Smoking as risk factor for chronic kidney disease : systematic review

Chronic kidney disease (CKD) and smoking are public health issues.1,2 The disturbances introduced in the health care system related to CKD stem from the increasing incidence of this condition and the significant costs involved in the treatment of individuals affected by it.3 Although more resources have been made Introduction: Chronic kidney disease (CKD) and smoking are public health problems. Objective: To assess smoking as a risk factor for progression of CKD. Methods: We conducted a systematic review in Medline, LILACS, SciELO, Google Scholar, Embase and Trials.gov with articles published until February/2013. Were included: cohort, clinical trials and case-control. Performed in humans, aged ≥ 18 years with smoking as a risk factor for progression of CKD. We excluded studies that reported no smoking and CKD in the title or had proposed to reduce smoking. Results: Among 94 citations, 12 articles were selected. Of these, six were multicenter conducted in developed countries, four were randomized. Males predominated 51-76%. There was associated with smoking progression in 11 studies. It was found that the consumption ≥ 15 packs/ year increases the risk of progression of CKD. Conclusion: Smoking is a risk factor for progression of CKD. AbstrAct


IntroductIon
Chronic kidney disease (CKD) and smoking are public health issues. 1,2he disturbances introduced in the health care system related to CKD stem from the increasing incidence of this condition and the significant costs involved in the treatment of individuals affected by it. 3Although more resources have been made available to treat patients with CKD, the conservative management of this disease still leads to high rates of morbidity and mortality. 4obacco smoke contains over 4,000 particles and gases, some of which nephrotoxic. 5Particles include heavy metals known to cause tubular injury such as cadmium and lead, which may reach serum concentrations above 40% in smokers. 6The action of nicotine on specific cholinergic receptors causes hemodynamic changes such as elevations in blood pressure, heart rate and peripheral vascular resistance. 7KD patients are subject to the deleterious effects of inflammation, oxidative stress and uremic toxins, which combined lead to rates of cardiovascular death 20-30 times greater than the general population. 8dditionally, smoking increases the incidence of heart failure, peripheral vascular disease, and death in patients with CKD. 9 In spite of these shocking evidences, few studies have assessed smoking as a risk factor for CKD.Two systematic reviews have been published to date on the matter.The first, by Burton-Jones et al., 10 (2007) reviewed the incidence of CKD in smokers.The second systematic review was published by Noborisaka 11 (2013) to look into the development of CKD in healthy populations and smokers.In Brazil, this issue was discussed in 2002 in a narrative review by Oliveira et al., 12 in which the knowledge then available on smoking as a risk factor for kidney disease was summarized.The present review 13 aims to examine smoking as a risk factor for progression of CKD.

SeARch StRAtegy
A systematic search for literature took place between November of 2011 and February of 2013.The following databases were used in this process: Medical Literature Analysis and Retrieval Online (Medline), Literature in the Health Sciences in Latin America and the Caribbean (LILACS), Google Scholar, Scientific Electronic Library Online (SciELO), EMBASE, and Trials.gov.A specific strategy was devised for each search portal to cross-reference descriptors and keywords in order to facilitate the retrieval of papers from the literature.
Medline was accessed via search portal Pubmed, using the following syntax: (("kidney failure, chronic" [Majr]) AND/OR ("renal insufficiency, chronic" [Majr]) AND "smoking" [Majr]).Descriptors "tabagismo" AND "doença renal crônica" AND "insuficiência renal crônica" were used in LILACS and SciELO.In Embase, the sarch was made via portal Evidence Based Medicine using the following syntax: 'smoking'/ exp/mj AND 'chronickidneydisease'/exp/mj.In Trials.gov, the syntax category was chronic kidney failure -> refine results -> smoking.In Google Scholar, the syntax was "allintitle kidney chronic smoking insufficiency OR failure OR disease." The resulting list of papers was then submitted to eligibility, inclusion, and exclusion criteria.Original articles in which smoking was deemed as a risk factor for chronic kidney disease, without linguistic or time restrictions were considered eligible.The PICOS (Population, Intervention, Comparison, Outcome, Study Design) search strategy was used. 10Selected papers included: (1) prospective/retrospective cohort studies, randomized and non-randomized clinical trials, and case-control studies; (2) studies in which smoking was considered a risk factor for CKD progression; (3) studies in humans aged 18 years and older; (4) studies in which CKD patients were offered conservative treatment, studies enrolling smokers with or without CKD who progressed to CKD as a result of smoking, and (5) studies comparing the outcomes of smokers and nonsmokers with CKD.Papers with no reference in their titles to smoking and CKD or studies proposing measures to combat smoking were excluded.

