Share this article on:

Aortic pulse wave velocity results depend on which carotid artery is used for the measurements

Dzeko, Mirela; Peters, Christian D.; Kjaergaard, Krista D.; Jensen, Jens D.; Jespersen, Bente

doi: 10.1097/HJH.0b013e32835b051b
ORIGINAL PAPERS: Blood vessels

Objective: Aortic pulse wave velocity (aPWV) is a gold standard noninvasive marker of arterial stiffness. aPWV is usually obtained as carotid–femoral pulse wave velocity by measurements on the common carotid artery and the femoral artery. The carotid arteries branch slightly differently from the aorta towards the right and left side of the neck. Theoretically, using the right or left carotid artery could influence aPWV results and there are no clear guidelines to support the choice of side. The aim of this study was to investigate whether aPWV results depend on right or left side carotid artery measurements in a group of healthy individuals.

Methods and results: Two different observers examined 50 individuals without known cardiovascular disease between 23 and 66 years of age. The measurements were performed with the SphygmoCor equipment using both right and left carotid arteries. We found that use of the right carotid artery provided significantly higher aPWV values compared with the left carotid artery, also when using different methods to estimate the travel length of the pulse wave (pooled data, subtracted distance: 0.2 ± 0.4 m/s, P = 0.003; direct distance: 0.2 ± 0.5 m/s, P = 0.001).

Conclusion: Using right or left carotid artery affects aPWV, as right-side measurements provided higher values. Attention to this side difference and use of the same carotid artery will increase the strength of intervention studies using aPWV as a surrogate endpoint.

Department of Renal Medicine, Aarhus University Hospital, Aarhus N, Denmark

Correspondence to Mirela Dzeko, Department of Renal Medicine, Aarhus University Hospital, Brendstrupgaardsvej 100, 8200 Aarhus N, Denmark. Tel: +45 6146 1393; fax: +45 7845 2430; e-mail:

Abbreviations: aPWV, aortic pulse wave velocity; BP, blood pressure; DBP, diastolic blood pressure; ECG, electrocardiogram; ICC, intraclass correlation coefficient; SD, standard deviation; SDP, systolic blood pressure

Received 31 May, 2012

Revised 26 September, 2012

Accepted 4 October, 2012

Back to Top | Article Outline


Over the past two decades, the clinical significance of arterial stiffness has become clear as an independent predictor of cardiovascular outcome. Several studies have shown that arterial pulse wave velocity (PWV), a noninvasive marker of arterial stiffness, is a predictor of cardiovascular events in patients with hypertension [1–3], end-stage renal disease [4,5] as well as in elderly hospitalized patients [6]. More recent population-based studies have shown that PWV is an independent predictor of coronary heart disease and stroke in apparently healthy individuals [7], and development of first cardiovascular events [8].

Increased arterial stiffness may lead to an elevation of systolic blood pressure (SBP), raising left ventricular afterload and causing subsequent hypertrophy, and a decrease in diastolic blood pressure (DBP) impairing coronary perfusion [9]. Arterial stiffness can be evaluated by measuring PWV between two sites in the arterial tree. Carotid–femoral PWV, also known as aortic pulse wave velocity PWV (aPWV), is a measure of stiffness of the thoracic and abdominal aorta [10].

aPWV is mostly assessed by consecutive measurements at the common carotid artery and the femoral artery. The newest guidelines on measurement of aPWV recommend that the measurements preferentially should be done at the right common carotid and common femoral arteries [11]. However, these recommendations are not based on any studies comparing the use of different recording sites on the same individuals. This is relevant because there is a small anatomical difference in the way the right and left carotid arteries branch from the aorta towards the neck and this could alter aPWV results.

The aim of the present study was to investigate whether aPWV measurements on either right or left carotid artery differed in a group of healthy individuals, as well as to study intraobserver and interobserver variability of these measurements.

Back to Top | Article Outline


Study population

Healthy individuals were recruited by advertisement at Aarhus University Hospital and in a local newspaper. Fifty-six volunteers with no history of cardiovascular disease were enrolled after giving informed consent. Blood samples were taken before examination to exclude major kidney, liver, haematological and infectious diseases. Five individuals were excluded because they had SBP greater than 140 mmHg and one because of use of cholesterol-lowering medication. The demographic data for the 50 remaining individuals are shown in Table 1. The Central Denmark Region Committee on Biomedical Research Ethics approved the study (M-20100101).

