This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Otto, S. J. Articles by de Koning, H. J. Search for Related Content PubMed Articles by Otto, S. J. Articles by de Koning, H. J. Social Bookmarking                            What's this?

Risk of Cardiovascular Mortality in Prostate Cancer Patients in the Rotterdam Randomized Screening Trial

Suzie J. Otto, Fritz H. Schröder, Harry J. de Koning

From the Department of Public Health and Department of Urology, Erasmus MC, University Medical Center Rotterdam, the Netherlands

Address reprint requests to Suzie J. Otto, PhD, Department of Public Health, Erasmus MC, University Medical Center Rotterdam, PO Box 2040, 3000 CA Rotterdam, the Netherlands, s.otto{at}erasmusmc.nl

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Purpose To estimate the risk of cardiovascular disease (CVD) mortality in prostate cancer patients in the Rotterdam section of European Randomized Study of Screening for Prostate Cancer, in both arms, and to compare this with the risk in the general population.

Methods Standardized mortality ratios (SMRs) of cardiovascular mortality for 2,211 prostate cancer patients were calculated including analyses for treatment subgroups (surgery, radiotherapy, watchful waiting, and hormone therapy). Cardiovascular mortality was defined as death as a result of all CVD and as a result of coronary heart disease, acute myocardial infarction, other diseases of the heart, and cerebrovascular accidents. The prevalence of self-reported comorbidities at entry of the trial was evaluated as well.

Results After a mean follow-up of 5.5 years, 258 prostate cancer patients (12%) had died. The SMR of all-cause mortality was 0.90 (95% CI, 0.79 to 1.01). The risk for cardiovascular mortality was low compared with that in the general population; the SMRs varied between 0.37 and 0.49. Low cardiovascular mortality risks were also seen within each treatment subgroup. CVD was the most frequently self-reported comorbidity at entry and prostate cancer patients undergoing radical prostatectomy reported the lowest rates (24%) compared with those receiving other therapies (40% to 42%).

Conclusion Although some self-selection has occurred, prostate cancer treatment did not increase the risk of dying as a result of cardiovascular causes in our cohort. The risk was significantly lower for all primary treatment modalities, suggesting that less emphasis should be put on CVD as a contraindication for aggressive (surgical) treatment for prostate cancer patients.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
During the last decade, prostate cancer has become the most common noncutaneous malignancy in many countries, accounting for 21% of all newly diagnosed cancers in males in the Netherlands.¹ Age-adjusted incidence increased from 62.8 per 100,000 men in 1989 to 93.2 in 2003. This rapid increase is likely the result of prostate-specific antigen (PSA) –based prostate cancer screening, which has become common practice,^2,3 despite the lack of evidence of the effectiveness of such screening. Currently, two randomized trials are underway in Europe (European Randomized Study of Screening for Prostate Cancer [ERSPC]) and the United States (the Prostate, Lung, Colorectal and Ovary cancer trials) to assess the effectiveness of prostate cancer screening, but the final outcomes have not yet been reported.^4,5

Generally, localized prostate cancer is treated by aggressive therapy intended to decrease the risk of disease recurrence/progression and cause-specific mortality.⁶ The choice of radical prostatectomy or radiotherapy often depends on stage, age, and presence of (chronic) comorbidities, due to possible perioperative morbidity and mortality. Concern for increased risk of long-term mortality from concurrent cardiovascular conditions has also been raised before for men treated for prostate cancer.^7-9 Satariano et al⁷ questioned whether prostate cancer or its treatment worsens the course of pre-existing cardiovascular disease (CVD).

In the ongoing randomized prostate cancer screening trials (ERSPC and Prostate, Lung, Colorectal and Ovary cancer trials), it is expected that possible differences in prostate cancer mortality between the screening and the control arms ultimately would be due to screening and early treatment, with the perspective of an equal pattern of other causes of death between the two arms. If prostate cancer treatment affected cardiovascular mortality adversely, an imbalance in the other causes of death pattern would arise between the two trial arms to the disadvantage of the screening arm, given that the number of cases detected and treated is far higher in the screening arm. For this reason, screening might be considered as hazardous, despite a possible beneficial effect on prostate cancer mortality reduction, particularly in view of the overdiagnosis and its effect on quality of life.

Therefore, the aim of this study primarily was to assess the cardiovascular mortality in the prostate cancer patients in the Rotterdam section of ERSPC in comparison with the general population, and secondarily to assess the cardiovascular mortality within the entire screening and control arm of the trial.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Study Population
The study population consists of the participants of the Rotterdam section of the ERSPC trial, which was started in June 1994, after a series of pilot studies were conducted in 1991 to 1993. Details of the study protocol and procedures have been described previously.¹⁰ Briefly, men age 55 to 74 years residing in the city of Rotterdam and 12 neighboring municipalities were identified through the corresponding population registries and invited to participate in the trial. After receipt of a full written informed consent form, men were randomly assigned to either the screening or control arm (recruitment rate was 49%). Between November 1993 and December 1999, 21,210 men were enrolled onto the screening arm and 21,166 men were enrolled onto in the control arm. These figures also include some men age 54 and 75 years (n = 454). Initially, the screening arm was offered PSA testing (cutoff, 4.0 ng/mL), digital rectal examination, and transrectal ultrasonography. Since February 1997, the initial and follow-up screenings in the Rotterdam center are based on PSA testing with a cutoff level of 3.0 ng/mL as the only referral for biopsy.¹¹

The attendance rate in the first round was 94%, and at the cutoff date of January 1, 2003, the attendance rate in the second round was approximately 81%,¹⁰ corrected for deceased men and those older than the upper age limit. Up to this cutoff date, a total of 2,227 patients with prostate cancer, either detected by screening or clinically diagnosed, have been registered in the trial database for the screening and control arm combined. Of these, 2,211 were age 55 to 74 at entry. Table 1 shows the characteristics of the prostate cancer patients by therapy. The most common initial treatment modality was radiotherapy (n = 888), followed by radical prostatectomy (n = 670), watchful waiting (n = 330), and hormone therapy (n = 120). At the time of these analyses, treatment modality was unknown for 203 of the prostate cancer patients. Almost all men undergoing radical prostatectomy or watchful waiting had localized tumors (92% and 93%, respectively), whereas hormone therapy seemed the initial treatment of choice for distant disease.

Vital status of the participants, up to December 31, 2002, was ascertained through record linkage with the Central Bureau of Genealogy (CBG) based on first (or initials) and last name, date of birth, and sex. To confirm correctness of the matched records, address, postal code, and place of residence reported by CBG for the deceased patients were compared with those registered in the trial database. Subsequently, death certificate number, date and place of death, and birth date were applied as key linkages to obtain the causes of death (underlying and contributory) from the Causes of Death Registry of Statistics Netherlands for the calendar years 1994 to 2002. The causes of death of seven deceased patients could not be retrieved (three in the screening arm and four in the control arm). Given that linkage with CBG is precise with respect to typographic errors, vital status of the nonmatched records was only occasionally established through linkage with the Causes of Death Registry of Statistics Netherlands using the unique 10-digit personal identification code.

Causes of death were coded according to the International Classification of Diseases (9th revision [ICD9]; 1994 to 1995) and ICD10 (from 1996 onward) and the grouping was based on the tabulation list for main primary causes of death of Statistics Netherlands.¹² All-cause mortality is the number of deaths from all causes. Cardiovascular mortality is defined as death from the circulatory system (codes 390 to 459/I00 to I99) and death from the subcategories coronary heart disease (CHD, codes 410 to 414/I20 to I25), acute myocardial infarction (codes 410/I21 to I22), the other forms of heart diseases (codes 420 to 423, 425 to 429/I30 to I33, I39 to I52; in our cohort, 90% of the cohort was assigned diagnostic codes relating to cardiomyopathy, cardiac arrest, arrhythmias, and heart failure), cerebrovascular accidents (codes 430 to 438/I60 to I69), and all of the remainder CVDs (93.5% of those classified with other CVDs consisted of diseases of the arterial and venous vessels, hypertensive diseases, and valve disorders) as underlying cause.

