Open Access

Low-dose CT measurements of airway dimensions and emphysema associated with airflow limitation in heavy smokers: a cross sectional study

  • Akkelies E Dijkstra1Email author,
  • Dirkje S Postma1,
  • Nick ten Hacken1,
  • Judith M Vonk1, 2,
  • Matthijs Oudkerk3,
  • Peter MA van Ooijen3,
  • Pieter Zanen4,
  • Firdaus A Mohamed Hoesein4,
  • Bram van Ginneken5,
  • Michael Schmidt6 and
  • Harry JM Groen1
Respiratory Research201314:11

DOI: 10.1186/1465-9921-14-11

Received: 24 October 2012

Accepted: 17 January 2013

Published: 28 January 2013



Increased airway wall thickness (AWT) and parenchymal lung destruction bothcontribute to airflow limitation. Advances in computed tomography (CT)post-processing imaging allow to quantify these features. The aim of thisDutch population study is to assess the relationships between AWT, lungfunction, emphysema and respiratory symptoms.


AWT and emphysema were assessed by low-dose CT in 500 male heavy smokers,randomly selected from a lung cancer screening population. AWT was measuredin each lung lobe in cross-sectionally reformatted images with an automatedimaging program at locations with an internal diameter of 3.5 mm, andvalidated in smaller cohorts of patients. The 15th percentilemethod (Perc15) was used to assess the severity of emphysema. Informationabout respiratory symptoms and smoking behavior was collected byquestionnaires and lung function by spirometry.


Median AWT in airways with an internal diameter of 3.5 mm(AWT3.5) was 0.57 (0.44 - 0.74) mm. Median AWT in subjectswithout symptoms was 0.52 (0.41-0.66) and in those with dyspnea and/orwheezing 0.65 (0.52-0.81) mm (p<0.001). In the multivariate analysisonly AWT3.5 and emphysema independently explained 31.1%and9.5%of the variance in FEV1%predicted, respectively,after adjustment for smoking behavior.


Post processing standardization of airway wall measurements provides areliable and useful method to assess airway wall thickness. Increased airwaywall thickness contributes more to airflow limitation than emphysema in asmoking male population even after adjustment for smoking behavior.


Airway dimensions Low-dose CT Respiratory symptoms Smoking Airflow limitation Emphysema


The quantification of airway dimensions by CT has become feasible with thedevelopment of multi detector computed tomography (CT) and new software tools forimage analysis [1, 2]. Assessment of airway dimensions by CT has been studied particularly inrelation to asthma, smoking and chronic obstructive pulmonary disease (COPD) [38], diseases generally associated with chronic or intermittent airflowlimitation. So far, airway wall thickness (AWT) measurements have been performed byselecting well quantifiable airways [912] or by standardizing dimensions to airways with a 10 mm internallumen perimeter (pi10, equivalent to about 3.2 mm internal airway lumendiameter) derived from a small number of airways [5, 6, 1315].

More recently, low-dose multi slice CT has become available, a technique with a goodquantifying performance that is preferred for monitoring of pulmonary and airwayabnormalities as compared to the high radiation exposures with high-resolutioncomputed tomography (high-dose CT). The cumulative radiation dose exposure withlow-dose CT remains very low, even when individuals are exposed multiple times [16]. Airway dimensions measured with low-dose CT have only been reported infew studies using the same diversity in analytic approaches as applied withhigh-dose CT measurements [7, 1719].

Airway dimensions and extent of emphysema are known to be associated with airflowlimitation [6, 914, 19, 20], although the influence of smoking behavior or signs of airway diseasesuch as cough, dyspnea, wheezing or mucus overproduction on airflow limitation isnot clear [5, 7, 21, 22].

The aims of this study are to quantify airway dimensions of the lung in multipleairway sections of each lobe in a novel manner and the extent of emphysema by usinglow-dose CT. These measurements were associated with the influence of airflowlimitation, respiratory symptoms and corrected for smoking behavior.



We randomly selected 500 current and former smokers participating in theGroningen cohort of a male population-based multi-centre lung cancer screeningstudy (NELSON). The Dutch ministry of health and the Medical Ethics Committee ofthe hospital approved the study protocol. Informed consent was obtained from allparticipants. Detailed inclusion criteria and characteristics have beendescribed elsewhere [23]. In short, subjects with a smoking history of at least 20 pack-yearswere included. Information on the presence or absence of respiratory symptomsand smoking (pack-years and former or current smoking) was obtained byquestionnaires. The question used to record respiratory symptoms was “doyou have experienced the following symptoms cough, sputum expectoration,wheezing or dyspnea for at least 3 months during the past year, even whenyou did not have a cold?”

Lung function

All participants performed standardized spirometry according to the EuropeanRespiratory Society guidelines [24], including forced expiratory volume in 1 sec (FEV1)and forced vital capacity (FVC) at the start of the study. In this population-based study we did not administer a bronchodilator.