Back to Top | Article Outline

Study protocol

Two different observers performed the examinations in a quiet room at a constant room temperature. Consumption of alcohol was not allowed 10 h prior to examination; intake of food, beverages containing caffeine and smoking were not allowed for 3 h before examination. All measurements were performed as double recordings with the individuals in a supine position. The first observer measured the individual's height and weight and took the blood sample. After 5 min of rest, the same observer measured the brachial blood pressure (BP) and aPWV, and the individual remained resting until the second observer arrived and repeated the aPWV measurements. It was randomly decided in which order the two observers examined every individual.

Back to Top | Article Outline

Blood pressure

The brachial BP was measured with an automatic BP monitor (Microlife; Microlife AG, Widnau, Switzerland) calibrated within the last year. BP was measured on both arms to rule out any side difference [12]. Measurements were done until they did not differ more than 5 mmHg in both SBP and DBP, and an average of the last two measurements was used [13].

Back to Top | Article Outline

Pulse wave velocity

All measurements of aPWV were performed using the SphygmoCor apparatus (Version 8.2; Atcor Medical, Sydney, Australia) by the same two observers in a random order. The method of aPWV measurements has been described in detail elsewhere [10]. Briefly, aPWV is found by sequential recordings of pressure waveforms at the carotid and femoral artery, and is expressed in meters per second (m/s). aPWV is defined as the distance between the two recording sites divided by the difference in pulse wave travel time. The length of the aorta was approximated by subtracting the distance between the carotid recording site and the suprasternal notch, from the distance between the femoral recording site and the suprasternal notch (subtracted distance). An electrocardiogram (ECG) was used to determine the start of the pulse wave defined as the R wave [14]. The distance was measured by a tape measure. Because the newest guidelines recommend the use of 80% of the direct distance between the femoral artery and the carotid artery recording site (carotid artery–femoral artery × 0.8), we also performed calculations, in which we compared the aPWV values obtained using the right and the left carotid artery, using the direct distance. The direct distance was calculated from the subtracted distance using the following equation [15]:

The distance used in the calculations was Xdirect × 0.8.

Back to Top | Article Outline

Recording sites

Each observer performed measurements on both the right and left carotid arteries on all participants. Measurements were performed twice by each observer. The first observer selected the femoral recording site with the best palpable pulse and marked this with an ‘x’. The right femoral artery was used as a recording site in 38 and the left femoral artery was used in the remaining 12 participants. Subsequently, both sides of the neck were examined and right and left carotid recording sites were also marked. aPWV was then measured using both the right and the left carotid arteries without changing the femoral artery. The second observer used the same recording sites (hence the markings).

Back to Top | Article Outline

Quality control

When performing the aPWV measurements, we strived to achieve visually acceptable waveforms, equal heart rate at both recording sites and an aPWV standard deviation (SD) of less than 10% of the mean aPWV. When this was not possible, the observers performed up to five measurements, and the two measurements with the lowest SD were chosen.

Back to Top | Article Outline


All data were analyzed using Stata 11 (StataCorp LP, College Station, Texas, USA). The assessment of the intra observer and interobserver reproducibility and the comparison of the aPWV measurements performed on the right and the left side of the neck were performed similarly. Statistically significant bias was ruled out using a variance ratio test. Bland–Altman plots (differences between the studied parameters plotted against their mean values) were used to check for asymmetry. The limits of agreement were calculated as suggested by Bland and Altman [16], and were considered as being 2 SDs of the mean differences, thus expressing the expected variation in 95% of the cases. Then, a paired two-tailed t-test analysis was performed to test for statistically significant differences. Furthermore, the variation was evaluated by considering the reliability of measurements in terms of the intraclass correlation coefficient (ICC). The ICC is defined as the ratio of the variance within the individuals to the total variance, which is a combination of the variance within and between individuals. If all the variation is explained by differences between individuals and methods and there is no variation within the individual, ICC = 1. If the ICC is close to zero, then all the variation is within individuals [17]. The data are presented as mean ± SD, unless otherwise stated. P values below 0.05 were considered significant.