Statistical Analysis
Man-years of observation were calculated for each participant at the midpoint of each year in trial from random assignment until death or end of follow-up (December 31, 2002), according to the calculations of the ERSPC Data Monitoring Committee.¹³ The analyses were restricted to men age 55 to 74 years at entry; hence, 454 men age 54 or 75 years were excluded.

Comparisons of age at entry among the treatment groups were done with the one-way analysis of variance, post hoc corrected with Bonferroni, and the ² test was used to examine the tumor stages and the comorbid variable, which were categorized as none, one, or at least two conditions.

Cardiovascular mortality among men with prostate cancer and within the two trial arms was compared with the general male population by means of standardized mortality ratios (SMRs; ie, the ratio observed to expected number of deaths). The expected number of deaths was obtained by multiplying the mortality rates of the male population in the Rotterdam area (13 municipalities) in the calendar years 1994 to 2002 by the man-years of observation in the trial by 5-year age bands.¹² The 95% CIs of the SMRs were calculated according to the Byar's approximation.¹⁴ We also used the ² test to examine the relationship between having a cardiovascular-related comorbidity (self-reported at entry) and dying as a result of cardiovascular-related cause among deceased prostate cancer patients.

Cox regression analysis was used for comparison of CVD mortality between the screening and the control arm, with adjustment for age (continuous variable) and CVD comorbidity (dichotomous variable) at entry, computing the hazard ratios (HRs) and the corresponding 95% CIs. Both the SMRs and the HRs of cardiovascular mortality in the two trial arms were evaluated first including and second excluding the prostate cancer patients. The ² test, one-way analysis of variance and the Cox regression analyses were performed using SPSS statistical software, version 11.01s (SPSS Inc, Chicago, IL).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
After a mean follow-up time of 5.5 years, a total of 4,144 participants (9.9%) had died: 2,068 in the control arm (113,891 man-years) and 2,076 in the screening arm (114,615 man-years). Of these, 258 were among men with prostate cancer. The most common official causes of death in the entire trial cohort were cancer (40%) and CVD (35%), followed by diseases of the respiratory system (7%), reflecting the pattern observed in the Netherlands and the Rotterdam area (39%, 35%, and 7%, respectively).¹² These major causes of death were comparable between the screening and control arm, with age-adjusted HRs of 1.02 (95% CI, 0.92 to 1.12; P = .733), 0.99 (95% CI, 0.89 to 1.12; P = .869), and 1.06 (95% CI, 0.84 to 1.34; P = .630), respectively, as well as all-cause mortality (HR, 1.00; 95% CI, 0.94 to 1.07; P = .929).

The frequency of the self-reported diseases at entry is presented in Table 2. The number of participants who reported none, one, or at least two conditions was quite similar between the participants in the control and the screening arm, and different from the prevalence reported in the Permanent Survey of Living Conditions conducted by Statistics Netherlands.¹² The diseases considered showed a comparable pattern of distribution in the trial arms and the prostate cancer patients. The most frequently reported conditions were CVD (34.6% to 35.3%) and diabetes (4.4% to 5.6%). It is evident from the self-reported conditions that CVD and diseases of the respiratory system play a role in treatment choice.

