Serum Cystatin C and Emphysema: Results from the NationalHealth and Nutrition Examination Survey (NHANES)

Haala K. Rokadia • Shikhar Agarwal

Received: 7 July 2011 / Accepted: 10 January 2012 / Published online: 28 January 2012

� Springer Science+Business Media, LLC 2012

Abstract

Background Cystatin C (CysC) is a potent nonorgan-

specific cysteine protease inhibitor and may contribute to

elastolysis and tissue destruction by a mechanism of pro-

tease–antiprotease imbalance. Given the prevalence of

CysC in the serum of smokers and its role in tissue

destruction, we aimed to evaluate the association between

CysC and emphysema.

Methods Pooled cross-sectional data from the National

Health and Nutrition Examination Survey 1999–2002 were

used. Emphysema and chronic bronchitis were defined by a

self-reported history ascertained using standardized ques-

tionnaires. Active smokers were defined as self-reported

current smokers or measured serum cotinine C10 ng/mL.

Nonactive smokers with a serum cotinine level [0.05 ng/

mL were defined as environmental tobacco smoke (ETS)-

exposed.

Results The prevalence (95% CI) of emphysema was

1.3% (range = 0.9–1.8%). The mean (SE) CysC level in

the emphysema group was significantly higher than in

normal controls [1,139 (22) vs. 883 (8) lg/L; p = 0.001].

Upon stratification of the study population by C-reactive

protein (CRP) concentrations, we demonstrated a progres-

sive increase in the mean serum CysC level with serially

increasing CRP concentrations. Active smokers with

emphysema had 115.4 (46.5) lg/L higher mean (SE) CysC

levels than the normal controls (p \ 0.001). Upon adjusted

analysis, we observed that nonactive smokers with signif-

icant ETS exposure had 31.2 (15.2) lg/L higher mean (SE)

serum CysC levels as compared to ETS unexposed non-

active smokers (p = 0.04).

Conclusion In a large representative noninstitutionalized

US population, we demonstrated an association between

emphysema and serum CysC. Active smokers with

emphysema had significantly higher CysC levels. These

findings suggest that CysC may play a role in the patho-

genesis of smoking-related emphysema.

Keywords Cystatin C � Emphysema � Chronic bronchitis �Chronic obstructive pulmonary disease � Smoking �National

Health and Nutrition Examination Survey

Introduction

Chronic obstructive pulmonary disease (COPD) is char-

acterized by significant airflow limitation and is an

important cause of lung-related morbidity and mortality.

COPD represents a group of chronic inflammatory dis-

eases, including chronic bronchitis and emphysema. The

histological hallmark of COPD is the presence of inflam-

matory cells, including neutrophils and alveolar macro-

phages, in the microvasculature, interstitium, and alveoli.

Several animal and cell culture-based studies have dem-

onstrated the role of serine and cysteine proteases in the

development of COPD [1–3]. In addition, several in vitro

studies have suggested that inhibition of these proteases

may attenuate the contribution of proteases to the patho-

genesis of COPD [4]. Furthermore, proteolytic enzymes

have been demonstrated to play a role in fibrotic lung

remodeling by direct action on transforming growth factor

[5, 6].

H. K. Rokadia

Department of Internal Medicine, Cleveland Clinic, Cleveland,

OH 44195, USA

S. Agarwal (&)

Department of Cardiovascular Medicine, Cleveland Clinic,

9500 Euclid Avenue, Mail Code J2, Cleveland, OH 44195, USA

e-mail: dr.shikhar.agarwal@gmail.com

123

Lung (2012) 190:283–290

DOI 10.1007/s00408-012-9374-z

Cathepsins are a family of cysteine proteases that are

responsible for macrophage-mediated extracellular matrix

degeneration [7]. Cathepsins B, H, L, and S have been

implicated in the pathogenesis of COPD. These cathepsins

are ubiquitous proteins secreted by inflammatory cells,

including neutrophils and alveolar macrophages. The

activity of these potent proteolytic enzymes is typically

inhibited by the cystatin family of proteins. Cystatin C

(CysC) is the most widespread and potent inhibitor of

cathepsins B and L, which are involved primarily in lung

tissue destruction [3, 8]. CysC binds to cathepsins in a

competitive fashion using noncovalent binding and results

in formation of inactive protease–cystatin complexes.

CysC is a nonorgan-specific cysteine protease inhibitor

that is secreted into the bloodstream by inflammatory cells,

including alveolar macrophages [9]. Although cathepsins

are measurable in the serum, the serum levels may not be

representative of the parenchymal cathepsin activity due to

primarily a local tissue action of these proteins [10, 11]. In

contrast, cystatins are secreted into the bloodstream and the

levels of these proteins may indirectly yet reliably predict

cathepsin activity.

CysC has been extensively studied in kidney disease and

cardiovascular disease and has been shown to be valuable

in assessing renal function and predicting cardiovascular

mortality, particularly in the elderly [12, 13]. Elevated

serum CysC has been associated with smoking, increased

age, male gender, C-reactive protein (CRP), and cardio-

vascular risk factors such as hypertension, low high-density

lipoprotein cholesterol and elevated body mass index [14].