Back to Top | Article Outline


Most aPWV recordings in the study were within the quality standard. However, a few individuals did not fulfil the quality standards in all the measurements. The mean SD for the pooled aPWV data was 6.5% (range 2.4–12.5%).

Back to Top | Article Outline

Intraobserver reproducibility

Calculation of intraobserver reproducibility was based on double recordings of aPWV, performed on the left and right carotid artery. Bland–Altman plots showed no sign of the mean differences being dependent on the underlying mean values. There was no significant difference between the first and second recordings performed on the same side of the neck by the same observer (Table 2). The ICCs were all above 76%. The mean differences and variations were less for observer 1 than observer 2.

Back to Top | Article Outline

Interobserver reproducibility

The interobserver calculations were based on the mean of two sets of 50 double recordings, performed on the left and right carotid artery by both observers. Bland–Altman plots showed no tendency of the mean differences being dependent on the underlying mean values. Observer 2 obtained higher aPWV values than observer 1, both when using the right and the left carotid artery. Though the differences between the two observers were small, a statistically significant interobserver difference was found between the measurements performed on the right side of the neck (P = 0.02) (Table 2). Both the ICCs were above 77%.

Back to Top | Article Outline

Right versus left carotid artery

The variation in aPWV measurements was calculated for each observer and for the pooled data, using both the subtracted and the direct distance × 0.8. Bland–Altman plots were made, and an example of this is shown in Figure 1 for the subtracted distance. None of the plots showed any skewed tendencies. Use of the right carotid artery resulted in significantly higher aPWV values than the left carotid artery, for both observers. Subtracted distance: observer 1 – 0.1 ± 0.4 m/s, P = 0.05; observer 2 – 0.2 ± 0.6 m/s, P = 0.009 and pooled data – 0.2 ± 0.4 m/s, P = 0.003, Table 3. Direct distance × 0.8: observer 1 – 0.2 ± 0.5 m/s, P = 0.02; observer 2 – 0.3 ± 0.7 m/s, P = 0.004 and pooled data – 0.2 ± 0.5 m/s, P = 0.001, Table 4. The ICCs were all above 85%.

To rule out a potential effect of the different femoral recording sites, separate calculations were made for the individuals in whom the right and the left femoral artery, respectively, was used as a recording site. In both groups, the tendency was the same as in the pooled data. Thus, the aPWV measurements performed using the right carotid artery were higher compared with the measurements based on the left carotid artery as a recording site. In the 38 individuals in whom the femoral recording site was on the right side, we found a mean difference in aPWV of (subtracted distance) 0.2 ± 0.3 m/s, P = 0.007 and (direct distance × 0.8) 0.2 ± 0.4 m/s, P = 0.001. In the remaining 12 individuals in whom the left femoral artery was used, there was a mean difference in aPWV measurements of (subtracted distance) 0.3 ± 0.6 m/s, P = 0.16 and 0.3 ± 0.7 m/s, P = 0.17 (direct distance × 0.8).

Back to Top | Article Outline


We demonstrated a higher aPWV when using the right compared to the left carotid artery as the recording site. The left and right common carotid arteries follow the same course with the exception of their origin. The right common carotid originates in the neck from the brachiocephalic trunk. The left arises from the aortic arch in the thoracic region. Thus, the actual carotid–femoral distance could differ between the two sides. Because the two carotid arteries have an asymmetrical origin and because the right subclavian artery emerges from the right carotid artery, it is possible that the flow in the right carotid artery could be more turbulent, which hypothetically could make the right carotid artery more prone to the development of arterial stiffness. This hypothesis is supported by several studies showing that plaque formation is most likely to occur at bifurcations, branchings and curvatures [18]. A strong association between arterial stiffness and atherosclerosis at different sites in the arterial tree has been shown in population-based studies [19] and in a population with 53% hypertensives [20].