dAtA AnAlysIs
Paper selection was carried out in three steps.In the first step, independent researchers VML and MAL read the papers blinded for their authors and journals on which they were published.Disagreements were resolved by consensus or, when needed, with the aid of a third researcher (UFEJ) called in to break the tie.
Papers failing to meet the inclusion criteria were excluded.In the second step, the independent researchers read the abstracts from the studies selected in the first step and excluded additional articles that did not meet the inclusion criteria.In the third step, the studies still remaining from the first two steps were read in full and selected to be included in this review (Figure 1).

results
Twelve of 94 papers were deemed potentially relevant, and were retrieved with their texts in full to be included in the review.On Medline, 27 of the 37 papers found via Pubmed were excluded based on their titles or due to duplicated occurrence of the reference.Another three were excluded after their abstracts were read, and seven were read in full.The two citations found on LILACS were excluded because of their titles.The six listed on SciELO were ruled out for the same reason.Thirteen of the papers listed on Embase were excluded because of their titles or due to duplicated occurrence of the reference.Seventeen of the 33 citations from Google Scholar were excluded because of their titles or due to reference duplicates, ten were left out after their abstracts were read, one after it was red in full, and four were included in the review.
The three references found on Trials.govwere excluded because of their titles.
Chart 1 shows the data and the results of the 12 included studies and reveals that the first study on smoking as a risk factor for progression of CKD was published in 1998.9][20][21] No papers from Latin America were included.Four were randomized controlled trials. 14,17,18,22ale patients were predominantly enrolled in the included studies, and accounted for 51%-77% of the subjects.Six studies [20][21][22][23][24][25] reported increased rates of CKD progression among smokers.Differently from others, one paper identified a transient increase of 0.32 ml/min/1.73m 2 /year in the glomerular filtration rate (GFR) of female smokers versus female nonsmokers.However, this difference disappeared seven years after they started smoking.This change in GFR was not statistically significant in males. 26Five studies [15][16][17][18]27 reported a dose-effect relationship in which higher risk of CKD progression was associated with heavier smoking, particularly when individuals smoked more than 15 packs of cigarettes a year.

dIscussIon the RelAtionShip between Smoking And ckd
The perception that smoking may cause renal injury is recent in the history of nephrology and, therefore, deserves some contextualization.The first causal association between smoking and CKD was discussed by Orth et al. 16 in 1998, with the publication of a multicenter trial involving nine research centers.This study looked into the outcomes of patients with primary IgA nephropathy (inflammatory model of CKD) and subjects with autosomal dominant polycystic kidney disease (non-inflammatory genetic model) all the way to stage-5 CKD.The authors concluded that male smokers with IgA nephropathy or autosomal dominant polycystic kidney disease were at a higher risk of progressing to stage-5 CKD when compared to nonsmoking patients. 16The United States Renal Data Systems (USRDS) published a paper that same year alerting the scientific community to the growth of a global epidemic of CKD. 25 Since this publication, expert committees from around the world have begun to scale up the attention given to the risk factors associated with CKD progression.In 2002, the guidelines from the National Kidney Foundation 1 already considered smoking as a factor in CKD progression.
This review found five case-control [15][16][17][18]25 and seven cohort studies 14,[20][21][22][23]26,28 with the analytical robustness required to produce risk assessments. Noetheless, the subjects in the case and control groups of the case-control studies were not given equal odds [15][16][17]21,24,26,28 , as they were not randomized.Additionally, most of the studies enrolled patients with established kidney disease, indicating the need for a prospective cohort study with CKD-free individuals.Two studies were designed with this method in mind, 21,28 but their outcomes are questionable due to methodological flaws.
In 2011, Hallan & Orth 21 carried out a study on CKD and smoking based on the population of the HUNT-2 cross-sectional study.However, the HUNT-2 study was not conceived to assess the habit of smoking.Instead, the Norwegian study was designed to evaluate the status of  public health issues such as cardiovascular disease (CVD), diabetes mellitus (DM), chronic obstructive pulmonary disease (COPD), osteoporosis, and mental health.As a consequence of methodological carelessness, patients with multiple cardiovascular comorbidities were included in the study on smoking and CKD, 21 a decision that introduced selection and survival biases in the studied sample of the population.