It has been suggested that CysC may more accurately

estimate glomerular filtration rate (GFR) than creatinine,

particularly in the setting of the elderly and those with mild

renal insufficiency [15, 16]. Furthermore, it has been pos-

tulated that CysC plays a role in pathogenesis of malig-

nancy and metastasis [17–21]. Cathepsins B and L and

their potent inhibitor CysC have been studied in the path-

ogenesis of multiple pulmonary processes, including

bronchiectasis [22], lung malignancy [23], and pleural

effusions [24].

It is suggested that a cathepsin–cystatin imbalance

contributes to tissue destruction by under-regulated prote-

ase activity [8, 25, 26]. It may be speculated that this

overactivity of proteases in the lung parenchyma upregu-

lates the formation of CysC in an attempt to attenuate lung

tissue destruction. Given the evidence of a role of CysC in

tissue destruction and its prevalence in smokers, there

might be a role of CysC in emphysema. There is a striking

paucity of data relating CysC and emphysema. With this

background, we aim to evaluate the association between

CysC and emphysema in a large representative noninsti-

tutionalized US population. Figure 1 is a causal diagram

demonstrating the relationship between CysC and

emphysema. In this figure we have considered several

different etiopathogenic pathways that may result in ele-

vation of CysC in subjects with emphysema.

Methods

Study Population

This study analyzed pooled data from the 1999–2000 and

2001–2002 National Health and Nutrition Examination

Surveys (NHANES) [27]. NHANES is an on-going cross-

sectional survey of the civilian, noninstitutionalized US

population designed to provide a representative sample

from which to make national estimates on health and

nutritional status.

Exposure and Outcome Assessment

Cigarette smoking status was determined using serum

cotinine measurements along with the questionnaire items,

including ‘‘Have you smoked at least 100 cigarettes in your

entire life?’’ and ‘‘Do you now smoke cigarettes?’’ In the

NHANES, serum cotinine was measured using isotope

dilution, high-performance liquid chromatography/atmo-

spheric pressure chemical ionization tandem mass spec-

trometry. Active smokers were defined as self-reported

cigarette smokers or those with a measured serum cotinine

C10 ng/mL. All nonactive smokers by self-report were

classified as former smokers and never smokers based on

survey question responses. All nonactive smokers were

Fig. 1 Causal diagram representing the relationship between cystatin

C, smoking, and emphysema. The possible pathways that may result in

elevation of serum cystatin C are demonstrated using dashed arrows.

This diagram demonstrates that cystatin C may be a partial mediator

in the association between cigarette smoking and emphysema. HTNhypertension, DM diabetes mellitus, CKD chronic kidney disease,

CAD coronary artery disease, CHF congestive heart failure

284 Lung (2012) 190:283–290

123

subcategorized into those with elevated cotinine levels

([0.05 ng/mL) and those with nonelevated cotinine levels.

Nonactive smokers with elevated cotinine levels were

defined as environmental tobacco smoke (ETS)-exposed.

CysC was assayed from stored serum samples from

NHANES 1999–2000 and 2001–2002 from all participants

aged 60 years or more and 25% random sampling of par-

ticipants aged 12–59 years. Samples were assayed using an

automated particle-enhanced nephelometric assay (N Latex

Cystatin C; Dade Behring, Deerfield, IL). The assay,

ranging from 0.23 to 7.25 mg/dl, is currently the most

precise automated assay across the clinical concentration

range.

The study outcomes of emphysema or chronic bronchitis

were determined historically using standardized question-

naires. Patients that reported only a history of asthma or

symptoms of asthma were not included.

Measurement of Confounders and Mediators

CysC has been associated with cardiovascular disease and

chronic kidney disease (CKD) in several studies [12–16].

Due to clustering of chronic diseases like cardiovascular

disease and CKD with COPD by virtue of common risk

factors, we adjusted for the presence of cardiovascular

disease, congestive heart failure, stroke, CKD, and tradi-

tional cardiovascular risk factors, including age, gender,

race, hypertension, hyperlipidemia, body mass index, and

diabetes. Since CRP has been implicated in the causation

of cardiovascular disease, we included it in our regression

modeling in order to serve two independent purposes. First,

it provided conservative estimates of association between

CysC and emphysema. Second, it helped elucidate if there

was a possibility of a noninflammatory mediated pathway

in the pathogenesis of emphysema.

Hypertension was defined as systolic blood pressure

[140 mmHg, diastolic blood pressure [90 mmHg, self-

reported diagnosis of hypertension by a physician, or self-

reported use of antihypertensive medication. Hyperlipidemia

was defined as a total blood cholesterol [240 mg/dl, self-

reported diagnosis of hyperlipidemia by a physician, or self-

reported use of cholesterol-lowering medication. Diabetes

was defined as self-reported diagnosis of diabetes by a phy-

sician or use of insulin or oral hypoglycemic medication.

GFR was estimated using the modification of diet in

renal disease study formula [28, 29] based on age, serum

creatinine, race, and gender. Serum creatinine was stan-

dardized across surveys based on previously published

calibration equations [30]. CKD was defined as an esti-

mated GFR \60 mL/min/1.73 m2. Body mass index was

calculated from self-reported current height and weight.

CRP was measured on stored venipuncture samples by

latex-enhanced nephelometry with results reported within

the range of 0.01–18.5 mg/dl.

Statistical Analysis

Statistical analysis was performed using Stata v10.0 (Stata

Corp., College Station, TX, USA). Data from NHANES

1999–2000 and 2001–2002 surveys were pooled using

standard methods and, subsequently, 4 year combined

weights were calculated. Survey statistics traditionally used

to analyze complex semirandom survey designs were

employed to analyze these data. Multivariate linear

regression analysis was performed with CysC as the out-

come measure in the regression models to obtain adjusted-

effect estimates and their 95% confidence intervals (CI)

after accounting for the above-mentioned confounders.

Results

The prevalence (95% CI) of emphysema and chronic bron-

chitis in the US population was estimated as 1.3%

(0.9–1.8%) and 5.5% (4.2–6.7%), respectively. Table 1

gives the demographics and clinical characteristics of the

study population. We observed that the subjects with self-

reported emphysema were significantly older than the nor-

mal controls (p \ 0.001). The chronic bronchitis group had

a female (p \ 0.001) and white (p = 0.03) predominance.

We observed that the emphysema group had a significantly

higher prevalence of coronary artery disease, cerebrovas-

cular disease, and CKD compared to the other groups

(p \ 0.001 for all comparisons). In addition to cardiovas-

cular disease, the emphysema group had a significantly

higher prevalence of traditional cardiovascular risk factors,

specifically hypertension (61.4%), hyperlipidemia (56.3%),

and diabetes (14.2%). In comparison to the control popula-

tion, the mean CRP was significantly higher in the emphy-

sema (p = 0.003) and chronic bronchitis (p = 0.02) groups.

There were significant differences noted in the distri-

bution of smoking characteristics in the study groups. In

comparison to the normal controls, we observed a signifi-

cantly higher prevalence of active smoking (48.5% vs.

28.5%; p \ 0.001) and former smoking (43.1% vs. 22.7%;

p \ 0.001) in the emphysema group. Only 8.4% of subjects

with self-reported emphysema were never smokers com-

pared to 48.8% of normal controls (p \ 0.001). Of the

never and former smokers (nonactive smokers), the pro-

portion (SE) of individuals with significant ETS exposure

was 48.3% (2.3%), 53.9% (3.1%), and 49.2% (2.4%)

among the normal control group, emphysema group, and

chronic bronchitis group, respectively.

Table 2 gives the distribution of mean CysC levels in

the study population. The mean (SE) CysC level in the

Lung (2012) 190:283–290 285

123

emphysema group was 1,139 (22) lg/L, which was sig-

nificantly higher than the mean CysC level in the normal

controls [883 (8) lg/L; p \ 0.001] or the chronic bronchitis

group [902 (22) lg/L; p \ 0.001]. When stratified by sex,

smoking status, cotinine levels, and CRP levels, the

emphysema group had a significantly higher mean CysC

level than the normal controls in all strata examined. In

contrast, individuals with self-reported chronic bronchitis

had mean CysC levels similar to that seen in the normal

controls. Upon stratification of the study population by

CRP levels, there was a trend toward increasing serum

CysC levels with increasing CRP concentrations. Figure 2

demonstrates the relationship between serum CysC and

CRP across 20 equal quantiles of CRP concentration. With

serially increasing CRP concentrations, we observed a

progressive increase in the mean serum CysC level.

Multivariate linear regression analysis adjusted for

demographic and clinical characteristics, including renal

function and CRP, demonstrated that subjects with

emphysema had a mean (SE) CysC level 89.4 (33.4) lg/L

higher than the normal controls (p \ 0.001) (Table 3).

Adjusted analysis with stratification of the emphysema

group by smoking status demonstrated that active smokers

with emphysema had 115.4 (46.5) lg/L higher mean (SE)

CysC levels than the normal controls (p \ 0.001)

(Table 4).

Upon evaluation of the nonactive smokers stratified by

ETS exposure in the entire study population, the mean (SE)

CysC level in ETS-unexposed nonactive smokers was 884

(12) lg/L. After adjusting for demographic and clinical

characteristics, we observed that nonactive smokers with

significant ETS exposure had 31.2 (15.2) lg/L higher mean

(SE) serum CysC levels than the ETS-unexposed nonactive

smokers (p = 0.04).