CT scanning

Low-dose CT images of the chest were acquired at full inspiration afterappropriate instruction on one CT scanner (Sensation 16, Siemens MedicalSolutions, Forchheim, Germany) [23] according to the following protocol: spiral acquisition at120 kV, 20mAs, rotation time 0.5 s, pitch 1.5 and collimation16×0.6 mm, field of view 300 to 350 mm, slice thickness1 mm and slice increment 0.7 mm. The effective radiation dose wasless than 0.8 mSv. Contrast medium was not used. The images werereconstructed to a pixel matrix of 512×512 using B30f kernel. Thus, thespatial resolution was 0.59 to 0.68 mm in x y plane, and 0.7 mm in z plane. The CT system wascalibrated routinely.

Quantification of AWT

AWT was measured in cross-sectionally reformatted images with an automatedresearch software prototype MEVIS Airway Examiner v1.0 (release 2009, FraunhoferMEVIS, Bremen, Germany) based on an algorithm by Weinheimer at locations with afixed internal diameter of 3.5 mm in each lung lobe [25]. This software automatically extracts airway centerlines, re-samplesimages perpendicular to the airway direction at equally spaced positions alongthe centerline and detects inner and outer airway wall borders in these images.The outer wall border is detectable when no adjacent tissue with similar CTdensity is present and is taken into account when the wall is detected in atleast 25% of the perimeter at a location. AWT and the fraction of perimeterwhere the outer wall border was identified (Assessed Perimeter Fraction, APF)are calculated for each location. Wall thickness quantification accounts forpartial volume effects by integrating Hounsfield units across the wall. Accuracyand reproducibility of this algorithm was tested previously under clinicalconditions using a similar protocol as used in our study [2]. Average wall thickness and cumulative APF of all detectable airwaylocations with a fixed lumen diameter is reported per lobe and for the wholelung. The borders of the lung lobes were automatically calculated by thesoftware in a standard way. All low-dose CT scans were visually evaluated forappropriate segmentation.

Quantification of emphysema and lung volume

Quantification of emphysema was based on density differences and measured with asoftware tool called Image Xplorer (Image Sciences Institute, Utrecht, theNetherlands) [16, 26]. This software produces automatically the lung volume. The extent ofemphysema was automatically performed at the 15th percentile (Perc15)of the Hounsfield density distribution. Perc15 is the threshold density valuewhere 15% of all voxels has a lower density [27]. A lower Perc15, i.e. closer to −1000 HU, means that moreemphysema is present. All scans were reconstructed with a soft reconstructionfilter (Philips B, Siemens B30f). Airways were automatically excluded to assessdensity of lung parenchyma exclusively and HU densities of the entire scan wererecalibrated using automatically measured average densities in the trachea andshifting the HU values of the entire scan so that air density in the tracheabecame −1000 HU. Additionally, the percentage of low attenuation area,defined as the proportion of low-density voxels below −950 HU(%LAA-950HU) was used. %LAA-950HU was log-transformed because of skeweddistribution.

Explorative studies

Prior to the research described above we have
  1. 1)

    established the optimal internal airway diameter, i.e. the internal airway diameter that allows the highest number of cumulatively Assessed Perimeter Fractions (APF) for the whole lung. Therefore we measured APF on 20 selected NELSON CT’s in airways with a lumen diameter of 2.5, 3.0, 3.5, 4.0, 4.5 and 5.0 mm (± 0.25 mm) divided into 3 groups: no emphysema and normal lung function (n = 8, p15 > −920 and FEV1/FVC > 85 %), moderate emphysema and normal lung function (n = 4, −940 < p15 < −960 and FEV1/FVC > 70 %) and no emphysema and severe airflow limitation (n = 8, p15 > −920 and FEV1/FVC < 50 %).

  2. 2)

    compared the mean AWT3.5, using the same method as described above, at the established internal lumen size with high- and low-dose CT in 8 NELSON subjects from whom high- and low-dose CT were available. These CT data were obtained in spiral mode with 16×0.75 mm collimation and in full inspiration with the same scanner (Sensation-16 Siemens Medical Solutions, Forchheim, Germany). Axial images were reconstructed with 1.0 mm thickness at 0.7 mm increments. All scans were reconstructed with a soft reconstruction filter (Siemens B30f) at a 512×512 matrix.

  3. 3)

    determined the generation where airways with the established optimal internal lumen size are present. AWT measurements at 3.5 mm internal lumen size were performed in 57 randomly selected low-dose CTs of NELSON subjects. A multi-planar reconstruction (MPR) was made in each case of the apical upper lobe bronchus (B1) and the posterior lower lobe bronchus (B10). Subsequently the location was projected on the segmentation image. Three-dimensional image moving created the opportunity to observe airways from various directions and to check bifurcations and count airway generations according to the method of Boyden [28].


Statistical analysis

Data are reported as mean ± standard deviation (SD) or median(25th - 75th percentile) values as appropriate. Themean AWT at 3.5 mm internal lumen size (AWT3.5) of all fivelobes per case was calculated taking into account the APF per lobe by thefollowing formulae: ((AWT left upper lobe × APF left upper lobe) + (AWTleft lower lobe × APF left lower lobe) + (AWT right upper lobe × APFright upper lobe) + (AWT right middle lobe × APF right middle lobe) + (AWTleft upper lobe × APF left upper lobe)) / sum of APF of all lung lobes.AWT3.5 for the whole population was skewed distributed, thereforewe report median AWT and range, and log-transformed AWT was used in theanalyses.

