its estimation from skinfold thickness measure-ments on 481 men and women aged from 16 to 72 years BrJ Nutr 1974 3277-97

22 Maroni BJ Steinman TI Mitch WE A method for estimatingnitrogen intake of patients with chronic renal failure KidneyInt 1985 2758-65

23 Agarwal R Chase SD Rapid fluorimetric-liquid chromato-graphic determination of malondialdehyde in biological sam-ples J Chromatogr B Analyt Technol Biomed Life Sci 2002 775121-126

24 Agarwal R Proinflammatory effects of oxidative stress inchronic kidney disease role of additional angiotensin IIblockade Am J Physiol Renal Physiol 2003 284F863-F869

25 Nelson KM Weinsier RL Long CL Schutz Y Prediction of restingenergy expenditure from fat-free mass and fat mass Am J ClinNutr 1992 56848-856

26 Leibel RL Rosenbaum M Hirsch J Changes in energy expendi-ture resulting from altered body weight N Engl J Med 1995332621-628

27 James WP Davies HL Bailes J Dauncey MJ Elevated metabolicrates in obesity Lancet 1978 11122-1125

28 Anonymous Comorbid conditions and correlations with mor-tality risk among 3399 incident hemodialysis patients Am JKidney Dis 1992 2032-38

29 Goldwasser P Mittman N Antignani A Burrell D Michel MA CollierJ Avram MM Predictors of mortality in hemodialysis patientsJ Am Soc Nephrol 1993 31613-1622

30 Panesar A Agarwal R Resting energy expenditure in chronickidney disease relationship with glomerular filtration rateClin Nephrol 2003 59360-366

31 Bucciante G Senesi G Piva M Donato D Velluti F Milito F SpisaniC Bernardi A Resting metabolic rate by indirect calorimetryin uremic patients Contrib Nephrol 1990 81214-219

32 Avesani CM Draibe SA Kamimura MA Dalboni MA Colugnati FACuppari L Decreased resting energy expenditure in non-dia-lysed chronic kidney disease patients Nephrol Dial Transplant2004 193091-3097

33 Schneeweiss B Graninger W Stockenhuber F Druml W Ferenci PEichinger S Grimm G Laggner AN Lenz K Energy metabolism inacute and chronic renal failure Am J Clin Nutr 1990 52596-601

34 Kuhlmann U Schwickardi M Trebst R Lange H Resting metabolicrate in chronic renal failure J Ren Nutr 2001 11202-206

35 Marcussen N Schumann J Campbell P Kjellstrand C Cytodiagnos-tic urinalysis is very useful in the differential diagnosis ofacute renal failure and can predict the severity Ren Fail 199517721-729

36 Halimi JM Giraudeau B Vol S Caces E Nivet H Lebranchu Y TichetJ Effects of current smoking and smoking discontinuation onrenal function and proteinuria in the general population [InProcess Citation] Kidney Int 2000 581285-1292

37 Clemens LH Klesges RC Slawson DL Bush AJ Cigarette smokingis associated with energy balance in premenopausal African-

Publish with BioMed Central and every scientist can read your work free of charge

BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime

Sir Paul Nurse Cancer Research UK

Your research papers will be

available free of charge to the entire biomedical community

peer reviewed and published immediately upon acceptance

cited in PubMed and archived on PubMed Central

yours mdash you keep the copyright

Submit your manuscript herehttpwwwbiomedcentralcominfopublishing_advasp

BioMedcentral

American adult women differently than in similarly agedwhite women Int J Obes Relat Metab Disord 2003 271219-1226

38 Warwick PM Baines J Energy expenditure in free-living smok-ers and nonsmokers comparison between factorial intake-balance and doubly labeled water measures Am J Clin Nutr1996 6315-21

39 Avesani CM Cuppari L Silva AC Sigulem DM Cendoroglo M SessoR Draibe SA Resting energy expenditure in pre-dialysis dia-betic patients Nephrol Dial Transplant 2001 16556-565

Pre-publication historyThe pre-publication history for this paper can be accessedhere

httpwwwbiomedcentralcom1471-2369613prepub

Abstract

Background

Methods

Results

Conclusion

Background

Methods

Subjects

Glomerular filtration rate

Resting energy expenditure

Lean body mass

Anthropometric method

Body impedance spectroscopy

Protein nitrogen appearance

Oxidative stress markers

Malondialdehyde (MDA)

Carbonyl measurements

Inflammation markers

Total leukocyte count (WBC count)

C-reactive Protein (CRP)

Interleukin 6 (IL-6)

Tumor Necrosis Factor-a (TNF-a)

Urinary Monocyte Chemotactic Protein-1 (MCP-1)

Statistical analysis

Results

Discussion

Competing interests

Authors contributions

Acknowledgements

References

Pre-publication history