View larger version (12K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 1. Standardized mortality ratio (SMR) of cardiovascular disease (CVD) in prostate cancer patients. All-cause mortality, death from all causes total; CHD, coronary heart disease; AMI, acute myocardial infarction; CVA, cerebrovascular accident. Other diseases of the heart are not shown because occurrences were observed only in the radiotherapy and the hormone therapy subgroup (SMR, 0.78; 95% CI, 0.31 to 1.61 and SMR, 0.57; 95% CI, 0.05 to 2.08, respectively).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Although the effectiveness of screening for prostate cancer is still unknown, its impact on disease-specific incidence¹⁵ and overdiagnosis¹⁶ has been well documented. The long-term treatment-related effects on urinary, bowel, and sexual functions,¹⁷ and the perioperative complications after radical prostatectomy,^18-22 are also well-known. In our study we assessed the risk of dying as a result of CVD among the cohort of prostate cancer patients in the Rotterdam section of the ERSPC trial. After a mean follow-up time of about 5.5 years (from random assignment onward), we found no indication of excess mortality from CVD among men with prostate cancer.

The prostate cancer patients initially were treated by surgery (30.3%), radiotherapy (40.2%), watchful waiting (15%), or hormone therapy (5%). Those who underwent surgery were on average younger and reported the lowest prevalence of cardiovascular comorbidities, and the majority had localized prostate cancer. This again emphasizes the preferential selection of younger, healthier patients with favorable tumors for surgery. The patients receiving hormone therapy had the most unfavorable tumor stages, 33% had metastatic disease at diagnosis compared with none in the other treatment groups, and were older than men treated surgically. Despite this dissimilarity in prognostic factors among the treatment groups, the observed cardiovascular-related deaths in all groups were lower than expected on the basis of the mortality rates in the general population. These results suggest that there is no convincing evidence that prostate cancer treatment modalities worsen possible concurrent cardiovascular conditions, as has been alluded to previously by Satariano et al.⁷ However, we should remark that Satariano et al⁷ compared prostate cancer patients who died as a result of prostate cancer with those who died from causes other than prostate cancer, whereas our comparison group is the general population. We also observed in the analyses among deceased prostate cancer patients that men with a cardiovascular comorbidity were more likely to die as a result of a cardiovascular-related cause.

When interpreting our findings, the potential effect of self-selection bias should be taken into account, given that recruitment in the Rotterdam section of ERSPC was based on volunteers. The current findings of the all-cause mortality by treatment group supports our previous observations in that our study cohort is somewhat healthier than the general population.^23,24 Except for the prostate cancer patients treated with hormone therapy, who had a three-fold increased risk of all-cause mortality likely due to the poor tumor stages at diagnosis, no excess all-cause mortality was observed in the other treatment groups. Comparison of the self-reported comorbidity patterns with the results of the Permanent Survey of Living Conditions conducted by Statistics Netherlands also emphasizes the selection issue, given that our participants seemed to have fewer self-reported comorbidities than the general Dutch population in same age group.

Our aforementioned findings suggest that prostate cancer treatment did not have an adverse effect on the all-cause mortality in the cohort of self-selected prostate cancer patients. Therefore, in a separate analysis, all participants enrolled onto the screening and the control arms were compared. We found that men enrolled onto the screening arm were at a high risk of dying as a result of acute myocardial infarction relative to their counterparts in the control arm. This high risk remained and even increased slightly from 17% to 32% (relative to the control arm) after exclusion of all prostate cancer patients from both trial arms and exclusion of those patients without prostate cancer in the screening arm who underwent biopsy. This observation instinctively may raise the concern of a possible unfavorable effect of the screening process itself (ie, PSA testing) and receipt of test results.²⁵ Nevertheless, the self-reported frequency of CVD at study entry was similar between the two trial arms and no excess cardiovascular deaths were observed in the screening arm or in the control arm in comparison with the general population, which refutes a possible harmful effect of the screening process. This is supported by the previous observation by Essink-Bot et al²³ that the screening process did not lead to considerable changes in health status or anxiety in screened men compared with nonparticipants.

Certain limitations should be considered in this study, particularly the use of self-reported data on comorbidities to explain the observed results. First, given that the prostate cancer treatments were not randomized prospectively, the findings should be interpreted with caution, and might be viewed as hypothesis generating. Second, men could have under-reported the prevalence of concurrent conditions, although chronic diseases, such as a CVD, would not likely be under-reported. Third, the development of other comorbidities during follow-up was not taken into account. However, it seems unlikely that aggravation or development of serious cardiovascular conditions would occur particularly in the screening arm. Data on comorbidity are collected for men in the screening arm at each screening visit, but we still lack comparative data on the control arm. Lastly, this study is based on data from death certificates, which might include deaths misattributed to CVD. Again, this would apply for both trial arms.

In conclusion, there was no evidence of a possible detrimental effect of prostate cancer diagnosis and subsequent treatment on cardiovascular mortality for any of the treatment modalities. Moreover, if age and tumor stage are taken into account, prostate cancer treatment did not adversely affect all-cause mortality in our cohort of self-selected prostate cancer patients within the ongoing trial; hence, aggressive therapy could be considered.

	Authors' Disclosures of Potential Conflicts of Interest

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Authors Employment Leadership Consultant Stock Honoraria Research Funds Testimony Other Fritz H. Schröder GlaxoSmithKline (A) GlaxoSmithKline (A); Schering (A)

Dollar Amount Codes (A) < $10,000 (B) $10,000-99,999 (C) > $100,000 (N/R) Not Required

	Author Contributions

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 

Conception and design: Suzie J. Otto, Fritz H. Schröder, Harry J. de Koning Financial support: Fritz H. Schröder, Harry J. de Koning Collection and assembly of data: Suzie J. Otto, Fritz H. Schröder Data analysis and interpretation: Suzie J. Otto, Fritz H. Schröder, Harry J. de Koning Manuscript writing: Suzie J. Otto, Fritz H. Schröder, Harry J. de Koning Final approval of manuscript: Suzie J. Otto, Fritz H. Schröder, Harry J. de Koning

 

	ACKNOWLEDGMENTS

 
We thank Statistics Netherlands (Causes of Death Registry) for supplying the mortality data.

	NOTES

 
Supported by ZonMw and the Dutch Cancer Society, and grants from Europe against Cancer and the 5th Framework Program of the European Union to the Rotterdam section of the ERSPC.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
1. Association of Comprehensive Cancer Centres: Cancer in the Netherlands, incidence and mortality. http://www.cancerregistry.nl

2. Sirovich BE, Schwartz LM, Woloshin S: Screening men for prostate and colorectal cancer in the United States: Does practice reflect the evidence? JAMA 289:1414-1420, 2003[Abstract/Free Full Text]

3. Melia J, Moss S, Johns L: Rates of prostate-specific antigen testing in general practice in England and Wales in asymptomatic and symptomatic patients: A cross-sectional study. BJU Int 94:51-56, 2004[CrossRef][Medline]

4. de Koning HJ, Auvinen A, Berenguer Sanchez A, et al: Large-scale randomized prostate cancer screening trials: Program performances in the European randomized screening for prostate cancer trial and the prostate, lung, colorectal and ovary cancer trial. Int J Cancer 97:237-244, 2002[CrossRef][Medline]

5. Prorok PC, Andriole GL, Bresalier RS, et al: Design of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Control Clin Trials 21:273S-309S, 2000[CrossRef][Medline]

6. Holmberg L, Bill-Axelson A, Helgesen F, et al: A randomized trial comparing radical prostatectomy with watchful waiting in early prostate cancer. N Engl J Med 347:781-789, 2002[Abstract/Free Full Text]

7. Satariano WA, Ragland KE, Van Den Eeden SK: Cause of death in men diagnosed with prostate carcinoma. Cancer 83:1180-1188, 1998[CrossRef][Medline]

8. Post PN, Hansen BE, Kil PJ, et al: The independent prognostic value of comorbidity among men aged < 75 years with localized prostate cancer: A population-based study. BJU Int 87:821-826, 2001[CrossRef][Medline]

9. Fouad MN, Mayo CP, Funkhouser EM, et al: Comorbidity independently predicted death in older prostate cancer patients, more of whom died with than from their disease. J Clin Epidemiol 57:721-729, 2004[CrossRef][Medline]

10. Roobol MJ, Kirkels WJ, Schroder FH: Features and preliminary results of the Dutch centre of the ERSPC (Rotterdam, the Netherlands). BJU Int 92:48-54, 2003 (suppl 2)

11. Beemsterboer PM, Kranse R, de Koning HJ, et al: Changing role of 3 screening modalities in the European randomized study of screening for prostate cancer (Rotterdam). Int J Cancer 84:437-441, 1999[CrossRef][Medline]

12. Statistics Netherlands: Statline. http://statline.cbs.nl/

13. Smith PH: The Data Monitoring Committee: Bridging the gap between urology and public health epidemiology. BJU Int 92:55-56, 2003 (suppl 2)[CrossRef][Medline]

14. Breslow NE, Day NE: Statistical Methods in Cancer Research: The Design and Analysis of Cohort Studies. Lyon, France, IARC, 1987

15. Hankey BF, Feuer EJ, Clegg LX, et al: Cancer surveillance series: Interpreting trends in prostate cancer, part I. Evidence of the effects of screening in recent prostate cancer incidence, mortality, and survival rates. J Natl Cancer Inst 91:1017-1024, 1999[Abstract/Free Full Text]

16. Draisma G, Boer R, Otto SJ, et al: Lead times and overdetection due to prostate-specific antigen screening: Estimates from the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst 95:868-878, 2003[Abstract/Free Full Text]

17. Korfage IJ, Essink-Bot ML, Borsboom GJ, et al: Five-year follow-up of health-related quality of life after primary treatment of localized prostate cancer. Int J Cancer 116:291-296, 2005[CrossRef][Medline]

18. Lu-Yao GL, McLerran D, Wasson J, et al: An assessment of radical prostatectomy: Time trends, geographic variation, and outcomes—The Prostate Patient Outcomes Research Team. JAMA 269:2633-2636, 1993[Abstract/Free Full Text]

19. Dillioglugil O, Leibman BD, Leibman NS, et al: Risk factors for complications and morbidity after radical retropubic prostatectomy. J Urol 157:1760-1767, 1997[CrossRef][Medline]

20. Begg CB, Riedel ER, Bach PB, et al: Variations in morbidity after radical prostatectomy. N Engl J Med 346:1138-1144, 2002[Abstract/Free Full Text]

21. Froehner M, Koch R, Litz R, et al: Preoperative cardiopulmonary risk assessment as predictor of early noncancer and overall mortality after radical prostatectomy. Urology 61:596-600, 2003[CrossRef][Medline]

22. Alibhai SM, Leach M, Tomlinson G, et al: 30-day mortality and major complications after radical prostatectomy: Influence of age and comorbidity. J Natl Cancer Inst 97:1525-1532, 2005[Abstract/Free Full Text]

23. Essink-Bot ML, de Koning HJ, Nijs HGT, et al: Short-term effects of population-based screening for prostate cancer on health-related quality of life. J Natl Cancer Inst 90:925-931, 1998[Abstract/Free Full Text]

24. Otto SJ, Schroder FH, de Koning HJ: Low all-cause mortality in the volunteer-based Rotterdam section of the European randomised study of screening for prostate cancer: Self-selection bias? J Med Screen 11:89-92, 2004[CrossRef][Medline]

25. Black WC, Haggstrom DA, Welch HG: All-cause mortality in randomized trials of cancer screening. J Natl Cancer Inst 94:167-173, 2002[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Otto, S. J. Articles by de Koning, H. J. Search for Related Content PubMed Articles by Otto, S. J. Articles by de Koning, H. J. Social Bookmarking                            What's this?