Differences between various categories were explored using chi-square tests(dichotomous data), 2-tailed unpaired Student’s t-tests for normallydistributed continuous data and Mann-Whitney U-tests for not normallydistributed continuous data. The difference in airway wall thickness betweenlung lobes was assessed with a Wilcoxon signed rank test. Univariate linearregression analyses was used to study the relationships between clinicalvariables and AWT, and those variables with FEV1%predicted. Next,multivariate linear regression analyses were performed on those clinicalvariables showing significance in the univariate regression analyses. Outcomesof these analyses have been described with beta’s and p-values.Bland-Altman plot was used to analyze the agreement between AWT by high- andlow-dose CT [29]. All statistical analyses were performed using SPSS 18.0 for Windows;P-values below 0.05 were considered statistically significant.


Population characteristics

After visual evaluation 8 out of the 500 randomly selected subjects were excludedbecause of (partial) missing of airway segmentation on CT. The mean age of thecohort was 59.4 (± 5.2) years, approximately 59% were current smokersand median pack-years was 34.0 (28.0 - 45.6). More than 51% of thepopulation reported at least one respiratory symptom (Table 1).
Table 1

Clinical and demographic characteristics of heavy smokers from thegeneral population cohort



Age, years

59.4 ± 5.2

Pack-years smoking

34.0 (28.0 - 45.6)

Current smoking,%


FEV1, liter

3.45 ± 0.76


98.2 ± 19.7


70.0 ± 10.7

Emphysema; Perc15, HU

−920 (−930 to −907)

Emphysema;%LAA −950 HU

2.5 (1.3 - 4.3)

Emphysema; >5%LAA −950 HU,%


Lung volume on CT, liter

6.5 ± 1.4

Chronic Mucus Hypersecretion,%








No respiratory symptoms,%


Mean ± standard deviation shown for continuous data and median(interquartile range) for non-parametric distribution.

Definition of abbreviations: HU = Hounsfield Units; Perc15= the threshold density value where 15% of all voxels has a lowerdensity; %LAA −950 HU = percentage low attenuation areas <−950 HU; >5%LAA −950 HU,% = percentage of thepopulation having >5% low attenuation areas < −950HU.

Airway wall thickness

Exploratory analyses

The highest numbers of cumulatively assessed perimeter fractions (APF) ofairways were reached at an internal lumen perimeter of 3.5 ±0.25 mm (Figure 1); therefore this diameter wasselected for further analysis. Median AWT3.5 on low-dose CT wascomparable with median AWT3.5 on high-dose CT, respectively 0.57(0.48 - 0.74) mm and 0.55 (0.47 - 0.73) mm (p = 0.89, n = 8). Thisdemonstrates that MEVIS software analysis of data from low-dose CT givessimilar results as from high-dose CT (Figure 2).Airways with an internal diameter of 3.5 mm appeared mainly in the5th generation (range, 3rd - 7thgeneration) in the upper lobe bronchus, and mainly in the 8th -10th generation (range, 6th - 12thgeneration) in the lower lobe bronchus. This distribution was observedirrespective of smoking, presence of airflow limitation (defined asFEV1/FVC <0.70) or chronic mucus hypersecretion (CMH)(Figure 3).
Figure 1

Determination of the optimal airway size. Mean cumulativeassessed perimeter fractions in the total lung, for different groupsof patients at different internal airway lumen diameters. APF wasmeasured on low-dose CT of 20 selected NELSON subjects divided into3 groups: subjects without emphysema and with normal lung function(n = 8, perc15 > −910 and FEV1/FVC >85 %), with moderate emphysema and normal lung function (n= 4, -940 < perc15 < −960 and FEV1/FVC> 70 %) and without emphysema and having severeairflow limitation (n = 8, perc15 > −920 andFEV1/FVC < 50 %), in airways with alumen diameter of 2.5, 3.0, 3.5, 4.0, 4.5 and 5.0 mm (±0.25 mm).
Figure 2

Comparison of airway wall thickness on high- and low-dose CT.AWT3.5 was measured on high and low-dose CT of 8NELSON subjects. Bland & Altman plot shows agreement betweenmean AWT3.5 determined by high- and low-dose CT. Dashedlines depict the 95% confidence interval.
Figure 3

Distribution of 3.5 mm sized airways over airwaygenerations. The distribution of 3.5 mm internal lumensized airways over the 2nd till 7th airwaygenerations in the apical upper lobe bronchus (B1) and over the6th till 12th generation in the lower lobebronchus (B10) in the right lung, for subjects with and withoutairway obstruction, for subjects with and without Chronic MucusHypersecretion (CMH) and for current and former smokers. The numberof cases (e.g. no CMH and CMH) in each group was limited but thedistribution was similar.

Airway wall thickness in the population

In the whole population the median (25th - 75th percentile)AWT3.5 was 0.57 (0.44 - 0.74) mm. The APF in the whole lungvaried from 142 to 295 (median 215). The results per lung lobe are presented inthe online Additional file 1: Table S1).