Discussion

In this study we have demonstrated a significant associa-

tion between emphysema and serum CysC in a large rep-

resentative US population. The emphysema group had

significantly higher mean CysC levels compared to the

normal controls. On adjusted analysis, after stratifying the

emphysema group by smoking status, the active smokers

Table 1 Baseline

characteristics

All values represent percentage

(standard error) of US

population, unless otherwise

stated

SE standard error

* 0.01 B p \ 0.05;

** 0.001 B p \ 0.01;

*** p \ 0.001a Composite cardiovascular

cerebrovascular disease

includes history of coronary

artery disease and history of

stroke

Normal controls

(N = 5,264)

Emphysema

(N = 120)

Chronic bronchitis

(N = 254)

Prevalence 93.2 (0.7) 1.3 (0.2) 5.5 (0.6)

Males 49.3 (1.1) 57.4 (3.9) 26.8 (4.3)***

Females 50.7 (1.1) 42.6 (3.8) 73.2 (4.3)***

Mean (SE) age (years) 41.1 (0.5) 64.8 (1.1)*** 46.6 (1.2)***

Whites 69.4 (1.9) 86.7 (4.0) 79.7 (3.8)*

Comorbidities

Hypertension 29.2 (1.3) 61.4 (4.2)** 50.5 (4.0)***

Hyperlipidemia 27.3 (1.8) 56.3 (5.2)** 39.1 (4.9)**

Diabetes by self-report 5.4 (0.3) 14.2 (2.6)** 11.1 (2.5)**

Mean (SE) C-reactive protein (mg/dl) 0.36 (0.01) 0.89 (0.13)** 0.64 (0.11)*

Mean (SE) body mass index (kg/m2) 27.1 (0.2) 27.9 (0.7) 31.5 (0.8)***

Chronic kidney disease

(GFR \60 mL/min/1.73 m2)

6.9 (0.4) 25.5 (3.8)*** 9.3 (2.1)

Coronary artery disease 5.9 (0.6) 25.0 (1.7)*** 11.8 (0.9)*

History of stroke 2.4 (0.3) 10.5 (0.8)*** 2.8 (0.9)

Composite cardiovascular

cerebrovascular diseasea7.4 (0.6) 29.4 (1.4)*** 12.8 (0.9)*

Congestive heart failure 1.9 (0.3) 22.1 (1.3)*** 4.6 (0.5)*

Smoking status

Never 48.8 (4.8) 8.4 (0.7)*** 43.1 (4.7)

Mean (SE) serum cotinine (ng/mL) 0.22 (0.03) 0.05 (0.01)*** 0.28 (0.03)

Former 22.7 (3.1) 43.1 (4.6)*** 24.5 (3.5)

Mean (SE) serum cotinine (ng/mL) 0.32 (0.07) 0.35 (0.08) 0.79 (0.11)

Active 28.5 (3.6) 48.5 (4.8)*** 32.4 (3.8)

Mean (SE) serum cotinine (ng/mL) 190.4 (6.7) 227.1 (20.2) 238.9 (10.3)*

286 Lung (2012) 190:283–290

123

with emphysema were observed to have higher CysC levels

compared to the nonsmokers. In addition, there was a

higher mean CysC level observed in the nonactive smokers

with significant ETS exposure in comparison to the normal

controls, suggesting that any active or second-hand smoke

exposure might contribute to increased serum CysC levels.

Cathepsins belong to the family of cysteine proteases

that catalyze the splitting of large proteins into amino acids

and play a significant role in pathogen killing and regula-

tion of inflammatory processes [31]. The activity of

cathepsins is closely regulated by a group of compounds

called cysteine protease inhibitors, which include stefins

and cystatins. The cystatin superfamily consists of three

types of compounds: type 1 cystatins (A and B) act intra-

cellularly, type 2 cystatins (C, D, E/M, F, G, and S) act

extracellularly, and type 3 cystatins are kininogens, which

are intravascular proteins [32]. Type 2 cystatins belong to a

heterogeneous group of ubiquitous proteins. CysC is the

most widespread type 2 cystatin and is secreted by virtually

all organs and is present in all body fluids. CysC inhibits

the action of cathepsins by noncovalent, reversible, com-

petitive binding and formation of inactive protease–cysta-

tin complexes. Tipping the balance in favor of proteases

may result in increased tissue destruction. It may be

speculated that increased tissue destruction resulting from

enhanced protease activity might lead to a compensatory

increase in the concentration of cystatins. The increase in

the concentration of cystatins as a maker of tissue

destruction has been utilized in several diseases.

Table 2 Mean (SE) cystatin C (lg/L) levels in study population

Normal

controls

Emphysema Chronic

bronchitis

Total 883 (8) 1139 (22)*** 902 (22)

Males 904 (8) 1167 (24)** 919 (23)

Females 863 (9) 1102 (25)*** 895 (20)

Smoking status

Never 870 (12) 1068 (21)** 904 (21)

Former 921 (13) 1217 (28)** 897 (24)

Active 907 (16) 1082 (28)* 902 (26)

Cotinine level (ng/mL)

\0.05 881 (12) 1138 (20)*** 895 (25)

0.05–10 867 (10) 1251 (26)* 908 (30)

C10 905 (15) 1082 (24)* 902 (22)

CRP level (mg/dL)

\0.1 826 (7) 975 (29)* 834 (22)

0.1–0.24 870 (9) 998 (28)** 877 (28)

0.25–0.49 919 (14) 1349 (32)** 911 (29)

C0.50 966 (16) 1200 (22)* 939 (26)

CRP C-reactive protein, SE standard error

* 0.01 B p \ 0.05; ** 0.001 B p \ 0.01; *** p \ 0.001

Fig. 2 Relationship between serum cystatin C and serum CRP. This

figure demonstrates the mean cystatin C level across 20 equal

quantiles of serum CRP concentrations in the entire study population.