chArt 1 chARActeRiSticS of obSeRvAtionAl StudieS And ASSociAtion between chRonic kidney diSeASe And Smoking
In 2013, Noborisaka et al. 28 evaluated 6,998 Japanese active workers without CKD, although with the following selection biases: (1) subjects who reported to have stopped smoking at baseline were categorized as former smokers; (2) the time for which the subjects had smoked was not accounted for; (3) patients with CKD were categorized from a single evaluation; 29 additionally, the KDIGO 29 was reported to have been the basis for categorization but its criteria were not followed; (4) proteinuria (indicative of kidney damage) was ruled out from the examination of a single sample tested with a reagent strip, a method subject to selection and information biases.

the countRieS of oRigin of the StudieS
This review predominantly included studies carried out in developed nations, possibly because of the greater number of epidemiologic surveillance efforts in effect in these countries and the longer life expectancy of their populations, 30 as is the case in Germany, 20 Austria, 16 Sweden, 18 the United States, 23 Norway, 26 and Japan. 28rojections for 2015 developed by the World Bank and the World Health Organization indicate that 6.4 million people will die of causes related to smoking -the equivalent to 10% of the deaths for all causes and a figure 50% greater than the number of deaths attributed to the human immunodeficiency virus (HIV). 31espite the preoccupying nature of this data and the impact smoking potentially has on the progression of CKD, no studies have been published on this matter in Latin America to date.The intersection between CKD and smoking must not be neglected.These issues combined may overwhelm health systems worldwide, particularly in poorer or developing nations such as Brazil.

contRoveRSieS oveR ckd And gendeR
The higher prevalence seen in male populations is consistent with the epidemiologic findings of other populations. 17,32For yet unknown reasons, males are at a higher risk of developing CKD. 33Evidence suggests that estrogen offers protection against renal disorders as it affects the conservation of nitric oxide -an important regulator of glomerular ultrafiltration -by modulating mesangial cells and afferent arterioles, 34 an effect not observed in postmenopausal women. 35However, data from the USRDS 36 have shown that the female population with CKD has grown between 1988 and 2010 and achieved a prevalence rate of 10.2% -versus 8.6% in males -due to increases in female life expectancy, obesity, and a greater number of cases of systemic hypertension and DM. 37,38nly the study by Kronborg et al. 26 reported improved GFR in female smokers based on the follow-up of the cohort enrolled in the Tromsø Study, 39 designed by its authors to determine high risk factors for cardiovascular death and develop strategies for the prevention of acute myocardial infarction (AMI) and stroke.However, it should be noted that the authors 26 had difficulties explaining this finding.The improvement observed in the GFR of the female patients was attributed to hyperfiltration and smoking-induced loss of muscle mass.This transient improvement disappears after seven years and the GFR then decreases progressively. 26oss of muscle mass and hyperfiltration are physiologically associated with decreases in serum creatinine, either by decreased synthesis or increased clearance.However, this explanation is far from being biologically plausible, particularly when the vascular alterations introduced by smoking are considered. 39espite the difference introduced by the transient improvement in the GFR of females, the Norwegian team 39 concluded that gender differences affect the interactions between predictors of GFR decline.Alcohol consumption at any level had a protective effect over the renal function of men, whereas in women such correlation was not statistically significant.Physical activity has been positively correlated with GFR increases in females, while DM and CVD have led to renal function deterioration in both genders. 3921,27 This finding appears to be similar to what has been reported for other diseases such as lung cancer in relation to the cumulative effects of smoking. 40 statistically significant correlation between the cumulative effects of smoking and renal function decline [15][16][17][18]21,27 has also been observed when patients with other comorbidities such as HIV 24 and lung transplant recipients were assessed, even more than six months after smoking cessation. 20 Smking cessation was assessed in ten studies included in this review and was shown to have been statistically correlated with renal function improvement.However, particularly among heavy smokers -defined as individuals who smoke more than 15 packs of cigarettes a yearthe risk of progression persisted, with a relative risk of 8.84.41 The method used in the studies that looked into the association between smoking and CKD may have introduced selection and information biases to the analysis.A great deal of these studies was designed to establish the risk factors for conditions such as severe hypertension, 22 lung transplant associated with use of calcineurin inhibitors, COPD, 20 and HIV. 24moking was one of the data points collected from enrolled individuals. Sming and CKD were not among the main objectives of these studies, although they were retrospectively identified as risk factors.[18][19][20][21]27 This review identified the need for studies to determine the pathophysiological mechanisms of kidney injury associated with smoking and the establishment of smoking as both a triggering agent and a risk factor for the progression of CKD.There still is no evidence as to whether the hemodynamic hypothesis (linked to nicotine) 8 or tubular toxicity (linked to heavy metals) 9 are the only mechanisms of renal injury.Perhaps there is a specific substance responsible for the entire process as in the bladder cancer model, in which a mutagen -3,4-benzopyrene -acts as the initiator of damage to cell DNA.42 new pAthophySiologicAl fRontieRS to explAin old hiStopAthologicAl pRoceSSeS In 2012, researchers from the University of Alabama at Birmingham (Nephrology and Pathology) showed that rats submitted to 5/6 nephrectomies had an oligomer known as α7 nicotinic acetylcholine receptor (α7nAChR) and changed the understanding of how nicotine acts on the kidneys.This protein led to increases of 100% in proteinuria but not in albuminuria, thus revealing impaired tubular reabsorption.The levels of NADPH oxidase 4 (NOX4), the main marker of oxidative stress in the renal cortex, were increased by 100%.Markers of fibrosis such as fibronectin and transforming growth factor beta (TGF-β) increased by 50% and 200%, respectively.Expression of α7nAChR was more pronounced in the proximal tubule and decreased toward the distal tubules.Levels of α7nAChR were not detected in glomeruli or in the intrarenal vasculature.The administration of methyllycaconitine (MLA), a specific α7nAChR blocker, led to reversals in proteinuria, hypertension, and expression of fibronectin, TGF-β, and NOX4 in rats submitted to 5/6 nephrectomy.
These findings have redirected research perspectives around the progression of CKD in smokers.Apparently, we are beginning to solve the puzzle of how nicotine acts on the kidneys.Experimental data indicate that nicotine activates α7nAChR in the proximal tubule and triggers the biosynthesis of profibrotic and proinflammatory cytokines. 43ccording to this theory, smoking not only causes substance production and deregulation in the endothelium, release of endothelin-1 (efferent arteriole vasoconstrictor), and decreased levels of nitric oxide (efferent arteriole vasodilator) 10 as previously thought, but it may also accelerate the progression of CKD through profibrotic and proinflammatory cytokines.Higher levels of vasoconstrictors would consequently cause intrarenal hypoxia, neovascularization, and formation of microaneurysms.The mesangial cells targeted by these cytokines would produce too much mesangial matrix, leading to the formation of focal nodules similar to what is found in diabetic nephropathy. 44f the theory involving proximal tubule inflammation and α7nAChR 43 is proven valid in humans, Science will be able to explain the findings identified in the study of idiopathic nodular glomerulosclerosis associated with smoking and hypertension. 44erhaps in the coming years the discussion of the toxic inflammatory effects of nicotine on the proximal tubule might guide further research for a selective receptor blocker or renal nicotineacetylcholine receptors.In the future, it will be possible to stabilize or reverse the progression of CKD with drug therapy.