Airway wall thickness and clinical characteristics

Significantly higher AWT3.5 values were observed in subjects withdyspnea and/or wheezing (n = 181, median AWT3.5 0.66 mm) orwith cough and/or CMH (n = 201, median AWT3.5 0.63 mm) comparedto subjects without dyspnea and/or wheezing (n = 309, median AWT3.50.53 mm, p<0.001) or without cough and/or CMH (n= 291, median AWT0.53 mm, p-values < 0.001). Current smokers and former smokers hadcomparable median AWT3.5 values, i.e. 0.58 mm and 0.56 mm(p = 0.113) respectively (Table 2).
Table 2

Airway wall thickness in subjects with and without respiratorysymptoms and in current and former smokers










0.62 (0.49-0.80)

no CMH


0.56 (0.44-0.71)




0.64 (0.50-0.84)

no cough


0.54 (0.43-0.70)




0.65 (0.51-0.80)

no dyspnea


0.54 (0.43-0.69)




0.66 (0.52-0.85)

no wheezing


0.53 (0.43-0.70)


CMH and/or cough


0.63 (0.49-0.79)

no CMH or cough


0.53 (0.43-0.70)


Dyspnea and/or wheezing


0.66 (0.52-0.83)

no dyspnea or wheezing


0.53 (0.42-0.67)


Cough, CMH, dyspnea and wheezing


0.69 (0.51-0.89)

no cough, CMH, dyspnea or wheezing


0.52 (0.41-0.66)


Current smoking


0.58 (0.45-0.75)

Former smoking


0.56 (0.43-0.71)


Data presented as median (interquartile range) values.

Definition of abbreviations: AWT3.5 = airwaywall thickness at 3.5 mm internal lumen size; CMH = chronicmucus hypersecretion.

Univariate linear regression analysis showed inverse associations betweenlog-AWT3.5 and FEV1 (b = −0.233, p < 0.001),FEV1/FVC (b = −0.015, p < 0.001),FEV1%predicted (b = −0.010, p < 0.001) and lung volume(b = −0.055, p<0.001) and positive associations betweenlog-AWT3.5, Perc15 and number of pack-years smoking, respectively(b = 0.003, p < 0.001) and (b = 0.003, p = 0.003) (Table 3).
Table 3

Univariate associations between (A) log-transformed AWT 3.5 and (B) FEV 1 % predicted, and clinical characteristics

Dependent variable

A. Log-AWT3.5

B. FEV1,% predicted









FEV1,% predicted




FEV1, liter










FVC,% predicted





Emphysema; Perc15





Emphysema; Log-%LAA −950 HU





Lung volume on CT, Liter










Smoking (former/current)





Age, years





Height, cm

























Definition of abbreviations: Log-AWT3.5 = logtransformed airway wall thickness at 3.5 mm diameter; CMH =chronic mucus hypersecretion; Log-%LAA −950 HU = logtransformed percentage of low attenuation areas < −950HU.

Multivariate analysis on all clinical variables significantly associated with AWTin univariate analyses revealed that log-AWT3.5 was independentlyassociated with lower FEV1%predicted (b = −0.010, p <0.001), higher Perc15 (b = 0.005, p < 0.001), and lung volume (b =−0.037, p = 0.005) respectively (Table 4).
Table 4

Multivariate linear regression: dependent variable is (A) logtransformed AWT 3.5 and (B) FEV 1 % predicted

Dependent variable

A. Log-AWT3.5

B. FEV1,% predicted





FEV1,% predicted








Emphysema; Perc15





Lung volume










Smoking (former/current)

























Definition of abbreviations: Log-AWT3.5 = logtransformed airway wall thickness at 3.5 mm diameter; CMH =chronic mucus hypersecretion.

Contribution of airway wall thickness and emphysema to airflow limitation

To study the contribution of AWT3.5, emphysema (Perc15), and clinicalvariables to airflow limitation, FEV1%predicted was taken asdependent variable. A significant negative association was found betweenFEV1%predicted and log-AWT3.5 (b = −31.21, p< 0.001), pack-years (b = −0.153, p = 0.004), cough, CMH, dyspnea andwheezing (b = −9.234, −8.17, −11.99, −13.26respectively, all p-values < 0.001), and a positive association betweenFEV%predicted and Perc15 (b = 0.204, p < 0.001) (Table 3).

Multivariate analysis on all clinical variables that were associated withFEV1%predicted in univariate analyses showed that higherFEV1%predicted was significantly associated with lowerAWT3.5 values (b = −31.3, p < 0.001) and with higherPerc15 (b = 0.342, p < 0.001) (Table 4).AWT3.5 and Perc15 explained 31.1% and 9.5% of thevariance of FEV1%predicted, respectively. The results of themultivariate regression analysis with emphysema expressed as %LAA−950HU as independent variable, are presented in the online Additional file 1: Table S2.


Low-dose CT is an appealing approach to quantify simultaneously pulmonary and airwayabnormalities. Our study shows that the use of low-dose CT combined with modern postprocessing software tools provides reliable information on airway wall thickness andthe extent of emphysema in a heavy smoking male population. Although CT does notprovide dynamic measurements, airway wall thickening and emphysema explainedrespectively 31.1% and 9.5% of the variance inFEV1%predicted, the most commonly used variable of airflowlimitation. Changes in AWT of more than 0.1 mm reflecting lumen surfacechanges over 8% measured at one air lumen level were observed between cases withand without respiratory symptoms.