The shaded square represents the mean cystatin C level and the

vertical lines represent the 95% confidence interval for each

respective quantile of CRP concentration

Table 3 Multivariate linear regression analysis with cystatin C

(lg/L) as outcome

Coefficient 95% confidence

interval

p value

Normal controls Reference

Emphysema 89.4 23.9 to 154.9 \0.001

Chronic bronchitis -11.8 -43.5 to 20.0 0.5

Age (years) 2.3 1.6 to 3.0 \0.001

Male sex 50.0 35.1 to 65.0 \0.001

Whites Reference

Blacks -17.2 -41.7 to 7.3 0.2

Mexican American -51.1 -71.4 to -30.7 \0.001

Other -38.6 -67.5 to -9.8 \0.001

Hypertension 20.1 1.7 to 38.4 \0.001

Hyperlipidemia -11.2 -32.2 to 9.8 0.3

Diabetes 15.7 -30.5 to 61.9 0.5

Cardiovascular disease 47.3 19.0 to 75.5 \0.001

GFR [90 mL/min/

1.73 m2Reference

GFR 60–90 mL/min/

1.73 m262.1 39.3 to 84.9 \0.001

GFR 30–59 mL/min/

1.73 m2371.7 327.4 to 416.0 \0.001

GFR \30 mL/min/

1.73 m23,155.9 2,369.6 to

3,942.3

\0.001

Logarithm transformed

CRP (mg/dL)

46.6 27.3 to 65.9 \0.001

BMI (kg/m2) 2.0 0.3 to 3.6 \0.001

Constant 664.6 589.4 to 739.8 \0.001

GFR glomerular filtration rate, CRP C-reactive protein, BMI body

mass index

Lung (2012) 190:283–290 287

123

In the current literature, CysC has been studied exten-

sively in kidney disease, cardiovascular disease, and solid

organ malignancy [12–21]. CysC has been studied as a

more sensitive marker for renal dysfunction than serum

creatinine [15, 16]. The proposed mechanism of CysC in

estimation of glomerular filtration arises from its low

molecular weight which allows it to be freely filtered by the

glomerulus and reabsorbed, but not secreted, by tubular

cells [33]. CysC has a proposed role in atherosclerosis and

may serve as an earlier marker for renal insufficiency

associated with adverse outcomes [34]. In addition, CysC

has been shown to be associated with increased cardio-

vascular and all-cause mortality [35].

Higher levels and activity of cathepsins and lower levels

of CysC have been shown to be present in higher-grade

malignancies and metastatic disease [20]. The suggested

mechanism of action is that cathepsins are secreted to

degrade the basement membrane and extracellular matrix,

in effect facilitating metastasis. With lower levels of CysC,

the inhibition of this function is insufficient [18–20].

There are multiple protease–antiprotease mechanisms

that have been implicated in the pathogenesis of pulmonary

disease. More widely known are the underregulated neu-

trophil elastase, a serine protease, and matrix metallopro-

teinase activity contributing to the pathogenesis of

emphysema [31]. Finlay et al. [36] demonstrated increased

levels of the two major matrix metalloproteinases, colla-

genase and gelatinase, in the bronchoalveolar lavage fluid

of emphysematous patients compared to smoking controls

and suggested that the presence of collagenase was a better

indicator of emphysema in smoking patients. CysC is the

most potent inhibitor of cathepsin B and L, which have

been implicated in pulmonary diseases [3, 8, 37, 38]. Both

these cysteine proteases and their inhibitors are secreted by

alveolar macrophages [24].

The protease–antiprotease activity of cathepsins and

CysC has been studied in multiple pulmonary diseases.

Smoking activates neutrophils and increases alveolar

macrophage activity in the lung [39]. Upon alveolar mac-

rophage activation, both proteases and antiproteases are

released [40, 41]. Warfel [25] demonstrated elevated CysC

and cathepsin B levels in the culture medium of alveolar

macrophages of smokers relative to nonsmokers. Abboud

[8] measured cathepsin L and CysC levels in bronchoal-

veolar lavage samples of smokers and found elevated

levels in smokers with emphysema compared to smokers

without emphysema. This suggestion that smokers that

develop clinically significant emphysema have higher

CysC levels compared to their nonemphysematous smok-

ing counterparts is also corroborated in our study. Lesser

[3] showed that intratracheal instillation of cathepsin B in

hamsters can induce emphysema. In a baboon model of

bronchopulmonary dysplasia, a neonatal disease of alter-

nating areas of emphysema and atelectasis, elevated

cathepsin activity was found without a concurrent elevation

in CysC activity, leading to a net increased cysteine pro-

tease activity [26]. These studies suggest that underregu-

lated cathepsin activity may lead to clinically relevant

pulmonary disease. Levels of total and unbound serum

cathepsin B and L were elevated in patients with pul-

monary malignancy [17, 19, 21, 42].

The finding of elevated CysC levels in smokers was

corroborated in our study, specifically showing elevated

CysC in emphysema patients who were active smokers. As

shown in Fig. 1, several mechanistic associations might

explain this observed relationship. CysC might be elevated

as a direct response to cigarette smoking or to the devel-

opment of emphysema or even to nonspecific lung tissue

destruction. The other, more plausible mechanism of CysC

elevation might be cathepsins–CysC imbalance in the

process of tissue destruction. Data suggest that elevated

cathepsin B and elevated CysC might still represent an

imbalance in favor of protease activity; despite elevated

Table 4 Multivariate linear regression analysis with stratification of

the emphysema group by smoking status

Coefficient 95% confidence

interval

p value

Normal controls Reference

Nonsmoker -37.2 -106.4 to 31.9 0.3

Former smoker 86.6 -43.4 to 216.7 0.2

Active smoker 115.4 24.2 to 206.6 \0.001

Age (years) 2.3 1.6 to 3.1 \0.001

Male sex 50.6 34.5 to 66.7 \0.001

Whites Reference

Blacks -18.2 -43.3 to 6.9 0.1

Mexican American -51.6 -71.5 to -31.6 \0.001

Other -32.8 -56.5 to -9.2 \0.001

Hypertension 21.0 2.6 to 39.5 \0.001

Hyperlipidemia -16.1 -39.8 to 7.6 0.2

Diabetes 1.2 -44.3 to 46.8 1.0

Cardiovascular disease 54.1 18.9 to 89.3 \0.001

GFR [90 mL/min/

1.73 m2Reference

GFR 60–90 mL/min/

1.73 m261.6 38.8 to 84.4 \0.001

GFR 30–59 mL/min/

1.73 m2372.5 324.0 to 421.1 \0.001

GFR \30 mL/min/

1.73 m23337.5 2594.6 to 4080.5 \0.001

Logarithm transformed

CRP (mg/dL)

43.5 23.7 to 63.3 \0.001

BMI (kg/m2) 2.5 0.7 to 4.3 \0.001

Constant 648.3 570.8 to 725.8 \0.001

GFR glomerular filtration rate, CRP C-reactive protein, BMI body

mass index

288 Lung (2012) 190:283–290

123

CysC levels, the cathepsin activity level might still be

active [3, 25]. Alternatively, the elevated levels of CysC

might reflect the body’s lagging response to counteract the

effect of increased concentration of the cathepsins induced

by smoking. However, the degree of elevation in CysC

might not be sufficient to inhibit the simultaneous

increased cathepsin levels. It has been demonstrated in

animal studies that the capacity of CysC to neutralize

excessive local proteolytic activity of the cathepsins might

be limited in the lung [26]. CysC might not achieve suf-

ficient levels in the microenvironment of the alveolar

macrophage where cathepsins are released by lysosomes

[43]. In addition, CysC has affinity to primarily cathepsin

B; other cathepsins might be less inhibited by CysC and

might contribute to tissue destruction.

Figure 2 demonstrates a dose-dependent trend of

increasing CysC levels with increasing serum CRP. As

demonstrated in Fig. 1, activation of inflammatory cells

might lead to activation of cathepsins with a consequent

increase in CysC. However, further analysis revealed that

the significant elevation in CysC levels tended to persist

despite an adjustment for serum CRP, suggesting that there

may be a non-CRP-mediated role of cathepsins and/or CysC

in the pathogenesis of emphysema. It is important to note

that by suggesting a non-CRP-mediated role of CysC, we do

not discount a concomitant CRP-mediated role of CysC. As

smoking and elevated CRP levels contribute to the inflam-

matory phenomenon of alveolar macrophage activation and

ultimately emphysema, there might be an additional role of

CRP within the inflammatory pathway and should be fur-

ther investigated in future studies. Furthermore, the non-

CRP-mediated role of CysC should be further elucidated to

clarify any preventative or therapeutic potential.

Limitations

Our data set was a semirandom sample that was based on

self-report and was susceptible to reporter’s bias. With this

cross-sectional study, an evaluation of causality could not

be made. In addition, as our study population was a non-

institutionalized civilian population, there was a potential

of survival bias as sicker patients with more severe risk

factors were likely not included in this study. However,

despite evaluating a potentially less sick group, our results

have likely underestimated the association between CysC

and emphysema.

We performed an analysis on nonsmokers with signifi-

cant ETS exposure based on reported nonsmokers with

serum cotinine levels [0.05 ng/mL. This categorization

might not capture all subjects with significant ETS expo-

sure, depending on the extent of exposure and the time

interval between the exposure and serum cotinine level

measurement.

A major limitation of our study was the use of self-

reported outcomes and the lack of spirometric or imaging

confirmation of emphysema. Due to this limitation, an

analysis of the outcomes stratified by severity of disease

was not possible. Given the association between CysC and

emphysema that we found in this study, further study into

this relationship, including correlation to radiographic

imaging and pulmonary function testing, would be of

interest.

