|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Increased Risk of Ischemic Stroke After Radiotherapy on the Neck in Patients Younger Than 60 YearsByFrom the Department of Neuro-Oncology, Epidemiology, Head and Neck Oncology, and Radiotherapy, the Netherlands Cancer Institute/Antoni van Leeuwenhoekhuis, Amsterdam, the Netherlands. Address reprints requests to Willem Boogerd, MD, PhD, Department of Neuro-Oncology, the Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands; email: newbo{at}slz.nl
PURPOSE: To estimate the risk of ischemic stroke in patients irradiated for head and neck tumors. PATIENTS AND METHODS: The incidence of ischemic stroke was determined in 367 patients with head and neck tumors (162 larynx carcinomas, 114 pleomorphic adenomas, and 91 parotid carcinomas) who had been treated with local radiotherapy (RT) at an age younger than 60 years. Relative risk (RR) of ischemic stroke was determined by comparison with population rates from a stroke-incidence register, adjusted for sex and age. Other risk factors for stroke (hypertension, smoking, hypercholesterolemia, diabetes mellitus [DM]) were registered. The median follow-up time after RT was 7.7 years (3,011 person-years of follow-up). RESULTS: Fourteen cases of stroke occurred (expected, 2.5; RR, 5.6; 95% confidence interval [CI], 3.1 to 9.4): eight in patients with laryngeal carcinoma (expected ,1.56; RR, 5.1; 95% CI, 2.2 to 10.1), four in pleomorphic adenoma patients (expected, 0.71; RR, 5.7; 95% CI, 1.5 to 14.5), and two in parotid carcinoma patients (expected, 0.24; RR, 8.5, 95% CI, 1.0 to 30.6). Five of six strokes in patients irradiated for a parotid tumor occurred at the ipsilateral side. Analysis of other risk factors for cerebrovascular disease showed hypertension and DM to cause an increase of the RR after RT. After more than 10 years’ follow-up, the RR was 10.1 (95% CI, 4.4 to 20.0). The 15-year cumulative risk of stroke after RT on the neck was 12.0% (95% CI, 6.5% to 21.4%). CONCLUSION: This is the first study to demonstrate an increased risk of stroke after RT on the neck. During medical follow-up, preventive measures should be taken to reduce the impact of the risk factors for cerebrovascular disease, to decrease stroke in these patients.
RADIOTHERAPY (RT) IS a common curative treatment modality for head and neck malignancies and may serve as adjuvant therapy in postoperative conditions for some benign tumors. Successful treatment increases survival but also puts the patient at risk for radiation-related side effects. In this respect, vascular side effects are serious and may be life-threatening. Already in 1899, Gassmann reported effects of RT on vascular tissue.1 Thereafter, several studies described radiation-induced atheromatous disease of major arteries in a variety of locations.2-4 Similar changes have also been reported involving the carotid artery after neck irradiation.5-9 Long-term vascular complications are usually contributed to accelerated atherosclerosis of the irradiated vessels, which increases the risk of vascular stenosis and thromboembolism. The main sequelae of carotid atherosclerosis include transient ischemic attacks, ischemic strokes, and amaurosis fugax.10 The reported intervals between irradiation and first symptoms range from 6 months to more than 20 years.10,11 Case reports combined with histopathologic studies suggest a possible relationship between RT and the occurrence of stroke.5-7 However, the extent of a possible increase of risk due to RT cannot be assessed from reports published so far. Moreover, most reported patients had also received surgical therapy for their head and neck tumors, which might have contributed to the carotid artery pathology.12 In only one study was the incidence of stroke in patients irradiated on the neck compared with the expected incidence in a population of similar age and sex.13 No significant increase of the risk of stroke after previous RT on the neck was found. However, inclusion of operated, not-operated, and elderly patients makes those study results hard to interpret. With this background, we investigated a consecutive series of patients protocollary treated for head and neck tumors at a relatively young age (< 60 years), to assess the relative risk (RR) of stroke after irradiation on the neck.
Patients The study group consisted of two categories of patients: those with laryngeal carcinoma primarily treated with RT, ie, T1 and T2 carcinomas, and patients with parotid gland tumors, ie, pleomorphic adenoma and parotid carcinoma. This patient selection was made to constitute a homogeneous study population with approximately the same treatment protocol and with an expected long-term survival and follow-up. Patients with T3 and T4 laryngeal carcinomas and patients with oropharyngeal carcinomas were excluded because of their limited survival time and because of operative intervention in the carotid area. Patients with laryngeal carcinoma received RT with a curative intent. Patients with pleomorphic adenomas underwent a superficial parotidectomy, and postoperative RT was indicated by tumor spill and/or incomplete microscopic clearance. Parotid carcinomas were surgically removed by a total (conservative) parotidectomy followed by RT. A retrospective cohort study was performed of patients treated between January 1977 and January 1998 at the Netherlands Cancer Institute/Antoni van Leeuwenhoekhuis. Patients 60 years old and older were excluded in order to constitute a relatively young study population with low risk of cerebrovascular disease and an expected long-term follow-up. A minimal follow-up time of 0.5 years was required in order to exclude acute complications of surgery followed by RT, which might harbor a different pathogenesis.5 Patients treated with surgery only for recurrence of a previously irradiated tumor were not excluded. During the period from January 1, 1977, to January 1, 1998, 383 patients fulfilled the inclusion criteria. Records were untraceable for 16 patients. Thus, this study was restricted to the 367 patients who received RT on the neck. From each patient chart, the following data were collected: sex, age, primary tumor, timing of RT, RT dosage and fraction size, risk factors of cerebrovascular disease, interval from RT to stroke occurrence, interval from start of RT to last visit during follow-up, and date and cause of death. The recorded risk factors included smoking, hypertension, diabetes mellitus (DM), and hypercholesterolemia. Hypertension was defined as either treatment for high blood pressure or a blood pressure that twice exceeded the limit of 95 mmHg diastolic blood pressure. Smoking was scored positive when the patient was currently smoking or had quit smoking less than 2 years before last follow-up. Each patient treated for DM was scored as positive. The cutoff value for hypercholesterolemia was 6.5 mmol/L. Criteria for cerebrovascular symptoms were a fixed neurologic deficit lasting for more than 24 hours (subarachnoidal and intracerebral hemorrhage, transient ischemic attack, and amaurosis fugax were excluded). In all cases, the clinical diagnosis was based on a neurologic examination, and in most cases, it was confirmed by a computed tomography or magnetic resonance imaging scan. In case of incomplete follow-up data in the patient record, the general practitioner was asked to provide relevant data via questionnaire that elicited risk factors and history of cerebrovascular disease.
RT The dose of RT in patients with pleomorphic adenoma was 50 Gy in fractions of 2 Gy, and the postoperative treatment for patients with parotid carcinoma consisted of 60 Gy in fractions of 2 Gy (five fractions per week). The radiation technique used was two lateral oblique-wedged fields with 4- to 8-MV photons encompassing the parotid gland area and upper neck nodal regions in patients with malignant tumors. Typically these fields included the ipsilateral carotid artery over an average length of 10 cm. The estimated dose of RT at the carotid level was 95% to 100% of the administered dose.
Statistical Analysis Expected numbers of ischemic stroke were computed by multiplying the accumulated age- and sex-specific person-years with corresponding age- and sex-specific incidence rates from the Oxfordshire Community Stroke Project.14 This British study was used because it covers much more detailed age- and sex-specific incidence rates than the smaller and less detailed Dutch Tilburg study, which is the only one available in the Netherlands.15 From a thorough inspection of both studies on differences in incidence rates, we learned that ischemic stroke rates seemed quite comparable in the two countries. Thus, with the survival experience (person-years) of the population of irradiated patients taken into account, RRs were calculated as observed-expected ratios overall and for separate categories of primary tumor, sex, follow-up interval, and several risk factors. Confidence limits of the ratios were obtained using the Poisson distribution of observed cases.16 The results were also used to calculate the absolute excess risk, by subtracting the expected number of cases from the number of observed, multiplying by 1,000, and finally dividing by person-years at risk. This risk, which estimates the excess number of strokes per 1,000 patients per year, is the most appropriate measure to judge which factors contribute most to the excess risk. Cumulative proportions of stroke incidence were estimated as a function of time since initial treatment. A life-table analysis was carried out according to the product-limit method first described by Kaplan and Meier.17
The characteristics of the 367 patients according to the type of tumor are described in Table 1. Thirty patients with laryngeal carcinoma underwent laryngectomy, including neck dissection, for recurrent disease.
Overall, 14 of 367 patients experienced a first ischemic stroke after RT. The overall risk of stroke after RT was significantly increased compared with the general population (RR, 5.6; 95% confidence interval [CI], 3.1 to 9.4). The median age at stroke was 62.7 years. The median interval from RT to occurrence of stroke was 10.9 years (range, 1.3 to 21.0 years). The absolute excess risk (AER), which estimates the excess number of strokes per 1,000 patients per year, amounted to 3.8 in the whole group. Table 2 shows the number of stroke cases, expected numbers, RR, and AER for the whole group and the separate groups. Table 3 lists the stroke characteristics and the outcomes of the patients after the occurrence of stroke. The two strokes observed in the group of patients with parotid carcinoma were both on the irradiated side. In the group of patients with pleomorphic adenoma, one out of four stroke patients experienced a stroke on the contralateral, unirradiated side. One out of eight patients with stroke in the group of laryngeal carcinoma patients had been treated 5 years after irradiation with a total laryngectomy and neck dissection for a local recurrence. The interval between the operation and the ischemic stroke was 6 years. In all 14 cases, the infarction area was in the carotid artery territory. Angiography revealed carotid artery stenosis or occlusion localized to the irradiated level, typically without signs of atherosclerosis at the nonirradiated segments (Fig 1).
In Table 4, the distribution of other risk factors for cerebrovascular disease is shown. Table 5 lists the RR and AER of stroke according to the risk profile. The RR was higher in patients with hypertension (RR, 12.5; 95% CI, 4.6 to 27.2) than in patients without hypertension (RR, 4.3; 95% CI, 1.8 to 8.4), although the difference was not statistically significant. The same tendency was seen in patients with DM. Smokers and nonsmokers had approximately similar RRs of stroke. According to Table 5, we found that patients with hypertension or DM who were irradiated on the neck had an AER of stroke of 11.6 and 17.7 per 1,000 patients per year, respectively, compared with the general population.
An analysis by duration of follow-up revealed increased RR and AER during follow-up time (Table 6). After more than 10 years’ follow-up, the RR was 10.1 (95% CI, 4.4 to 20.0) and the AER was 14.0 per 1,000 patients per year. In the group of patients younger than 50 years old during RT, the RR was 9.8 (95% CI, 3.2 to 22.9) compared with 4.5 (95% CI, 2.1 to 8.6) in the group of patients older than 50 years during RT.
Figure 2 shows the cumulative risk of stroke by follow-up time. After 15 years of follow-up, the cumulative risk amounted to 12.0% (95% CI, 6.5% to 21.4%).
Several case reports and histopathologic studies suggest a relation between RT and large-vessel disease.2,4,9,10,18,19 To the best of our knowledge, our study is the first to show a significant increased risk of ischemic stroke in an irradiated patient population with head and neck tumors. The risk of stroke after RT on the neck has already been investigated by Elerding et al,13 but they were not able to demonstrate a significant correlation between RT and stroke. In comparison with Elerding et al, we studied a more homogenous group of patients who were treated with approximately the same dose of RT and who had a relatively favorable prognosis for vascular events because of their age, ie, below 60 years at RT. We excluded patients with T3 and T4 laryngeal carcinomas, oropharyngeal carcinomas, and lymphomas because their treatment protocols consist of RT in combination with chemotherapy and/or operative intervention in the carotid area. These procedures may be associated with an increased incidence of vascular disease,12,20,21 which could make the results hard to interpret. Recently, a meta-analysis on the long-term survival of patients irradiated for early breast cancer showed a significantly increased vascular mortality after the first decade following RT.22 This study underlines the long-term effect of RT on the development of radiation-induced vascular disease. In our study we also found a significantly increased risk of ischemic stroke after more than ten years of follow-up The assumed pathogenesis of radiation-induced vascular disease is an acceleration of the atherosclerotic process probably due to endothelial cell damage, fibrosis of the intima-media layer and development of atheromatous plaques, which is increased by occlusive changes in the vasa vasorum leading to ischemia of the arterial wall. This contributes to additional damage of the carotid artery.9,18 The risk of ischemic stroke is presumably related to the increase of the intima-media thickness and narrowing of the vascular lumen. This is reflected in the high excess number of strokes of 14.0 per 1,000 patients per year after 10 years following RT on the large carotid artery compared to a 1.9 excess number of strokes when the interval from RT is shorter than 10 years. This seems also reflected in the more exponential course of the curve of the actuarial risk of stroke from 10 years following RT (Fig 2). Besides, an increased incidence of stroke in the control group of the same age probably contributes to the course of the curve at advanced age. Five out of the six strokes occurring at the ipsilateral side of unilaterally irradiated parotid tumors corroborates the hypothesis that radiation-induced atherosclerosis of the carotid artery is the pathogenic factor. Age at RT may be of importance. Hancock et al23 observed a highly significant increased risk of cardiovascular complications after mediastinal RT for Hodgkin’s disease in patients treated before they were 20 years old compared with those treated when they were older than 50 years. In the present study, the RR of stroke was 9.8 in patients treated before 50 years old compared with 4.5 in older patients; however, the difference was not significant. The applied radiation dose in the present study varied between 50 Gy and 66 Gy. The minimal dose for RT damage of the cervical arteries was assumed to be 40 Gy.24 Chung et al,11 evaluating the degree of carotid stenosis on postradiation MR scans, found no dose-effect relationship when RT dose was increased to 60 Gy. The risk factor hypertension increased the risk (RR, 12.5) of stroke after RT. No difference in the RR of stroke was observed between smokers and nonsmokers. However, this finding should be interpreted with caution, because registration of smoking habits in a retrospective analysis is probably inaccurate. The statement that stroke in patients with head and neck tumors should be ascribed to their lifestyle, including excessive smoking, does not hold true for patients with a parotid tumor. Moreover, a significant proportion of the patients with laryngeal carcinoma stopped smoking at the diagnosis of cancer. In view of the median interval of 7 years between RT and occurrence of stroke in these patients, and the observed similar RR of stroke in patients with laryngeal carcinoma and parotid tumors, smoking habits in our population of patients presumably did not play a significant role in the observed increased risk of stroke compared with the matched normal population. Hypercholesterolemia may also play a role in the development of stenosis after RT. Silverberg et al25 found hypercholesterolemia to be a contributory risk factor for radiation-accelerated atherosclerosis, as was also demonstrated in animal studies.26,27 In our population, only a few patients (22.6%) were screened for cholesterol levels, which made data inconclusive. As treatment for malignancies becomes more effective, more patients will survive and long-term cerebrovascular disease will become more prominent. This highlights the need for long-term follow-up after curative treatment for head and neck cancer. During follow-up, more attention should be paid to reduction of risk factors for atherosclerosis in patients who received RT on the neck. Routine performance of carotid noninvasive studies, such as ultrasound, may be worthwhile. In this respect, in case of amaurosis fugax, transient ischemic attack, or stroke, the patient needs a neurologic evaluation to investigate possible therapeutic interventions. RT-induced carotid artery lesions, as seen on angiograms, can easily be differentiated from normal atherosclerotic lesions. The lesions seem to have a disproportionate involvement of the distal carotid artery, according to the irradiated area, are usually long, and are not confined to the bifurcation.28 In addition, they are often associated with occlusion of other cervical arteries within the radiated portal.29 Therapeutic options consist of percutaneous transluminal angioplasty, carotid endarteriectomy, and bypass surgery. However, these techniques remain difficult because of arterial, periarterial, and cutaneous sclerosis. Complications are diverse and included scar disruption, prosthetic infection, anastomotic breakdown, wound-healing problems, and restenosis.29 Reports about carotid artery repair after neck radiation include fewer than 100 patients, and most have been isolated cases. Recently, Kashyap et al30 presented the results of two decades of follow-up of carotid artery repair in 24 patients. They concluded that carotid surgery after neck radiation was a safe and durable treatment. Furthermore, they found this intervention to guarantee the protection against neurologic events similar to the results obtained in endarterectomy in the absence of RT.30 Although surgery is an option for intervention, we would rather emphasize prevention of cerebrovascular disease after RT, which can be achieved in part by reduction of risk factors. The benefit of prophylactic antiplatelet therapy and the use of statins or angiotensin-converting enzyme inhibitors remain uncertain in this matter. Indications for postoperative RT in case of incomplete surgical removal of pleomorphic adenoma and/or tumor spill should be re-evaluated and weighed against the increased risk of neurologic complications.31
1. Gassmann A: Zur histologie der roentgenulcera. Fortschr Geb Roentgenstr 2: 199-207, 1899 2. Butler MJ, Lane RH, Webster JH: Irradiation injury to large arteries. Br J Surg 67: 341-343, 1980[Medline] 3. Taylor PJ, Cooper GG, Sarkar TK: Upper-limb arterial disease in women treated for breast cancer. Br J Surg 82: 1089-1091, 1995[Medline] 4. Pettersson F, Swedenborg J: Atherosclerotic occlusive disease after radiation for pelvic malignancies. Acta Chir Scand 156: 367-371, 1990[Medline] 5. McCready RA, Hyde GL, Bivins BA, et al: Radiation-induced arterial injuries. Surgery 93: 306-312, 1983[Medline] 6. Conomy JP, Kellermeyer RW: Delayed cerebrovascular consequences of therapeutic radiation: A clinicopathologic study of a stroke associated with radiation-related carotid arteriopathy. Cancer 36: 1702-1708, 1975[CrossRef][Medline]
7.
Levinson SA, Close MB, Ehrenfeld WK, et al: Carotid artery occlusive disease following external cervical irradiation. Arch Surg 107: 395-397, 1973 8. Osgood CP, Karunaratne ER, Sandler R, et al: Carotid occlusion in a 27-year-old man: A case report. Neurosurgery 18: 459-460, 1986[Medline] 9. Atkinson JL, Sundt TMJr, Dale AJ, et al: Radiation-associated atheromatous disease of the cervical carotid artery: Report of seven cases and review of the literature. Neurosurgery 24: 171-178, 1989[Medline]
10.
Murros KE, Toole JF: The effect of radiation on carotid arteries: A review article. Arch Neurol 46: 449-455, 1989 11. Chung TS, Yousem DM, Lexa FJ, et al: MRI of carotid angiopathy after therapeutic radiation. J Comput Assist Tomogr 18: 533-538, 1994[Medline] 12. Nosan DK, Gomez CR, Maves MD: Perioperative stroke in patients undergoing head and neck surgery. Ann Otol Rhinol Laryngol 102: 717-723, 1993[Medline] 13. Elerding SC, Fernandez RN, Grotta JC, et al: Carotid artery disease following external cervical irradiation. Ann Surg 194: 609-615, 1981[Medline]
14.
Bamford J, Sandercock P, Dennis M, et al: A prospective study of acute cerebrovascular disease in the community: The Oxfordshire Community Stroke Project–1981-86—2. Incidence, case fatality rates and overall outcome at one year of cerebral infarction, primary intracerebral and subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 53: 16-22, 1990
15.
Herman B, Leyten AC, van Luijk JH, et al: Epidemiology of stroke in Tilburg, the Netherlands: The population-based stroke incidence register—2. Incidence, initial clinical picture and medical care, and three-week case fatality. Stroke 13: 629-634, 1982 16. Pearson ES, Hartley HO: Biometrika Tables for Statisticians. London, United Kingdom, Biometrica Trust, 1976 17. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958[CrossRef] 18. Fonkalsrud EW, Sanchez M, Zerubavel R, et al: Serial changes in arterial structure following radiation therapy. Surg Gynecol Obstet 145: 395-400, 1977[Medline]
19.
Moritz MW, Higgins RF, Jacobs JR: Duplex imaging and incidence of carotid radiation injury after high-dose radiotherapy for tumors of the head and neck. Arch Surg 125: 1181-1183, 1990
20.
Doll DC, Ringenberg QS, Yarbro JW: Vascular toxicity associated with antineoplastic agents. J Clin Oncol 4: 1405-1417, 1986
21.
Meinardi MT, Gietema JA, van der Graaf WT, et al: Cardiovascular morbidity in long-term survivors of metastatic testicular cancer. J Clin Oncol 18: 1725-1732, 2000 22. Favourable and unfavourable effects on long-term survival of radiotherapy for early breast cancer: An overview of the randomised trials—Early Breast Cancer Trialists’ Collaborative Group. Lancet 355:1757-1770, 2000
23.
Hancock SL, Tucker MA, Hoppe RT: Factors affecting late mortality from heart disease after treatment of Hodgkin’s disease. JAMA 270: 1949-1955, 1993 24. Graus F, Rogers LR, Posner JB: Cerebrovascular complications in patients with cancer. Medicine (Baltimore) 64: 16-35, 1985[Medline] 25. Silverberg GD, Britt RH, Goffinet DR: Radiation-induced carotid artery disease. Cancer 41: 130-137, 1978[CrossRef][Medline] 26. Gold H: Production of arteriosclerosis in the rat: Effect of x-ray and high-fat diet. Arch Pathol 71: 268-273, 1961[Medline] 27. Lamberts H, de Boer W: Contributions to the study of immediate and early x-ray reactions with regard to chemoprotection: VII. X-ray-induced atheromatous lesions in the arterial wall of upper cholesterolemic rabbits. Int J Radiat Biol 6: 343-350, 1963
28.
Loftus CM, Biller J, Hart MN, et al: Management of radiation-induced accelerated carotid atherosclerosis. Arch Neurol 44: 711-714, 1987 29. Melliere D, Becquemin JP, Berrahal D, et al: Management of radiation-induced occlusive arterial disease: A reassessment. J Cardiovasc Surg (Torino) 38: 261-269, 1997[Medline] 30. Kashyap VS, Moore WS, Quinones-Baldrich WJ: Carotid artery repair for radiation-associated atherosclerosis is a safe and durable procedure. J Vasc Surg 29: 90-96, 1999[CrossRef][Medline] 31. Leverstein H, van der Wal JE, Tiwari RM, et al: Surgical management of 246 previously untreated pleomorphic adenomas of the parotid gland. Br J Surg 84: 399-403, 1997[CrossRef][Medline] Submitted January 22, 2001; accepted August 24, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