Smoking and clinical practice
Regardless of the mechanism of renal injury, the evidence derived from observational studies indicates that CKD patients with comorbidities (primary glomerulonephritis, 16 COPD, 20 hypertension, 25 HIV, 24 proteinuria 28 ) tend to experience faster declines in GFR and increased risk of death. 44,45The evidence on CKD progression, mortality, tobacco dose-effect, and renal function deterioration leaves no doubt over the need to combat smoking.The main tool available for smoking cessation is a combination between treatment with a multidisciplinary team and drug therapy.
In our opinion, antismoking medications such as bupropion hydrochloride (Bup ® ) and varenicline tartrate (Champix ® ) combined with chewing gum and nicotine patches (Niquitin ® ) should be added to the therapeutic arsenal of nephrologists.

a new perSpective in nephrology
In our view, physicians -nephrologists in particular -should change their approach to smoking.Smoking needs to be seen as a public health issue, as the treatment of smokers does not lie only in the hands of psychiatrists and pulmonologists.We need to intervene upon this modifiable element in the progression of CKD.To that end, the following are needed: 1) guidelines and refresher courses to train nephrologists on how to combat smoking; 2) change our perspectives in relation to smoking; 3) offer antismoking medication in CKD units; 4) address smoking cessation responsibly during patient visits to the CKD unit.
In short, smoking cessation should be valued and treated in the same manner in which one deals with other renal disease progression factors such as anemia, hyperphosphatemia, hypertension, proteinuria, and secondary hyperparathyroidism.

conclusIon
This systematic review revealed a correlation between smoking and progression of CKD.This positive correlation became more pronounced for individuals smoking 15 of more packs of cigarettes a year.

Figure 1 .
Figure 1.Flowchart showing the steps taken in the selection of the papers included in the systematic review.Source: designed by the authors.