As comorbidities, including cardiovascular disease,

renal disease, and smoking, may affect serum CysC levels,

these confounding factors were identified and controlled

for in our analysis. We performed a rigorous multivariate

linear regression analysis that controlled for demographic

information and clinical comorbidities, including renal

disease by stages of CKD, cardiovascular disease, and

cardiac risk factors. Importantly, despite controlling for

CRP level and stratifying by smoking history and despite a

robust analysis to control for confounding factors that

might affect CysC levels, we continued to find a significant

association between serum CysC and emphysema. We

acknowledge the possibility of residual confounding due to

unmeasured factors which might affect our analysis.

Conclusions

Our study demonstrated a significant association between

serum CysC levels and emphysema in a large representative

noninstitutionalized US population. Active smokers with

emphysema were observed to have significantly higher

levels of CysC in comparison to the normal controls. In

addition, never smokers with significant ETS exposure had

significantly higher CysC compared to those without a

significant ETS exposure. Due to a significant association of

CRP and CysC in our study population, it might be specu-

lated that elevation of CysC occurs secondary to inflam-

matory processes in the lung. However, persistence of

significant association between emphysema and CysC

despite adjustment for serum CRP might indicate a nonin-

flammatory pathway in the etiopathogenesis of emphysema.

Conflict of interest The authors have no conflicts of interest to

disclose.

References

1. Jannoff A, Sloan B, Weinbaum G et al (1977) Experimental

emphysema induced with purified human neutrophil elastase:

tissue localization of the instilled protease. Am Rev Respir Dis

115:461–478

2. Owen CA, Campbell EJ (1999) The cell biology of leukocyte-

mediated proteolysis. J Leukoc Biol 65:137–150

Lung (2012) 190:283–290 289

123

3. Lesser M, Padilla ML, Cardozo C (1992) Induction of emphy-

sema in hamsters by intratracheal instillation of cathepsin B. Am

Rev Respir Dis 145:661–668

4. Chughtai B, O’Riordan TG (2004) Potential role of inhibitors of

neutrophil elastase in treating diseases of the airway. J Aerosol

Med 17:289–298

5. Chua F, Laurent GJ (2006) Neutrophil elastase: mediator of

extracellular matrix destruction and accumulation. Proc Am

Thorac Soc 3:424–427

6. Lungarella G, Cavarra E, Lucattelli M, Martorana PA (2008) The

dual role of neutrophil elastase in lung destruction and repair. Int

J Biochem Cell Biol 40:1287–1296

7. Owen CA (2008) Roles for proteinases in the pathogenesis of

chronic obstructive pulmonary disease. Int J Chronic Obstr Pulm

Dis 3:253–268

8. Abboud RT, Vimalanathan S (2008) Pathogenesis of COPD. Part

1. The role of protease–antiprotease imbalance in emphysema. Int

J Tuberc Lung Dis 12:361–367

9. Kos J, Lah TT (1998) Cysteine proteinases and their endogenous

inhibitors: target proteins for prognosis, diagnosis, and therapy in

cancer. Oncol Rep 5:1349–1361

10. Chapman HA, Stone OL (1984) Comparison of live human

neutrophils and alveolar macrophage elastolytic activity in vitro.

Relative resistance of macrophage elastolytic activity to serum

and alveolar proteinase inhibitors. J Clin Invest 74:1693–1700

11. Abbott DE, Margaryan NV, Jeruss JS et al (2010) Reevaluating

cathepsin D as a biomarker for breast cancer: serum activity

levels versus histopathology. Cancer Biol Ther 9:23–30

12. Meng L, Yang Y, Qi LT, Wang XJ, Xu GB, Zhang BW (2011)

Elevated serum cystatin C is an independent predictor of car-

diovascular events in people with relatively normal renal func-

tion. J Nephrol. doi:10.5301/jn.5000020

13. Bevc S, Hojs R, Ekart R, Gorenjak M, Puklavec L (2011) Simple

cystatin C formula compared to sophisticated CKD-EPI formulas

for estimation of glomerular filtration rate in the elderly. Ther

Apher Dial 15:261–268

14. Kottgen A, Selvin E, Stevens LA et al (2008) Serum cystatin C in

the United States: the third National Health and Nutrition

Examination Survey (NHANES III). Am J Kidney Dis 51:

385–394

15. Kyhse-Andersen J, Schmidt C, Nordin G et al (1994) Serum

cystatin C, determined by a rapid, automated particle-enhanced

turbidimetric method, is a better marker than serum creatinine for

glomerular filtration rate. Clin Chem 40:1921–1926

16. Dharnidharka VR, Kwon C, Stevens G (2002) Serum cystatin C

is superior to serum creatinine as a marker of kidney function: a

meta-analysis. Am J Kidney Dis 40:221–226

17. Zore I, Krasovec M, Cimerman N et al (2001) Cathepsin

B/Cystatin C complex levels in sera from patients with lung and

colorectal cancer. Biol Chem 382:805–810

18. Poteryaeva ON, Falameyeva OV, Korolenko TA et al (2000)

Cysteine proteinase inhibitor level in tumor and normal tissues in

control and cured mice. Drugs Exp Clin Res 26:301–306

19. Krepela E, Prochazka J, Karova B, Cermak J, Roubkova H (1998)

Cysteine proteases and cysteine protease inhibitors in non-small

cell lung cancer. Neoplasma 45:318–331

20. Kos J, Werle B, Lah T, Brunner N (2000) Cysteine proteinases

and their inhibitors in extracellular fluids: markers for diagnosis

and prognosis in cancer. Int J Biol Markers 15:84–89

21. Werle B, Schanzenbacher U, Lah TT et al (2006) Cystatins in

non-small cell lung cancer: tissue levels, localization and relation

to prognosis. Oncol Rep 16:647–655

22. Buttle DJ, Burnett D, Abrahamson M (1990) Levels of neutrophil

elastase and cathepsin B activities, and cystatins in human spu-

tum: relationship to inflammation. Scand J Clin Lab Invest

50:509–516

23. Luthgens K, Ebert W, Trefz G, Gabrijelcic D, Turk V, Lah T

(1993) Cathepsin B and cysteine proteinase inhibitors in bron-

choalveolar lavage fluid of lung cancer patients. Cancer Detect

Prev 17:387–397

24. Werle B, Sauckel K, Nathanson CM et al (2003) Cystatins C,

E/M, and F in human pleural fluids of patients with neoplastic and

inflammatory lung disorders. Biol Chem 382:281–287

25. Warfel AH, Cardozo C, Yoo OH, Zucker-Franklin D (1991)

Cystatin C and cathepsin B production by alveolar macrophages

from smokers and nonsmokers. J Leukoc Biol 49:41–47

26. Altiok O, Yasumatsu R, Bingol-Karakoc G et al (2006) Imbal-

ance between cysteine proteases and inhibitors in a baboon model

of bronchopulmonary dysplasia. Am J Respir Crit Care Med

173:318–326

27. Centers for Disease Control and Prevention (2010) NHANES–

National Health and Nutrition Examination Survey homepage.

Available at http://www.cdc.gov/nchs/nhanes.htm/. Accessed 3

August 2010

28. Levey AS, Coresh J, Balk E et al (2003) National Kidney

Foundation practice guidelines for chronic kidney disease: eval-

uation classification and stratification. Ann Intern Med 139:

137–147

29. Coresh J, Selvin E, Stevens LA et al (2007) Prevalence of chronic

kidney disease in the United States. JAMA 298:2038–2047

30. Selvin S, Manzi J, Stevens LA et al (2007) Calibration of serum

creatinine in the National Health and Nutrition Examination

Surveys (NHANES) 1988–1994, 1999–2004. Am J Kidney Dis

50:918–926

31. Korkmaz B, Horwitz MS, Jenne DE, Gauthier F (2010) Neutro-

phil elastase, proteinase 3, and cathepsin G as therapeutic targets

in human disease. Pharmacol Rev 62:726–759

32. Abrahamson M (1994) Cystatins. Methods Enzymol 244:

685–700

33. Salgado JV, Neves FA, Bastos MG et al (2010) Monitoring renal

function: measured and estimated glomerular filtration rates—a

review. Braz J Med Biol Res 43:528–536

34. Lafarge JC, Naour N, Clement K, Guerre-Millo M (2010)

Cathepsins and cystatin C in atherosclerosis and obesity. Bio-

chimie 92:1580–1586

35. Shlipak MG, Marnak MJ, Katz R et al (2005) Cystatin C and the

risk of death and cardiovascular events among elderly persons.

N Engl J Med 352:2049–2060

36. Finlay GA, Russell KJ, McMahon KJ et al (1997) Elevated levels

of matrix metalloproteinases in bronchoalveolar lavage fluid of

emphysematous patients. Thorax 52:502–506

37. Henskens YM, Veerman EC, Amerongen AV (1996) Cystatins in

health and diseases. Biol Chem Hoppe Seyler 377:71–86

38. Abrahamson M, Alvarez-Fernandez M, Nathanson CM (2003)

Cystatins. Biochem Soc Symp 70:179–199

39. Finkelstein R, Fraser RS, Ghezzo H, Cosio MG (1995) Alveolar

inflammation and its relation to emphysema in smokers. Am J

Respir Crit Care Med 152:1666–1672

40. Takeyabu K, Betsuyaku T, Nishimura M et al (1998) Cysteine

proteinases and cystatin C in bronchoalveolar lavage fluid from

subjects with subclinical emphysema. Eur Respir J 12:1033–1039

41. Chapman HA, Reilly JJ, Yee R, Grubb A (1990) Identification of

cystatin C, a cysteine proteinase inhibitor, as a major secretory

product of human alveolar macrophages in vitro. Am Rev Respir

Dis 141:698–705

42. Sloane BF, Rozhin J, Robinson D, Honn KV (1990) Role for

cathepsin B and cystatins in tumor growth and progression. Biol

Chem Hoppe Seyler 371:193–198

43. Woischnik M, Bauer A, Aboutaam R et al (2008) Cathepsin H

and napsin A are active in the alveoli and increased in alveolar

proteinosis. Eur Respir J 31:1197–1204

290 Lung (2012) 190:283–290

123