PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
BJU Int. Author manuscript; available in PMC 2013 July 18.
Published in final edited form as:
PMCID: PMC3715084
NIHMSID: NIHMS440869

Prostate angiosarcoma: is there any association with previous radiation therapy?

Abstract

For the current review a literature search was carried out using Pubmed, EmBase, and Cochrane databases. All cases of prostate angioscaroma reported to date and observational studies evaluating the radiation associated cancer occurrence were reviewed. Despite the rarity, prostate angiosarcomas display remarkable clinical and pathological heterogeneity, and a treatment challenge. We found the association of prostate angiosarcoma with radiation therapy to be weak based upon the results from observational studies and case reports. Although radiation exposure has been suggested etiology of prostate angiosarcomas, assumption of such association is not supported by the current literature.

Keywords: prostate angiosarcoma, radiation therapy, adjuvant chemotherapy, multidisciplinary management

Sarcomas are malignant mesenchymal tumours. Angiosarcomas usually effect older individuals and encompass 2% of soft-tissue sarcomas. They frequently involve the skin, breast and soft tissue [1]. These very rare malignant neoplasms (only 10 identified in the current world literature) originate from the blood vessel endothelium and are distinguished by atypical, solid or multilayered endothelial proliferation [1]. In the present paper, we provide a review of prostate angiosarcomas, highlighting their epidemiology, aetiology, clinical presentation, histological features, prognostic factors and current treatment options.

EPIDEMIOLOGY

Rhabdomyosarcomas are the most frequent sarcomas of the prostate, accounting for >75% of cases, and are typically seen in infants, children and young adults [2,3]. Prostate angiosarcoma is an almost non-existent entity and the disease progression and prognosis of these tumours is poorly understood. A peak incidence has been noted between the ages of 40–50 and 70–80 years (mean age: 40 years) in the literature. One case even involved a 2-year-old child. Smith et al. [4] reported two cases and found two additional cases in the literature from 1889 to 1986 [5,6]. Chan et al. [7] reported a fifth case of prostate angiosarcoma in a 35-year-old Chinese man, and Oliva Encina et al. [8] found a sixth case in a 31-year-old male. Chandan and Wolsh [9] reported the seventh case of angiosarcoma of the prostate in a patient who received radiation therapy for prostate adenocarcinoma 10 years earlier. Lee et al. [10] reported the eighth case, which was a 19-year-old male with prostate teratoma, resistant to chemotherapy, who was later found to have prostatic angiosarcoma. Guo et al. [11] found the ninth case of angiosarcoma in a patient who had undergone radiation therapy 4 years earlier. Khaliq et al. [12] reported the 10th case, a 73-year-old man who had undergone external beam radiation therapy with brachytherapy boost for prostate adenocarcinoma 8 years previously and later presented with prostate angiosarcoma along with recurrent adenocarcinoma.

AETIOLOGY AND RISK FACTORS

Previous radiation exposure is a well-known risk factor for angiosarcomas. The direct oncogenic effects of ionizing radiation and prolonged cellular stimulation during repair of tissue damage resulting from radiation-induced ischaemic change are thought to play a role in the development of angiosarcoma [13]. Other factors that have been linked to angiosarcomas include: chronic lymphoedema and chemical exposure such as arsenic, thorium dioxide and vinyl chloride. None of the reported cases had a history of exposure to the above-mentioned chemicals.

Cahan et al. [14] proposed that radiation-induced sarcoma may occur in a previously irradiated area within a latent period of as long as 7 years. Although the association of angiosarcoma with radiation exposure has been described previously [15,16], only three out of 10 reported cases of prostate angiosarcoma had previous radiation exposure where PSA levels were within normal range or undetectable [9,11,12].

Only one of these three cases involved a recurrent adenocarcinoma of the prostate [12]. In the case report by Lee et al. [10], it was thought to be a malignant transformation within a pre-existing teratoma lesion; therefore, it is not clear whether post-radiation angiosarcoma stems from a dedifferentiated prostate cancer or signifies instead a second mesenchymal neoplasm. The estimated lifetime risk of developing post-irradiation sarcoma at any site with long-term follow-up appears to be 0.03% to 0.8% [17].