In this study, we have shown high reproducibility for aPWV measurements performed by the same observer. Reproducibility of aPWV has been validated using the SphygmoCor equipment in a mixed population by Wilkinson et al.[21] and in a healthy population by Frimodt-Moller et al.[22]. Both the studies showed a high intraobserver and interobserver reproducibility. Our results are consistent with this regarding intraobserver reproducibility. However, our observer 2 found higher aPWV results than observer 1 and significantly so in the aPWV measurements based on the right carotid artery. This one-sided difference could maybe be explained by a few outliers in the dataset from observer 2. To exclude a possible effect of recording points, the two observers used the same recording sites in all individuals. By comparing these interobserver results with similar studies [21], we speculate that this variation could have been avoided by letting both the observers measure the distances separately, without marking the recording points. Nevertheless, our study suggests the use of the same observer whenever possible in the case of repeated measurements in intervention studies.

We chose, deliberately, only one femoral recording site in each individual. If both the right and the left femoral artery had been used corresponding to the carotid recording sites when measuring aPWV, it would have been impossible to determine whether upper or lower anatomy or vascular status caused a difference in aPWV. Our approach minimized any variation in aPWV arising from a possible anatomical difference between the right and the left femoral artery. Conversely, any difference in aPWV because of differences between the right and the left carotid artery could be demonstrated. The results were evaluated for the two observers separately and for the pooled data, and aPWV measurements were consistently higher when the right carotid artery was used. These findings strongly suggest that one side of the neck should be used, and although the ipsilateral or contralateral femoral artery could probably be randomly chosen, this should be confirmed by another study. Use of the femoral artery in the present study was based on which was most palpable for the first observer. Calculations were made for these two groups (right and left femoral artery) separately and showed the same tendencies as the pooled data. We can therefore exclude that the difference in aPWV measurements in our study could be because of the use of different femoral recording sites.

In our aPWV calculations, we used the subtracted carotid–femoral distance (carotid–suprasternal notch distance subtracted from the suprasternal notch–femoral distance). Weber et al.[23] showed that this estimation of pulse wave travel distance has the best agreement with invasive measurements of aPWV compared to various other distance-measuring techniques. Furthermore, PWV calculated using the suprasternal notch-to-femoral distance minus suprasternal notch-to-carotid distance provides the strongest relationship to cardiovascular mortality as shown by Nemeth et al.[24]. However, the recent guidelines [11] published after we designed our study advocate the use of the direct distance between the carotid and femoral arteries × 0.8. In order to see whether the use of direct distance × 0.8 altered the results of our study, we converted the measured subtracted distances into direct distances, using the equation published by Vermeersch et al.[15] and used by The Reference Values for Arterial Stiffness’ Collaboration group [25]. The aPWV values obtained using the calculated direct distance were higher than the aPWV values obtained using the subtracted distance, but the conversion did not change the overall finding of a higher aPWV when using the right compared to the left carotid artery (Tables 3 and 4). Obviously, the use of an equation is not ideal, and conversion of the measured subtracted distances will only introduce a conversion factor which of course alters the resulting PWV result but not the overall finding of a difference between the right and left carotid artery. Consequently, our results are limited by the use of subtracted distances. In order to fully clarify whether there is a side difference in aPWV values when using direct distances, we propose a new study with accurate measurement of the direct distance.

Although our study is the first to examine the side differences of aPWV measurements, the difference in plaque formation between the two carotid arteries has been examined and the results are conflicting. Chaubey et al.[26] examined the possibility that the effect of modifiable cardiovascular risk factors is different in the two carotid arteries in a cross-sectional population-based study, involving a random sample of 425 men and 367 women aged 56–77 years. The study investigated whether there was a side difference in the presence of plaques at the right and left carotid artery. They found that the prevalence of atherosclerotic plaques was most often bilateral in the carotid arteries, but in the cases of unilateral plaque formation there was a slightly higher incidence of plaques in the right carotid artery in both men and women. BP increases were associated with increased presence of plaque in the right carotid artery (9%) when compared to the left carotid artery (2%) (P = 0.016). Dalager et al.[27] performed an autopsy study on 100 individuals aged 20–82 years, in which 27 died of coronary artery disease. They found that the distribution of atherosclerotic lesions (plaques, foam cell and intermediate lesions) was higher in carotid than femoral arteries, but they did not find a difference between right and left carotids.