Our study confirmed that increased AWT is associated with lowerFEV1%predicted. This lower FEV1%predicted depends onthe airway size in which the measurements of AWT are being performed [12, 17] and on the characteristics of the study population [7, 20]. Our population consisted of rather healthy elderly males from a randomlyrecruited Dutch heavy smoking population and still we were able to find significantassociations between thicker airway walls and more severe airflow limitation. Incontrast with the study of Nakano we found a significant negative associationbetween AWT and FEV1/FVC illustrating the sensitivity of our method [9].

The significant but weak negative association between airway wall thickness andemphysema has also been reported in other studies [7, 13, 30] but was not found in the study by Nakano [9]. Loss of elastic recoil may contribute to collapse of the airwaysresulting in a more proximal localization of airways with 3.5 mm internallumen diameter. As these more proximal airways have thicker airway walls thisphenomenon contributes to the weak negative association. Another possibleexplanation for this negative association may be that there are subjects withpredominantly airway wall thickening and others with predominantly emphysematouschanges. Particularly subjects with relatively more airway wall thickening areresponsible for the negative association and subjects with predominantlyemphysematous changes do hardly contribute. Apparently, in our population ofsubjects with normal lung function and with mild airflow limitation, the bronchiticphenotype is already present in the very early stages of smoking-induced lungdisease. Discrepancies between the study of Nakano [9] and our study may be due to the composition and size of the studypopulations, respectively predominantly emphysema versus predominantly healthysmokers with respiratory symptoms.

Importantly, we observed that the contribution of AWT3.5 to airflowlimitation was larger than the emphysema component. Moreover, AWT3.5 andemphysema together only explained about 40% of the variance inFEV1%predicted in this smoking male population. This unexpected lowcontribution of AWT3.5 and emphysema to FEV1%predicted maybe that the CT images were obtained at full inspiration, while FEV1reflects expiratory airflow limitation. One explanation for this observation is thatairflow limitation is not only due to reduced airway diameter at one level butshould be evaluated as an integral of all airways at all lumen diameters. This isdifficult to achieve and therefore we took the smallest measurable lumen diameterthat provides the largest contribution to airflow limitation. A more obviousphysiological explanation may be the presence of the heterogeneity in airwayventilation interrupting the symmetry in parallel airways leading to large clustersof poorly ventilated lung units [31].

In the univariate analysis, increased AWT3.5 was associated withrespiratory symptoms. However, AWT3.5 was not associated with thepresence of any respiratory symptom in the multivariate analysis after adjustmentfor FEV1%predicted, emphysema and smoking behavior. This findingcorresponds with other studies [7] and is consistent with the idea that inflammation and airway remodeling,associated with chronic bronchitis, is located in the more central airways [32]. The study of Martinez et al. showed a positive association betweenairway dimensions and questionnaires, the BODE index [33] and the St. George’s respiratory questionnaire [34] including questions about BMI, respiratory symptoms, exercise capacityand lung function. Also Camiciottoli et al. found a positive association betweenBODE and airway wall thickness [35]. Our study also showed that including respiratory symptoms in themultivariate model with AWT3.5 and emphysema has no impact on airflowlimitation.

Lung volume depends on height, weight and sex and as a consequence each person hasdifferent airway dimensions. Therefore, airway dimensions should be corrected forlung volume. Actually, volume-corrected AWT is the best parameter to use. In thisstudy lung volume does not change the multivariate model because FEV1%predictedis already corrected for lung volume by correcting for patient height.

It has been shown that the automated imaging program (MEVIS Airway Examiner) based ona method by Weinheimer et al. performed much better than the often used“full-width-at-half-maximum” method in a silicon tube phantom, regardingthe blurring effect of CT [25, 36]. Usually it is better to use sharper kernels for airway quantification.However, it was shown in the study by Schmidt et al. that the MEVIS airway examinerprovides reproducible quantitative results across different reconstruction kernels(B30f and B50f) and repeated acquisitions [2]. Moreover, the “full-width-at-half-maximum” techniquesystematically overestimates AWT, particularly in small airways [36]. Because low-dose CT and the automated imaging program (MEVIS AirwayExaminer) had not been used previously in clinical studies, we firstly optimized ourpost processing measurements in smaller cohorts of patients before applying it inthe population study. We demonstrated that the highest number of AWT measurementscould be performed on airways with an internal diameter of 3.5 mm,irrespective of the presence of airflow limitation or emphysema. In addition wedemonstrated that differences in AWT3.5 are not explained by differencesin airway generations. Finally, we demonstrated that low-dose CT imaging providedsimilar AWT results as high-dose CT imaging.

In a non-biased way we were able to evaluate 230 cumulatively assessed perimeterfractions (APF) per CT, ranging from 27-641 APF. In contrast to the commonly usedpi10 method, in which a secondary derived variable from few, mostly 6 selectedairways is used to estimate the airway wall thickness [5, 6, 1315], we obtained many direct airway wall measurements. To our opinion directmeasurements assessed over all lobes provide a better overall reflection of AWT thana limited number of secondary AWT measurements.

A limitation of this study is, inherent to general population-based studies, thatonly male smokers with mild emphysema and/or airflow limitation were included.Strengths of our study is the non-biased way of analyzing a high number of airwaysections in all lobes that makes our approach more suitable for combined airway wallthickness and emphysema measurements on one low-dose CT scan. Such approach allowsmonitoring of intervention effects on both parameters. This is important when newtreatment modalities will become available for clinical testing.

In the future further developments may involve measurements of thickness of airwaywalls along the full length of the bronchial tree at in- and expiration scans.Possibly more airflow variability will be explained.

In conclusion, post processing standardization of large numbers of airway wallmeasurements in all lung lobes is feasible, reliable and an useful method to assessairway wall thickness. We have demonstrated that increased airway wall thicknesscontributes more to airflow limitation than emphysema in a smoking male populationeven after adjustment for smoking behavior.



Cumulative assessed perimeter fractions


Airway wall thickness


Airway wall thickness at 3.5 mm internal lumendiameter


Body mass index


Chronic mucus hypersecretion




Forced expiratory volume in 1 sec


Ratio of forced expiratory volume in 1 sec and forcedvital capacity


(Ratio of present FEV1 andexpected FEV1) x 100 %


Forced vital capacity


theGlobal Initiative for Chronic Obstructive Lung Disease


High resolutioncomputed tomography


Hounsfield unit


Percentage of low attenuationareas


the Dutch-Belgian Randomized Lung Cancer Screening Trial


thethreshold density value where 15% of all voxels has a lower density


Pulmonary function test


Standard deviation.



MEVIS Medical Solutions provided the software used for the airway examiner.


The study was part of the EU-grant HEALTH F2-2007-201379 COPACETIC.

Authors’ Affiliations

Department of Pulmonary Diseases, University of Groningen, University Medical Center Groningen, GRIAC research institute
Department of Epidemiology, University of Groningen, University Medical Center Groningen
Department of Radiology, University of Groningen, University Medical Center Groningen
Division of Heart & Lungs, University Medical Center Utrecht
Department of Radiology, Radboud University Nijmegen Medical Centre
Fraunhofer MEVIS, Institute for Medical Image Computing


  1. Takasugi JE, Godwin JD: Radiology of chronic obstructive pulmonary disease. Radiol Clin North Am. 1998, 36: 29-55. 10.1016/S0033-8389(05)70006-3.PubMedView ArticleGoogle Scholar
  2. Schmidt M, Kuhnigk JM, Krass S, Owsijewitsch M, de Hoop B, Peitgen HO: Reproducibility of airway wall thickness measurements. Proceedings of the SPIE - The International Society for OpticalEngineering. 2010, 762410-762419.Google Scholar
  3. Kurashima K, Hoshi T, Takayanagi N, Takaku Y, Kagiyama N, Ohta C, Fujimura M, Sugita Y: Airway dimensions and pulmonary function in chronic obstructive pulmonarydisease and bronchial asthma. Respirology. 2011, 17: 79-86.View ArticleGoogle Scholar
  4. Matsumoto H, Niimi A, Tabuena RP, Takemura M, Ueda T, Yamaguchi M, Matsuoka H, Jinnai M, Chin K, Mishima M: Airway wall thickening in patients with cough variant asthma and nonasthmaticchronic cough. Chest. 2007, 131: 1042-1049. 10.1378/chest.06-1025.PubMedView ArticleGoogle Scholar
  5. Grydeland TB, Dirksen A, Coxson HO, Pillai SG, Sharma S, Eide GE, Gulsvik A, Bakke PS: Quantitative computed tomography: emphysema and airway wall thickness by sex,age and smoking. Eur Respir J. 2009, 34: 858-865. 10.1183/09031936.00167908.PubMedView ArticleGoogle Scholar
  6. Kim WJ, Silverman EK, Hoffman E, Criner GJ, Mosenifar Z, Sciurba FC, Make BJ, Carey V, Estepar RS, Diaz A, Reilly JJ, Martinez FJ, Washko GR, NETT Research Group: CT metrics of airway disease and emphysema in severe COPD. Chest. 2009, 136: 396-404. 10.1378/chest.08-2858.PubMedPubMed CentralGoogle Scholar
  7. Mair G, Maclay J, Miller JJ, McAllister D, Connell M, Murchison JT, Macnee W: Airway dimensions in COPD: Relationships with clinical variables. Respir Med. 2010, 104: 536-542.View ArticleGoogle Scholar
  8. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS, GOLD Scientific Committee: Global strategy for the diagnosis, management, and prevention of chronicobstructive pulmonary disease. NHLBI/WHO Global Initiative for ChronicObstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med. 2001, 163: 1256-1276.PubMedView ArticleGoogle Scholar
  9. Nakano Y, Muro S, Sakai H, Hirai T, Chin K, Tsukino M, Nishimura K, Itoh H, Pare PD, Hogg JC, Mishima M: Computed tomographic measurements of airway dimensions and emphysema insmokers. Correlation with lung function. Am J Respir Crit Care Med. 2000, 162: 1102-1108.PubMedView ArticleGoogle Scholar
  10. Matsuoka S, Kurihara Y, Yagihashi K, Hoshino M, Nakajima Y: Airway dimensions at inspiratory and expiratory multisection CT in chronicobstructive pulmonary disease: correlation with airflow limitation. Radiology. 2008, 248: 1042-1049. 10.1148/radiol.2491071650.PubMedView ArticleGoogle Scholar
  11. Nishimura M: Application of three-dimensional airway algorithms in a clinical study. Proc Am Thorac Soc. 2008, 5: 910-914. 10.1513/pats.200809-109QC.PubMedView ArticleGoogle Scholar
  12. Hasegawa M, Nasuhara Y, Onodera Y, Makita H, Nagai K, Fuke S, Ito Y, Betsuyaku T, Nishimura M: Airflow limitation and airway dimensions in chronic obstructive pulmonarydisease. Am J Respir Crit Care Med. 2006, 173: 1309-1315. 10.1164/rccm.200601-037OC.PubMedView ArticleGoogle Scholar
  13. Patel BD, Coxson HO, Pillai SG, Agustí AG, Calverley PM, Donner CF, Make BJ, Müller NL, Rennard SI, Vestbo J, Wouters EF, Hiorns MP, Nakano Y, Camp PG, Nasute Fauerbach PV, Screaton NJ, Campbell EJ, Anderson WH, Paré PD, Levy RD, Lake SL, Silverman EK, Lomas DA: Airway wall thickening and emphysema show independent familial aggregation inchronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2008, 178: 500-505. 10.1164/rccm.200801-059OC.PubMedView ArticleGoogle Scholar
  14. Lee YK, Oh YM, Lee JH, Kim EK, Lee JH, Kim N, Seo JB, Lee SD: Quantitative assessment of emphysema, air trapping, and airway thickening oncomputed tomography. Lung. 2008, 186: 157-165. 10.1007/s00408-008-9071-0.PubMedView ArticleGoogle Scholar
  15. Martinez CH, Chen YH, Westgate PM, Liu LX, Murray S, Curtis JL, Make BJ, Kazerooni EA, Lynch DA, Marchetti N, Washko GR, Martinez FJ, Han MK, COPDGene Investigators: Relationship between quantitative CT metrics and health status and BODE inchronic obstructive pulmonary disease. Thorax. 2012, 67: 399-406. 10.1136/thoraxjnl-2011-201185.PubMedPubMed CentralView ArticleGoogle Scholar
  16. Mets OM, de Jong PA, van Ginneken B, Gietema HA, Lammers JW: Quantitative computed tomography in COPD: possibilities and limitations. Lung. 2012, 190: 133-145. 10.1007/s00408-011-9353-9.PubMedPubMed CentralView ArticleGoogle Scholar
  17. Yamashiro T, Matsuoka S, Estepar RS, Dransfield MT, Diaz A, Reilly JJ, Patz S, Murayama S, Silverman EK, Hatabu H, Washko GR: Quantitative assessment of bronchial wall attenuation with thin-section CT:An indicator of airflow limitation in chronic obstructive pulmonarydisease. AJR Am J Roentgenol. 2010, 195: 363-369. 10.2214/AJR.09.3653.PubMedView ArticleGoogle Scholar
  18. Washko GR, Dransfield MT, Estepar RS, Diaz A, Matsuoka S, Yamashiro T, Hatabu H, Silverman EK, Bailey WC, Reilly JJ: Airway wall attenuation: a biomarker of airway disease in subjects withCOPD. J Appl Physiol. 2009, 107: 185-191. 10.1152/japplphysiol.00216.2009.PubMedPubMed CentralView ArticleGoogle Scholar
  19. Bon JM, Leader JK, Weissfeld JL, Coxson HO, Zheng B, Branch RA, Kondragunta V, Lee JS, Zhang Y, Choi AM, Lokshin AE, Kaminski N, Gur D, Sciurba FC: The influence of radiographic phenotype and smoking status on peripheralblood biomarker patterns in chronic obstructive pulmonary disease. PLoS One. 2009, 4: e6865-10.1371/journal.pone.0006865.PubMedPubMed CentralView ArticleGoogle Scholar
  20. Achenbach T, Weinheimer O, Biedermann A, Schmitt S, Freudenstein D, Goutham E, Kunz RP, Buhl R, Dueber C, Heussel CP: MDCT assessment of airway wall thickness in COPD patients using a new method:correlations with pulmonary function tests. Eur Radiol. 2008, 18: 2731-2738. 10.1007/s00330-008-1089-4.PubMedView ArticleGoogle Scholar
  21. Orlandi I, Moroni C, Camiciottoli G, Bartolucci M, Pistolesi M, Villari N, Mascalchi M: Chronic obstructive pulmonary disease: thin-section CT measurement of airwaywall thickness and lung attenuation. Radiology. 2005, 234: 604-610. 10.1148/radiol.2342040013.PubMedView ArticleGoogle Scholar
  22. Grydeland TB, Dirksen A, Coxson HO, Eagan TM, Thorsen E, Pillai SG, Sharma S, Eide GE, Gulsvik A, Bakke PS: Quantitative CT measures of emphysema and airway wall thickness Are relatedto respiratory symptoms. Am J Respir Crit Care Med. 2009, 34: 858-865.Google Scholar
  23. van Klaveren RJ, Oudkerk M, Prokop M, Scholten ET, Nackaerts K, Vernhout R, van Iersel CA, van den Bergh KA, van ’t Westeinde S, van der Aalst C, Thunnissen E, Xu DM, Wang Y, Zhao Y, Gietema HA, de Hoop BJ, Groen HJ, de Bock GH, van Ooijen P, Weenink C, Verschakelen J, Lammers JW, Timens W, Willebrand D, Vink A, Mali W, de Koning HJ: Management of lung nodules detected by volume CT scanning. N Engl J Med. 2009, 361: 2221-2229. 10.1056/NEJMoa0906085.PubMedView ArticleGoogle Scholar
  24. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault JC: Lung volumes and forced ventilatory flows. Report working partystandardization of lung function tests, European community for steel andcoal. Official statement of the European respiratory society. Eur Respir J Suppl. 1993, 16: 5-40.PubMedView ArticleGoogle Scholar
  25. Weinheimer O, Achenbach T, Bletz C, Duber C, Kauczor HU, Heussel CP: About objective 3-d analysis of airway geometry in computerizedtomography. IEEE Trans Med Imaging. 2008, 27: 64-74.PubMedView ArticleGoogle Scholar
  26. Gietema HA, Schilham AM, van Ginneken B, van Klaveren RJ, Lammers JW, Prokop M: Monitoring of smoking-induced emphysema with CT in a lung cancer screeningsetting: detection of real increase in extent of emphysema. Radiology. 2007, 244: 890-897. 10.1148/radiol.2443061330.PubMedView ArticleGoogle Scholar
  27. Parr DG, Sevenoaks M, Deng C, Stoel BC, Stockley RA: Detection of emphysema progression in alpha 1-antitrypsin deficiency using CTdensitometry; methodological advances. Respir Res. 2008, 9: 21-10.1186/1465-9921-9-21.PubMedPubMed CentralView ArticleGoogle Scholar
  28. Boyden EA: A synthesis of the prevailing patterns of the bronchopulmonary segments inthe light of their variations. Dis Chest. 1949, 15: 657-668. 10.1378/chest.15.6.657.PubMedView ArticleGoogle Scholar
  29. Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinicalmeasurement. Lancet. 1986, 1: 307-310.PubMedView ArticleGoogle Scholar
  30. Kim S, Choi IS, Kim YJ, Kim CS, Han ER, Park DJ, Kim DE: Airway responsiveness to inhaled aspirin is influenced by airwayhyperresponsiveness in asthmatic patients. Korean J Intern Med. 2010, 25: 309-316.PubMedPubMed CentralView ArticleGoogle Scholar
  31. Venegas JG, Winkler T, Musch G, Vidal Melo MF, Layfield D, Tgavalekos N, Fischman AJ, Callahan RJ, Bellani G, Harris RS: Self-organized patchiness in asthma as a prelude to catastrophic shifts. Nature. 2005, 434: 777-782. 10.1038/nature03490.PubMedView ArticleGoogle Scholar
  32. Hogg JC, Chu F, Utokaparch S, Woods R, Elliott WM, Buzatu L, Cherniack RM, Rogers RM, Sciurba FC, Coxson HO, Pare PD: The nature of small-airway obstruction in chronic obstructive pulmonarydisease. N Engl J Med. 2004, 350: 2645-2653. 10.1056/NEJMoa032158.PubMedView ArticleGoogle Scholar
  33. Ong KC, Earnest A, Lu SJ: A multidimensional grading system (BODE index) as predictor ofhospitalization for COPD. Chest. 2005, 128: 3810-3816. 10.1378/chest.128.6.3810.PubMedView ArticleGoogle Scholar
  34. Jones PW, Quirk FH, Baveystock CM, Littlejohns P: A self-complete measure of health status for chronic airflow limitation. TheSt. George’s Respiratory Questionnaire. Am Rev Respir Dis. 1992, 145: 1321-1327.PubMedView ArticleGoogle Scholar
  35. Camiciottoli G, Bigazzi F, Bartolucci M, Cestelli L, Paoletti M, Diciotti S, Cavigli E, Magni C, Buonasera L, Mascalchi M, Pistolesi M: BODE-index, modified BODE-index and ADO-score in chronic obstructivepulmonary disease: relationship with COPD phenotypes and CT lung densitychanges. COPD. 2012, 9: 297-304. 10.3109/15412555.2012.661000.PubMedView ArticleGoogle Scholar
  36. Mets OM, Buckens CF, Zanen P, Isgum I, van Ginneken B, Prokop M, Gietema HA, Lammers JW, Vliegenthart R, Oudkerk M, van Klaveren RJ, de Koning HJ, Mali WP, de Jong PA: Identification of chronic obstructive pulmonary disease in lung cancerscreening computed tomographic scans. JAMA. 2011, 306: 1775-1781. 10.1001/jama.2011.1531.PubMedView ArticleGoogle Scholar


© Dijkstra et al.; licensee BioMed Central Ltd. 2013

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (, whichpermits unrestricted use, distribution, and reproduction in any medium, provided theoriginal work is properly cited.