Huang et al. [18] reported an enhanced risk of soft-tissue sarcomas after adjuvant radiotherapy among patients with breast cancer in Surveillance of Epidemiology and End Results (SEER) data. This risk was especially increased for angiosarcomas and peak incidence was reported 5–10 years after the radiation therapy [18]. A concurrent lymphoedema, secondary to breast cancer treatment, was thought to be a potential confounder in this association [19,20]. A similar association between radiotherapy and subsequent angiosarcomas has also been reported in the gynaecological cancer literature [19,21]. Kim et al. [21] found 66 reported cases of radiation-associated angiosarcoma where the most common indication for radiation therapy was breast cancer (44%), followed by gynaecological cancer (21%). Kim et al. [21] also reported that 85% of radiation-associated angiosarcomas were detected in the skin with a median latency period of 8 years. In the same study, the median age at diagnosis was 65 years and median survival was 12 months [21].

Another large population-based cohort study evaluating the risk of angiosarcoma among all patients with cancer found an increased risk of truncal angiosarcoma among women with breast and gynaecological cancers [19]. This study did not find a strong relationship with radiotherapy, age or male gender. Although ionizing radiation is a well documented aetiology for angiosarcoma, Müller et al. [22] found no increased risk of secondary malignancy in the literature from the mid-1980s to 2007 after adjustment for age and follow-up duration; however, they reported an enhanced number of secondary cancers of the bladder, rectum, lung and sarcoma after prostate irradiation.

Moon et al. [23] reported enhanced risk of second primary cancer of the bladder, rectum, gastrointestinal tract, brain and lung, lymphoma and leukaemia among patients with prostate cancer 5 years after radiation therapy as compared with those who did not receive radiation in SEER data. In the same study, men who received radiation therapy in the form of radioactive implants or isotopes did not have an increased risk of a second primary cancer [23]. In this large cohort study, no enhanced risk of prostate angiosarcoma was reported.

CLINICAL PRESENTATION AND DIAGNOSTIC EVALUATION

Primary symptoms of prostate angiosarcoma include dysuria, haematuria and pain. Other reported symptoms are urinary frequency, decreased urinary stream, bladder spasm, constipation, weight loss, decreased libido, hematospermia and perineal pain (Table 1 [4-12]). Physical examination of the prostate was only significant or reported in four cases. Affected prostate glands were described as enlarged, firm, tender, boggy and containing a large solid mass. Diagnostic assessment was made using TURP in three cases, radical cystoprostatectomy in three cases, and incisional biopsy, rectal biopsy, and prostate biopsy in other cases. Seven out of 10 cases had no metastases and one case was lost to follow-up. One case had local metastases and one had metastases to the stomach, lung, spleen and liver. Three cases had a history of prostate adenocarcinoma with previous prostatic radiation therapy; however, serum PSA levels were either low or undetectable in those cases (Table 1).

TABLE 1
Case reports of prostate angiosarcoma in the literature

PATHOLOGICAL FINDINGS

Although a hallmark of angiosarcomas is abnormal, pleomorphic, malignant endothelial cells, other variants that have been reported include: round, polygonal, fusiform or epithelioid cells [24]. In well-differentiated cancer, abnormal endothelial cells form vascular sinusoids dissecting between collagen bundles and are associated with monocyte infiltration. In aggressive disease, the tissue architecture becomes more chaotic, forming multilayered papillary projections in the vascular lumen. Mitotic bodies and cytoplasmic clusters of erythrocytes are also common in aggressive disease. In poorly differentiated tumours, areas of haemorrhage and necrosis among the continuous sheets of malignant endothelium impair the ability to differentiate these tumours from anaplastic carcinoma or melanoma [25]. Immunohistochemistry of the tumour is essential for diagnosis because angiosarcomas express endothelial markers like vascular endothelial growth factor (VEGF), Ulex europaeus agglutinin 1, factor VIII, CD31, and CD34. Factor VIII, Ulex europaeus agglutinin 1, and CD31 are the most useful markers in poorly differentiated angiosarcomas [26]. Among review of the reported cases, positive factor VIII staining was reported in five cases, positive CD34 staining in three cases, positive CD31 staining in three cases and positive vimentin in two cases (Table 1).

PROGNOSTIC FACTORS

Mortality was very high for prostate angiosarcoma in the reported cases (Table 1). Six patients died within 9 months of diagnosis; three were disease-free for a period of 16, 24 and 36 months; and one patient was lost to follow-up. Five-year survival is 50–60% for primary soft-tissue sarcoma [27] and 35% for angiosarcomas at any site [28-30]. Suspected poor prognostic factors for other soft-tissue sarcomas are tumour size (>5 cm) and grade, advanced age, visceral and retroperitoneal locations, metastases and poor patient performance status [25].

TREATMENT

Management of angiosarcoma is a challenge. The two most important determinants of treatment options are size of tumour and presence of metastases. The small number of cases and lack of clinical trials are the major limitations for site-specific angiosarcoma treatment recommendations. Treatment for prostate angiosarcoma follows the general guidelines for other soft-tissue sarcoma management as suggested by the European Society for Medical Oncology (ESMO) and the National Comprehensive Cancer Network (NCCN). Although several vascular-targeted therapies are being tested in clinical trials, the present review focuses primarily on current available management. Among reported cases of prostate angiosarcomas, radical cystoprostatectomy was performed in six patients, whereas two patients had subtotal resections of the prostate (probably as palliative therapy only). One patient died shortly after hospitalization and one was lost to follow-up. Out of six cases where radical cystoprostatectomy was performed, clean surgical margins were only reported in one case. Three patients with radical cystoprostatectomy received adjuvant chemotherapy, consisting of doxorubicin in one case, experimental thalidomide in one case, and ifosfamide plus doxorubicin in the other. None of the reported cases received radiation therapy as a part of multimodal treatment because of suspected radiation-induced angiosarcoma.

LOCAL DISEASE

The primary treatment of choice for localized prostate angiosarcoma is radical surgery with complete (R0) resection and clean margins, but clean margins with R1 or R2 resection can sometimes be a challenge because of tumour size, tissue invasion, and relationship to adjacent critical organs, all of which confer poor prognosis [28,30-32]. Lahat et al. [33] reported improved survival among patients with recurrent angiosarcoma, who underwent surgery with pathological complete resection for localized disease. Although adjuvant radiotherapy with larger doses (>50 Gy) is generally recommended for treatment of sarcoma owing to the high risk of local recurrence, randomized radiotherapy trials have not been conducted. However, evidence from retrospective series suggests improved local control and overall survival with adjuvant radiotherapy [32]. Nevertheless, radiation therapy has not been recommended for radiation-induced angiosarcomas.

Use of adjuvant chemotherapy for risk reduction of metastasis in sarcoma has also been controversial owing to lack of clinical trials. Naka et al. [34] reported extended survival in patients who received adjuvant actinomycin-D after surgery although the number of patients was very small. We did not find prospective studies suggesting improved survival with the use of anthracycline-based adjuvant chemotherapy in soft-tissue sarcomas [25]. The only trial showing a survival benefit is the Italian and Scandinavian Sarcoma Groups study that focused on high-grade non-metastatic extremity osteosarcomas [35]. A meta-analysis of randomized controlled trials of adjuvant chemotherapy use for localized resectable soft-tissue sarcoma found an overall survival benefit only with combined doxorubicin and ifosfamide [36]. However, given the uncertainty surrounding this issue, the NCCN guidelines still recommend evaluating chemotherapy on a case-by-case basis. Use of chemotherapy for abdominal or retroperitoneal sarcomas has also not shown any survival benefit. Hence the use of chemotherapy in the neoadjuvant or adjuvant setting after definitive surgery for angiosarcoma remains controversial.

METASTATIC DISEASE

Cytotoxic chemotherapy with anthracyclines, ifosfamide, and taxanes has been the primary treatment for metastatic angiosarcoma. There are no evidence-based recommendations for treatment of metastatic prostate angiosarcoma because of small numbers of cases and associated comorbidities limiting use of chemotherapeutic agents. Use of combination therapy has been associated with enhanced toxicity and failed to improve the overall survival [27]. The use of doxorubicin and ifosfamide as single agents has demonstrated response rates of 16–36% in soft-tissue sarcoma [27]. Taxanes have been viewed as promising therapy for treatment of angiosarcoma owing to their anti-angiogenic properties, but the response rates have been variable in other soft-tissue sarcomas [37,38].

There are a few case reports about thalidomide (an anti-angiogenic and immunomodulatory agent) use in advanced or metastatic angiosarcoma [39]. Other treatments with encouraging outcomes include the VEGF-A monoclonal antibody, bevacizumab [40] and interferon-α [41]. Sorafenib, a VEGF receptor small molecule inhibitor, has shown promising results against metastatic angiosarcoma [42]. Interferon-α, an immune modulator with anti-angiogenic activity has shown inhibition of transformed murine endothelial cell lines in preclinical studies [41]. There are also some case reports documenting a response to interferon-α in combination with doxorubicin for advance cutaneous angiosarcoma [43]. Interleukin-2 has also been used as single agent, or in combination with chemotherapy or radiotherapy, suggesting that combined systemic and local treatment might improve survival [44]. Interferon-α and interleukin-2 use is currently not recommended for treatment outside clinical trials owing to insufficient data.

CONCLUSION

From a review of the literature, including both prospective and retrospective studies, we did not find a strong association between radiation therapy and prostatic angiosarcoma although this possibility cannot be excluded. Out of all the cases of prostate angiosarcoma, only three of 10 patients had a history of radiation therapy for prostate adenocarcinoma. Currently the main curative treatment method for prostatic angiosarcoma is wide surgical resection with clean margins. However, wide resection with clean margins can be a challenge because of the invasive nature of the tumour and proximity to critical organs. In addition, radiation therapy has never been used to treat prostatic angiosarcoma as recommended by general treatment guidelines for sarcoma by NCCN and ESMO. Despite a significant risk of subsequent metastatic disease, use of adjuvant chemotherapy for localized sarcomas or angiosarcoma is also not recommended as it has not shown any survival benefit. Based on available evidence, metastatic prostatic angiosarcoma can be treated with single-agent doxorubicin, paclitaxel, sorafenib or bevacizumab depending upon the functional status of the patient. At this time, patients should be offered participation in a clinical trial when available. Barring trial participation, patients should be referred to specialist centres where they can be managed by a team of radiation, medical and surgical oncologists.

What’s known on the subject? and What does the study add?

Angiosarcomas are histological subtype of sarcomas and rarely involve the prostate gland. Only ten cases of prostate angiosarcoma have been reported in the literature to date. Occurrence of post-irradiation prostate angiosarcoma is rare considering the frequency of radiotherapy used for treatment of prostate adenocarcinoma. We provide a brief review of all cases of prostate angiosarcoma and describe the epidemiology, etiology, clinical presentation, histopathology, prognostic factors and current treatment options for prostate angiosarcoma.

Acknowledgments

The authors would like to acknowledge Regina Landis, Dr Alejandro Necochea and Dr Charles Guo for their assistance with the manuscript preparation.

Abbreviations

SEER
Surveillance of Epidemiology and End Results
VEGF
vascular endothelial growth factor
ESMO
European Society for Medical Oncology
NCCN
National Comprehensive Cancer Network

Footnotes

CONFLICT OF INTEREST

None declared.

References

1. Schoen FJ, Mitchell NR. Blood vessels. In: Kumar V, Abbas AK, Fausto N, Aster JC, editors. Robbins and Cotran Pathologic Basis of Disease. 8. Philadelphia: W.B. Saunders Company; 2009. pp. 487–527.
2. Young RH, Srigley JR, Amin M, et al. Tumors of the Prostate Gland, Seminal Vesicles, Male Urethra and Penis. Washington, DC: Armed Forces Institute of Pathology; 2000. pp. 257–88.Atlas of Tumor Pathology. fascicle 10(3rd series)
3. Hansel DE, Herawi M, Montgomery E, Epstein JI. Spindle cell lesions of the adult prostate. Mod Pathol. 2007;20:148–58. [PubMed]
4. Smith DM, Manivel C, Kapps D, Uecker J. Angiosarcoma of the prostate: report of 2 cases and review of the literature. J Urol. 1986;135:382–84. [PubMed]
5. Botesco M. Sarcome primitif de la prostate chez un enfant de deux ans et neuf mois. Bull Mem Soc Chir Bucares. 1902;5:77.
6. Salleras J, Vilar G. Consideraciones a proposito de un sarcoma primitivo de la prostata en un sujeto de 32 anos prostatectomia perineal. Semana Med Buenos Aires. 1924;31:1275–9.
7. Chan KW. Angiosarcoma of the prostate. An immunohistochemical study of a case. Pathology. 1990;22:108–10. [PubMed]
8. Oliva Encina J, Gil Martinez P, Allepuz Losa C, Andrés Lázaro V, Rioja Sanz LA. Prostatic angiosarcoma: report of a new case. Actas Urol Esp. 2001;25:78–80. [PubMed]
9. Chandan VS, Wolsh L. Postirradiation angiosarcoma of the prostate. Arch Pathol Lab Med. 2003;127:876–78. [PubMed]
10. Lee HM, Song SY, Park JO, Kim BH. Primary immature teratoma of the prostate with angiosarcoma component: its unusual response to chemotherapy. Int J Urol. 2006;13:305–7. [PubMed]
11. Guo CC, Pisters LL, Troncoso P. Prostate cancer invading the rectum: a clinicopathological study of 18 cases. Pathology. 2009;41:539–43. [PubMed]
12. Khaliq W, Meyer CF, Uzoaru I, Wolf RM, Antonarakis ES. Prostate angiosarcoma: A case report and literature review. Med Oncol. 2012 doi: 10.1007/s12032-012-0188-x. in press. [PMC free article] [PubMed] [Cross Ref]
13. Chen KT, Hoffman KD, Hendricks EJ. Angiosarcoma following therapeutic irradiation. Cancer. 1979;44:2044–48. [PubMed]
14. Cahan WG, Woodard HQ, Higinbotham NL, Stewart FW, Coley BL. Sarcoma arising in irradiated bone. Cancer. 1948;1:3–29. [PubMed]
15. Nanus DM, Kelsen D, Clark DG. Radiation-induced angiosarcoma. Cancer. 1987;60:777–79. [PubMed]
16. Navon JD, Rahimzadeh M, Wong AK, Carpenter PM, Ahlering TE. Angiosarcoma of the bladder after therapeutic irradiation for prostate cancer. J Urol. 1997;157:1359–60. [PubMed]
17. Mark RJ, Poen J, Tran LM, Fu YS, Selch MT, Parker RG. Postirradiation sarcomas: a single-institution study and review of the literature. Cancer. 1994;73:2653–62. [PubMed]
18. Huang J, Mackillop WJ. Increased risk of soft tissue sarcoma after radiotherapy in women with breast carcinoma. Cancer. 2001;92:172–80. [PubMed]
19. Virtanen A, Pukkala E, Auvinen A. Angiosarcoma after radiotherapy: a cohort study of 332,163 Finnish cancer patients. Br J Cancer. 2007;97:115–17. [PMC free article] [PubMed]
20. Karlsson P, Holmberg E, Samuelsson A, Johansson KA, Wallgren A. Soft tissue sarcoma after treatment for breast cancer–a Swedish population-based study. Eur J Cancer. 1998;34:2068–75. [PubMed]
21. Kim MK, Huh SJ, Kim DY, et al. Secondary angiosarcoma following irradiation: case report and review of the literature. Radiat Med. 1998;16:55–60. [PubMed]
22. Müller AC, Ganswindt U, Bamberg M, Belka C. Risk of Second Malignancies after Prostate Irradiation? Strahlenther Onkol. 2007;183:605–9. [PubMed]
23. Moon K, Stukenborg GJ, Keim J, Theodorescu D. Cancer incidence after localized therapy for prostate cancer. Cancer. 2006;107:991–98. [PubMed]
24. Koch M, Nielsen GP, Yoon SS. Malignant tumors of blood vessels: angiosarcomas, hemangioendotheliomas, and hemangioperictyomas. J Surg Oncol. 2008;97:321–9. [PubMed]
25. Young RJ, Brown NJ, Reed MW, Hughes D, Woll PJ. Angiosarcoma. Lancet Oncol. 2010;11:983–91. [PubMed]
26. Ohsawa M, Naka N, Tomita Y, Kawamori D, Kanno H, Aozasa K. Use of immunohistochemical procedures in diagnosing angiosarcoma. Evaluation of 98 cases. Cancer. 1995;75:2867–74. [PubMed]
27. Mocellin S, Rossi CR, Brandes A, Nitti D. Adult soft tissue sarcomas: conventional therapies and molecularly targeted approaches. Cancer Treat Rev. 2006;32:9–27. [PubMed]
28. Fury MG, Antonescu CR, Van Zee KJ, Brennan MF, Maki RG. A 14-year retrospective review of angiosarcoma: clinical characteristics, prognostic factors, and treatment outcomes with surgery and chemotherapy. Cancer J. 2005;11:241–47. [PubMed]
29. Mark RJ, Poen JC, Tran LM, Fu YS, Juillard GF. Angiosarcoma. A report of 67 patients and a review of the literature. Cancer. 1996;77:2400–06. [PubMed]
30. Fayette J, Martin E, Piperno-Neumann S, et al. Angiosarcomas, a heterogeneous group of sarcomas with specific behavior depending on primary site: a retrospective study of 161 cases. Ann Oncol. 2007;18:2030–36. [PubMed]
31. Abraham JA, Hornicek FJ, Kaufman AM, et al. Treatment and outcome of 82 patients with angiosarcoma. Ann Surg Oncol. 2007;14:1953–67. [PubMed]
32. Pawlik TM, Paulino AF, McGinn CJ, et al. Cutaneous angiosarcoma of the scalp: a multidisciplinary approach. Cancer. 2003;98:1716–26. [PubMed]
33. Lahat G, Dhuka AR, Lahat S, et al. Outcome of locally recurrent and metastatic angiosarcoma. Ann Surg Oncol. 2009;16:2502–59. [PubMed]
34. Naka N, Ohsawa M, Tomita Y, et al. Prognostic factors in angiosarcoma: a multivariate analysis of 55 cases. J Surg Oncol. 1996;61:170–76. [PubMed]
35. Ferrari S, Smeland S, Mercuri M, et al. Neoadjuvant chemotherapy with high-dose Ifosfamide, high-dose methotrexate, cisplatin, and doxorubicin for patients with localized osteosarcoma of the extremity: a joint study by the Italian and Scandinavian Sarcoma Groups. J Clin Oncol. 2005;23:8845–52. [PubMed]
36. Pervaiz N, Colterjohn N, Farrokhyar F, Tozer R, Figueredo A, Ghert M. A systematic meta-analysis of randomized controlled trials of adjuvant chemotherapy for localized resectable soft-tissue sarcoma. Cancer. 2008;113:573–81. [PubMed]
37. Verweij J, Lee SM, Ruka W, et al. Randomized phase II study of docetaxel versus doxorubicin in first- and second-line chemotherapy for locally advanced or metastatic soft tissue sarcomas in adults: a study of the european organization for research and treatment of cancer cancer soft tissue and bone sarcoma group. J Clin Oncol. 2000;18:2081–86. [PubMed]
38. Penel N, Bui BN, Bay JO, et al. Phase II trial of weekly paclitaxel for unresectable angiosarcoma: the ANGIOTAX Study. J Clin Oncol. 2008;26:5269–74. [PubMed]
39. Raina V, Sengar M, Shukla NK, et al. Complete response from thalidomide in angiosarcoma after treatment of breast cancer. J Clin Oncol. 2007;25:900–01. [PubMed]
40. Koontz BF, Miles EF, Rubio MA, et al. Preoperative radiotherapy and bevacizumab for angiosarcoma of the head and neck: two case studies. Head Neck. 2008;30:262–66. [PubMed]
41. Taylor KL, Oates RK, Grane R, Leaman DW, Borden EC, Lindner DJ. IFN-alpha1,8 inhibits tumor-induced angiogenesis in murine angiosarcomas. J Interferon Cytokine Res. 2006;26:353–61. [PubMed]
42. Maki RG, D’Adamo DR, Keohan ML, et al. Phase II study of sorafenib in patients with metastatic or recurrent sarcomas. J Clin Oncol. 2009;27:3133–40. [PMC free article] [PubMed]
43. Burns BT, Blakey SA, Harris WB. Complete response of metastatic angiosarcoma to liposomal doxorubicin and interferon-2a. Proc Am Soc Clin Oncol. 2002;21:2939. abstr.
44. Ohguri T, Imada H, Nomoto S, et al. Angiosarcoma of the scalp treated with curative radiotherapy plus recombinant interleukin- 2 immunotherapy. Int J Radiat Oncol Biol Phys. 2005;61:1446–53. [PubMed]