Thus, our findings of higher aPWV in the right carotid artery could be because of different lengths of the right and left arteries, respectively. However, a higher arterial stiffness on the right side seems plausible, suggesting increased plaque formation on the right side. It could be discussed whether the difference in aPWV for the pooled data of 0.2 ± 0.4 m/s (subtracted distance) is clinically relevant. It is, however, highly relevant for future intervention studies using aPWV as an endpoint to be aware of this side difference and use the same side of the neck and if possible the same observer.

In conclusion, our study demonstrated a higher aPWV when using the right compared to the left side of the neck in the same healthy individuals, which is novel. This could be because of the anatomical difference per se, regional increased stiffness on the right side because of the flow pattern or both. Our results provide evidence that aPWV should indeed be measured on the right side of the neck as suggested in the latest recommendations. However, awareness of the documented side difference is suggested.

Back to Top | Article Outline


The authors thank the staff at the Research Laboratory, Department of Renal Medicine, Aarhus University Hospital, Denmark, for their support and technical assistance.

This work was supported by the Danish Agency for Science and the Humanitarian Fund of Svend Fælding.

Back to Top | Article Outline

Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline

Reviewers’ Summary Evaluations Reviewer 1

The study was performed to test for differences in aortic pulse wave velocity depending on the carotid artery recorded from. The study shows that values on aortic pulse wave velocity are higher when recordings are taken from the right carotid artery and provides evidence that standardization of aortic pulse wave velocity needs to include the recording site to ensure reproducibility of measurements on aortic stiffness.

Back to Top | Article Outline

Reviewer 2

The reproducibility of an operator-dependent measurement, as is the case of aortic pulse wave velocity, is increased in direct proportion to the standardization of the method employed. This paper provides evidence to recommend the use of the same carotid (right carotid preferentially), when arterial stiffness is estimated by the use of a carotid femoral method. Although this study is based on a nonprobabilistic sample, in normal, nonhypertensive and relatively young people, their quality standards and the statistical method employed are sufficient to support this recommendation.

Back to Top | Article Outline


1. Laurent S, Boutouyrie P, Asmar R, Gautier I, Laloux B, Guize L, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001; 37:1236–1241.
2. Boutouyrie P, Tropeano AI, Asmar R, Gautier I, Benetos A, Lacolley P, et al. Aortic stiffness is an independent predictor of primary coronary events in hypertensive patients: a longitudinal study. Hypertension 2002; 39:10–15.
3. Laurent S, Katsahian S, Fassot C, Tropeano AI, Gautier I, Laloux B, et al. Aortic stiffness is an independent predictor of fatal stroke in essential hypertension. Stroke 2003; 34:1203–1206.
4. Blacher J, Guerin AP, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness on survival in end-stage renal disease. Circulation 1999; 99:2434–2439.
5. Blacher J, Safar ME, Guerin AP, Pannier B, Marchais SJ, London GM. Aortic pulse wave velocity index and mortality in end-stage renal disease. Kidney Int 2003; 63:1852–1860.
6. Meaume S, Benetos A, Henry OF, Rudnichi A, Safar ME. Aortic pulse wave velocity predicts cardiovascular mortality in subjects >70 years of age. Arterioscler Thromb Vasc Biol 2001; 21:2046–2050.
7. Mattace-Raso FU, van der Cammen TJ, Hofman A, van Popele NM, Bos ML, Schalekamp MA, et al. Arterial stiffness and risk of coronary heart disease and stroke: the Rotterdam Study. Circulation 2006; 113:657–663.
8. Mitchell GF, Hwang SJ, Vasan RS, Larson MG, Pencina MJ, Hamburg NM, et al. Arterial stiffness and cardiovascular events: the Framingham Heart Study. Circulation 2010; 121:505–511.
9. Nichols WW, O’Rourke MF, Vlachopoulos C. McDonald's blood flow in arteries – theoretical, experimental and clinical principles, 6th ed. Don MIlls, ON, Canada: Hodder Arnold Publishers; 2011.
10. Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, et al. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27:2588–2605.
11. Van Bortel LM, Laurent S, Boutouyrie P, Chowienczyk P, Cruickshank JK, De Backer T, et al. Expert consensus document on the measurement of aortic stiffness in daily practice using carotid–femoral pulse wave velocity. J Hypertens 2012; 30:445–448.
12. O’Brien E, Asmar R, Beilin L, Imai Y, Mancia G, Mengden T, et al. Practice guidelines of the European Society of Hypertension for clinic, ambulatory and self blood pressure measurement. J Hypertens 2005; 23:697–701.
13. Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves J, Hill MN, et al. Recommendations for blood pressure measurement in humans and experimental animals: Part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Hypertension 2005; 45:142–161.
14. Chiu YC, Arand PW, Shroff SG, Feldman T, Carroll JD. Determination of pulse wave velocities with computerized algorithms. Am Heart J 1991; 121:1460–1470.
15. Vermeersch SJ, Rietzschel ER, De Buyzere ML, Van Bortel LM, Gillebert TC, Verdonck PR, et al. Distance measurements for the assessment of carotid to femoral pulse wave velocity. J Hypertens 2009; 27:2377–2385.
16. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310.
17. Kirkwood BR, Sterne JAC. Essential medical statistics, 2nd ed. Oxford, UK: Blackwell Publishing; 2010.
18. Glagov S, Zarins C, Giddens DP, Ku DN. Hemodynamics and atherosclerosis. Insights and perspectives gained from studies of human arteries. Arch Pathol Lab Med 1988; 112:1018–1031.
19. Van Popele NM, Grobbee DE, Bots ML, Asmar R, Topouchian J, Reneman RS, et al. Association between arterial stiffness and atherosclerosis: the Rotterdam Study. Stroke 2001; 32:454–460.
20. Zureik M, Temmar M, Adamopoulos C, Bureau JM, Courbon D, Thomas F, et al. Carotid plaques, but not common carotid intima-media thickness, are independently associated with aortic stiffness. J Hypertens 2002; 20:85–93.
21. Wilkinson IB, Fuchs SA, Jansen IM, Spratt JC, Murray GD, Cockcroft JR, et al. Reproducibility of pulse wave velocity and augmentation index measured by pulse wave analysis. J Hypertens 1998; 16 (12 Pt 2):2079–2084.
22. Frimodt-Moller M, Nielsen AH, Kamper AL, Strandgaard S. Pulse-wave morphology and pulse-wave velocity in healthy human volunteers: examination conditions. Scand J Clin Lab Invest 2006; 66:385–394.
23. Weber T, Ammer M, Rammer M, Adji A, O’Rourke MF, Wassertheurer S, et al. Noninvasive determination of carotid–femoral pulse wave velocity depends critically on assessment of travel distance: a comparison with invasive measurement. J Hypertens 2009; 27:1624–1630.
24. Nemeth ZK, Studinger P, Kiss I, Othmane Tel H, Nemcsik J, Fekete BC, et al. The method of distance measurement and torso length influences the relationship of pulse wave velocity to cardiovascular mortality. Am J Hypertens 2011; 24:155–161.
25. Reference Values for Arterial Stiffness’ Collaboration. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’. Eur Heart J 2010; 31:2338–2350.
26. Chaubey S, Nitsch D, Altmann D, Ebrahim S. Differing effect of modifiable cardiovascular risk factors on intima-media thickening and plaque formation at different sites of the arterial vasculature. Heart 2010; 96:1579–1585.
27. Dalager S, Paaske WP, Kristensen IB, Laurberg JM, Falk E. Artery-related differences in atherosclerosis expression: implications for atherogenesis and dynamics in intima-media thickness. Stroke 2007; 38:2698–2705.

Cited By:

This article has been cited 1 time(s).

Journal of Hypertension
Methodological aspects in the measurement of pulse wave velocity by means of applanation tonometry
Salvi, P; Parati, G
Journal of Hypertension, 31(1): 35-38.
PDF (1454) | CrossRef
Back to Top | Article Outline

anatomical difference; arterial stiffness; carotid artery; pulse wave velocity; vascular remodelling

© 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins