Elective popliteal aneurysms: does venous availability has an impact on indications?

AIM: The aim of this retrospective study was to evaluate the patency and limb salvage rates after prosthetics or venous bypasses for asymptomatic or mildly symptomatic popliteal aneurysms, in order to determine if small uncomplicated aneurysms (caliber <300%) should be operated or periodically controlled when a venous conduit is not available. METHODS: During a 18 years period, 100 popliteal aneurysms, including 85 asymptomatic and 15 associated with intermittent claudication, were operated on: group I consisted of 80 venous bypasses, and group II consisted of 20 prosthetic bypasses. RESULTS: Demographics and risks factors were similar in both groups. Local complications were more frequent in group I (17% vs 10% p=NS). Early vascular complications were rare in both groups. Late arterial complications were more frequent in group II. Primary patency, assisted primary patency, and secondary patency rates at 2 years were 94.3%, 97.3% and 98.7% in group I, and 61.5%, 89% and 88.4% in group II (p<0.05). In contrast, the limbs salvage rates were not significantly different for each type of graft (98.7% vs 100% p=NS). CONCLUSIONS: Our data shows that aneurysms treated with a prosthetic graft are at higher risk of late occlusion than those operated with a vein graft. This should be taken into account when facing a small uncomplicated aneurysm without available venous conduit. The presence of a suitable vein should be checked before deciding to operate a small uncomplicated popliteal aneurysm.

Subintimal angioplasty: feasible and durable.

OBJECTIVE: To assess prospectively the feasibility and durability of subintimal angioplasty (SA) clinically and by duplex scans every 3 months. PATIENTS AND METHODS: Within a period of 54 months, we selected 96 patients with 100 occlusions (mean length: 11.5 cm) of femoro-popliteal or tibial arteries, for SA. RESULTS: The technical success rate was 88% and seven out of 12 failures were treated by conventional surgery. Five below-the-knee amputations were performed despite a patent recanalization. The following complications occurred: arterial perforation (6), arterial thrombosis (4), extensions beyond the planned re-entry site (5), and arterial dissection (2). Primary, assisted-primary and secondary patency rates were 61, 68 and 74%, respectively at 24 months. The 24 month-limb salvage and survival rates were 78 and 85%, respectively. Duplex imaging demonstrated 10 restenosis (five symptomatic >70%, five asymptomatic 30-70%), seven occlusions (five asymptomatic, two symptomatic treated by a bypass) and one asymptomatic dilatation. CONCLUSION: In a selected group of patients SA is feasible with a high initial technical success rate. SA is a good alternative in patients who are poor candidates for bypass surgery.

Robotically assisted aorto-femoral bypass grafting: lessons learned from our initial experience.

OBJECTIVE: The da Vinci trade mark Surgical System (Intuitive Surgical Inc., Sunnyvale, CA) is a computer-enhanced telemanipulator that may help to overcome some limitations of traditional laparoscopic instruments. This prospective study was performed to assess the safety and feasibility of robotically assisted aorto-femoral bypass grafting (AF). METHODS: Five patients undergoing elective AF were enrolled in this study. In three patients, a laparotomy of 6 cm was first performed, the aorta being exposed using an Omnitract degrees retractor. In two patients, aortic dissection was performed with laparoscopy, with the patient in a modified right lateral decubitus position. In all patients, the proximal anastomosis was attempted with the da Vinci trade mark system by a remote surgeon. The role of the assistant at the patient's side was limited to exposure, haemostasis and maintaining traction on the running sutures performed by the robot. Six weeks after the operation, all patients underwent a duplex scan of the graft. RESULTS: Mean operative time was 188 min. Robotically assisted aortic anastomoses were successfully completed in four out of five patients. In these four patients, adequate blood flow was observed within the graft with no need for conversion for haemostasis. In the fifth patient, despite an adequate laparoscopic aortic dissection, the anastomosis was impossible to perform due to external conflicts between the robotic arms. A conversion using conventional suture was successfully performed. No robot-related complications were noted. Six weeks after the operation, the duplex scans demonstrated a graft patency of 100%. CONCLUSION: Robotically assisted anastomoses are possible by their unique ability to combine conventional laparoscopic surgery with stereoscopic 3D magnification and ultra-precise suturing techniques due to the flexibility of the robotic-wristed instruments using different motion scaling of surgeon hand movements. In addition, prior training in laparoscopic aortic surgery is not necessary for surgeons to obtain the level required for suturing. Further clinical trials are needed to explore the clinical potential and value of robotically assisted AF.

[Feasibility of ambulatory endovenous laser for the treatment of greater saphenous varicose veins: one-month outcome in a series of 20 outpatients]. / Peut-on traiter les grandes saphènes variqueuses par laser endoveineux en ambulatoire? Résultats á l mois d'une étude de faisabilité sur 20 patients traités en salle de consultation.

The purpose of this feasibility study was to demonstrate that endovenous laser can be a useful alternative to conventional surgery for ambulatory treatment of advanced varicose veins. We assessed an open, non-randomized series of patients treated in one center by the same operator. The study protocol was approved by the local ethics committee. Twenty patients with stage II or III varicose veins in the Porter classification gave their informed consent to participate in the study. The patients were treated with endovenous laser by the first author in the outpatient clinic of the Henri Mondor University Hospital vascular surgery department. All procedures were conducted under local anesthesia. A 980 nm laser diode optic fiber was introduced into the vein percutaneously. Laser beams were fired from the sapheno-femoral junction to just under the genu, withdrawing the fiber 3 mm every 1.5 sec. Clinical evaluation with a quality-of-life questionnaire and duplex-scan was performed at days 3, 8 and 30 post-op. Complete occlusion and retraction of the treated vein was observed at day 3 and 30, from the point of introduction to the sapheno-femoral junction in 18 of the 20 patients. The branches of the greater saphenous vein remained patent with physiological flow in the stump which remained patent 1 to 2 cm upstream from the sapheno-femoral junction. The length of the patent stump dependend on the level of the anterior or posterior branch. There were no adverse effects related to the local anesthesia. Pain was low to mild during treatment and the days following the procedure, requiring 8 tablets of acetominophen at most. Hematomas were minimal and had completely resolved by the end of the first month. No work stoppage was required for the 14 patients with occupational activities. There were no cases of deep or superficial vein thrombosis. Complete occlusion and retraction of the varicose vein at one month suggests this treatment has a long-lasting effect. Long-term evaluation is required. Treatment of advanced varicose veins with endovenous laser can be an alternative to surgical treatment providing the advantage of outpatient ambulatory treatment.

[Acute pancreatitis after surgical treatment of abdominal aortic aneurysm]. / Pancréatite aiguë après traitement chirurgical des anévrismes de l'aorte abdominale.

INTRODUCTION: Acute pancreatitis after surgical treatment of non ruptured aneurysm of abdominal aorta is a rare complication, considered to be due to pancreatic ischemia or peroperative trauma of pancreas. The aim of this study is to describe 4 new cases of this complication and to discuss its etiology. PATIENTS AND METHODS: From January 1995 to November 2000, 365 patients underwent elective surgery for a non ruptured abdominal aorta aneurysm. Four (1.1%) men, aged 66 to 79 years and operated for an aneurysm which diameter ranged from 60 to 77 mm, developed postoperative acute pancreatitis. The abdominal approach was a midline incision in 3 cases and a retroperitoneal lombotomy in one case. Superior pole of the aneurysm always adjoined or involved the right renal artery. The aortic clamping was supra-renal in 3 cases and celiac in one case. Diagnosis of acute pancreatitis was established at days 2, 4, 12, and 23 after surgery on abdominal computed tomography in 3 cases and at reoperation in one case. RESULTS: Three patients died, including 2 from early multiple organ failure and one peroperatively during surgical attempt to treat a prostheto-digestive fistula. One patient was alive and asymptomatic with a 2-years follow-up. CONCLUSION: Acute pancreatitis is a rare and serious complication after surgical treatment of abdominal aorta aneurysm. Its diagnosis is often delayed. The main etiological factor of this complication could be trauma of pancreas during supra-renal clamping through a midline incision.

Late reoperations after surgical and endovascular lower limb revascularisation.

OBJECTIVES: to evaluate the rate of late reoperations after primary lower limb revascularisation. METHODS: a prospectively collected database of 639 consecutive patients, undergoing 763 primary arterial reconstruction for lower limb occlusive disease during the five-year period 1989 through 1993, was analysed with regard to the rates and locations of late (after 30 days) re-operation. The reconstruction was supra-inguinal in 527 (69%) cases, surgical in 391 (Group IA) and endovascular in 136 (Group IB). Infra-inguinal reconstruction had been performed in 236 (31%) cases, surgical in 160 (Group IIA) and endovascular in 76 (Group IIB). The 499 men and 140 women were followed during an average of 32 (range 1-122) months. RESULTS: the 5-year cumulative reoperation rates were 25% after supra-inguinal surgery and 28% after supra-inguinal angioplasty, respectively (NS). The rates were 47% and 33% after infra-inguinal surgery and angioplasty, respectively ( p=0.04). CONCLUSION: limbs treated for occlusive arterial disease carry a high risk of late reoperation whether the initial procedure is open surgery or angioplasty. Reoperations were twice as common for infra-inguinal procedures.

Percutaneous transluminal angioplasty in patients with ischemic tissue necrosis is worthwhile.

PURPOSE: Ischemic tissue necrosis is usually associated with long or sequential arterial obstructions. As a result, the role of percutaneous transluminal angioplasty (PTA), which addresses only short lesions, in patients presenting with trophic changes remains questionable. The purpose of this study was to evaluate the effectiveness of PTA in diabetic and non-diabetic patients presenting with grade 4 Fontaine's classification. METHOD: Between January 1992 and December 1997, 1352 patients with aorto-iliac and/or infrainguinal occlusive diseases were admitted to our institution. Three hundred and ten patients who presented with distal gangrene (95.5%) or ischemic ulcers (4.5%) were identified. The patients consisted of 117 diabetics and 193 non-diabetics. PTA alone was performed in 26 diabetics (group 1) and in 30 non-diabetics (group 2). Their charts were retrospectively reviewed and the patients were recalled for clinical examination and non-invasive monitoring. RESULTS: Follow-up was available for all patients and ranged from 1 to 4 years. The survival rate was significantly higher in diabetic patients than in non-diabetic patients (96% vs 77% at 1 year; p<0.05 and 91% vs 66% at 3 years; p<0.05). In group 1, the primary cumulative patency rate at 1 and 3 years was 76%. In group 2, the primary cumulative patency rate at 1 and 3 years were 85% and 80%, respectively. Three patients in group 1 required a redo PTA at 4 months, resulting in an assisted primary patency rate at 1 and 3 years of 88%. In contrast, no patients in group 2 required additional PTA. In group 1, the limb salvage rate at 1 and 3 years was 84%; and in group 2, 80% and 75%, respectively. CONCLUSION: The results of PTA in both groups were encouraging. Dilation of one or two short stenoses, despite multiple distal lesions, may improve distal flow sufficiently to promote wound healing. Thus, this procedure may be recommended in selected patients suffering from ischemic tissue loss. However, during the first 6 months following PTA, diabetic patients should be followed carefully with Duplex as they are prone to early restenosis.

Injury in vascular surgery--the intimal hyperplastic response.

Intimal hyperplasia is extensively studied in order to improve arterial reconstruction outcome. The mechanisms leading to stenosis or restenosis may vary according to the technique used for arterial reconstruction. Lesions are mostly made of an accumulation of smooth muscle cells and fibroblasts, with only sparse inflammatory cells. The accumulated material reduces the graft lumen and ultimately induces thrombosis. Intimal hyperplasia with smooth muscle cell and matrix accumulation is the prominent feature in all these situations with evidences of intense cell proliferation and cell death. The purpose of this review is to present the biology of intimal hyperplastic response based on the recently published data. Experiments in the rabbits have shown that the vein wall thickening is mainly regulated by the tangential wall stress which is applied transversely to the vein wall as a blood pressure. Experiments in the rat carotid artery balloon injury suggested that heparin could be used as a treatment to prevent intimal hyperplasia. Treatments for preventing restenosis after angioplasty or stenoses development in bypasses have been disappointing clinical evaluation suffers from insufficient prospective randomized studies. Intimal hyperplasia is the major cause of failure after arterial reconstruction. The biology of intimal hyperplasia is complex, and treatment disappointing. Some types of hyperplasia may need to be preserved in order to prevent functional atrophy and aneurysmal dilatation of vein grafts.

Unilateral surgery for primary hyperparathyroidism on the basis of technetium Tc 99m sestamibi and iodine 123 subtraction scanning.

HYPOTHESIS: Parathyroid scanning, based on simultaneous recording of technetium Tc 99m sestamibi and iodine 123 images, is able to identify patients with multiple parathyroid gland disease and is a safe imaging technique for unilateral parathyroid surgery. DESIGN: Scintigraphic criteria of eligibility for unilateral surgery were prospectively tested against findings of conventional bilateral surgery. SETTING: Patients referred to an endocrine surgeon in a university hospital. PATIENTS: Seventy consecutive patients with primary hyperparathyroidism had dual-isotope scanning before conventional surgery. Forty-one patients had scan findings compatible with unilateral surgery, with a single focus of high intensity seen on the anterior and lateral views. The remaining 29 patients had 1 or more criteria of ineligibility: (1) scan findings pointing to multiple gland disease, (2) no well-identified focus, (3) contralateral thyroid nodule requiring surgical management, or (4) family history of hyperparathyroidism or multiple endocrine disease. MAIN OUTCOME MEASURES: Number of enlarged parathyroid glands at surgical inspection and calcemia follow-up. RESULTS: None of the 41 patients, with a single well-defined focus on the scan image, showed evidence of multiple parathyroid involvement. Each parathyroid adenoma was resected from the precise site predicted by the subtraction scan. Nine patients (13%) had surgical findings of multiple parathyroid gland disease. All 9 were ineligible based on preoperative image findings. CONCLUSIONS: Unilateral surgery can be safely offered to 60% of patients with primary hyperparathyroidism, on the basis of simultaneous (99m)Tc-sestamibi and (123)I scanning. This may reduce the length of the operation, anesthesia requirements, and hospital stay, and the risks of hypoparathyroidism and injury to the recurrent laryngeal nerve.

[Surgery of the internal carotid: locoregional or general anesthesia?]. / Chirurgie de la carotide interne: anesthésie locorégionale ou générale?

AIM OF THE STUDY: The aim of this retrospective study was to assess the advantages of regional anesthesia over general anesthesia in carotid artery surgery. PATIENTS AND METHOD: From January 1989 to December 1998, 670 patients with severe internal carotid artery stenosis were operated in the same center and were classified into two groups according to the type of anesthesia: group I, general anesthesia (n = 312) and group II, regional anesthesia (n = 358). Characteristics of the two groups were almost similar except for a higher rate of unstable heart disease in group I and bypass grafts in group II. RESULTS: A shunt was used in 16.3% of cases in group I and in 8.4% in group II. Complications resulting from the use of a shunt and intraoperative complications observed with regional anesthesia were reported. There was a conversion from regional to general anesthesia in 6 patients. Median duration of clamping was longer in group II (30 min vs 25 min). Cardiac complication rates were similar in the two groups, particularly cardiac mortality (0.6%). There were more pulmonary and miscellaneous complications in group I. Neurological complications were more frequent in group I, particularly fatal strokes (1% versus 0%). Neurological mortality and morbidity cumulative rates were 3.1% and 1.5%, respectively, not significantly different. CONCLUSION: These results, in agreement with those of the literature, confirm that carotid artery endarterectomy is associated with a low rate of neurological mortality and morbidity. Although regional anesthesia was associated with a lower rate of complications, we are not allowed to conclude to its superiority, as the present study was retrospective and the difference was not statistically significant.

Acute occlusion of popliteal and/or tibial arteries: the value of percutaneous treatment.

OBJECTIVES: to describe early and mid-term results with a percutaneous therapeutic protocol including thromboaspiration, thrombolysis, and correction of the underlying lesion by PTA. METHODS: thirty-three consecutive selected patients with recent (<1 month) reversible acute ischaemia associated with popliteal and/or tibial occlusion were studied. The primary endpoints were technical success (defined as residual mural thrombus less than 20% of the lumen and the presence of at least one tibial artery on angiogram), patient survival and limb salvage at 1 and 12 months. Secondary endpoints included complications, primary, assisted primary and secondary patency determined by duplex scan at 1, 6 and 12 months. RESULTS: technical success was achieved in 27 patients (82%). Twenty patients were treated by thrombo- aspiration+/-thrombolysis only, and seven required additional PTA (26%). In six patients (18%), percutaneous techniques failed, and embolectomy was performed in two, bypass in one and major amputations in three (9%). For the entire series, the survival rate was 100% at 1 month and 94% at 1 year. The limb salvage rate was 91% at 1 month and 1 year. The cumulative primary patency, assisted primary patency and secondary patency rates were 81%, 81% and 86% respectively at 1 month and 66%, 72% and 77%, respectively, at 12 months. Early complications occurred in 10 patients (30%): five groin haematomas (15%), four compartment syndromes (12%) and one haemoglobinuria (3%). CONCLUSION: percutaneous techniques offer excellent early and mid-term results in selected patients presenting with acute ischaemia with popliteal and/or tibial arteries occlusion.

The underestimated advantages of iliofemoral endarterectomy.

Iliofemoral endarterectomy was invented 50 years ago, but it is seldom practiced today for two reasons. The first is that it is technically challenging and the second is that outcome in early series was poor. Our preliminary experience having been more encouraging, we have continued to perform iliofemoral endarterectomy for the past 20 years. The purpose of this retrospective study was to evaluate our results and compare them with results of alternative techniques described in recent literature. We have performed a total of 176 iliofemoral endarterectomies in patients with normal or nearly normal aortas. The procedure involved the entire network including the common iliac artery, external iliac artery, and common femoral artery in 108 cases (group I), the common iliac artery with or without the external iliac artery in 40 cases (group II), and the external iliac arteries and the common femoral artery with or without the deep femoral artery in 28 cases (group III). From our results we conclude that iliofemoral endarterectomy should be used as a first-choice modality in patients with normal or nearly normal aortas who present with iliac lesions that are either too long for balloon angioplasty or impossible to recanalize. It eliminates the risk of graft infection and false aneurysm. Restenosis can be treated by balloon angioplasty. It also saves the cost of a prosthesis.

Peroperative factors influencing the outcome of elective abdominal aorta aneurysm repair.

PURPOSE: To identify perioperative variables which may influence mortality of elective abdominal aneurysm repair (AAA). METHOD: prospective study of patients undergoing elective AAA repair between 1986 and 1997. RESULTS: Four hundred and seventy patients (438 men, 32 females) with a mean age of 69.4+/-13 years and aneurysms with a diameter of 60+/-3 mm were operated on with a 1-month mortality rate of 5.3%. Multivariate analysis identified the following independent risk factors for mortality: age >70 (p<0.0001), a past history of myocardial infarction (p<0.0001), preoperative renal insufficiency (p<0.0001), reoperation (p<0.0001), colonic necrosis (p<0.0001), and severe postoperative medical complications (p<0.0001). CONCLUSION: Intra- and postoperative events affect the outcome of AAA repair, independently of preoperative factors, and should be described when presenting the results of AAA repair.

Mid-term results of endovascular versus open repair for abdominal aortic aneurysm in patients anatomically suitable for endovascular repair.

OBJECTIVES: to prospectively evaluate the mid-term results of endovascular and open repair in patients with abdominal aortic aneurysm (AAA) anatomically suitable for endovascular repair. MATERIAL AND METHODS: between January 1995 and March 1999, among 438 patients treated for AAA, 180 (41%) were suitable for endovascular repair as assessed by computed tomography (CT) scan and angiogram. Seventy-three were treated by various commercially available endovascular grafts (EV) and 107 by open repair (OR). Postoperatively, patients were followed every 6 months with clinical examination, duplex scan and in the EV group, CT scans. Patients>> demographic data, intra- and postoperative events were recorded prospectively in a computerised database and compared for each group. RESULTS: median age, sex ratio, preoperative risk factors and aneurysm diameters were not statistically different between the two groups. Respectively in the EV and OR, the average duration of operation was 149+/-73 mn, and 133+/-44 mn (NS), blood loss 96 ml+/-28 and 985 ml+/-113 (p<0.01), duration of hospitalisation 7 days+/-2 and 13 days+/-7 (p<0.01). The one-month mortality was 2.7% (n=2) for EV and 2.8% (n=3) for OR. The rate of cardiac and pulmonary complications was significantly higher in the OR group (6. 9% versus 19.6%, p=0.017). At a mean follow-up of 1 year, the cumulative survival rate was 82.2%+/-7.5 for EV and 96%+/-2.12 for OR (log-rank test p=0.043). No patients died of rupture, but three patients had to be converted to open surgery. Twenty-two percent (n=16) patients in the EV and 7.5% (n=8) in the OR were submitted to a subsequent minor or major reintervention (p=0.007). At 1 year, the cumulative rates free of any reintervention were respectively 78. 8%+/-6.7% and 92.9%+/-2.7% (p=0.001). In the EV there were 17 early endoleaks (23.3%). At the end of patient's follow-up seven endoleaks (9.6%) persisted. The primary success rate defined by the absence of endoleak and the absence of reintervention was 54 (74%) with EV and 101 (94%) with OR (p=0.001). CONCLUSION: EV is a promising technique. However, with current devices and indications the immediate benefits, mainly less blood loss, fewer cardiac and pulmonary complications, and shorter hospitalisation time, are outweighed by a higher rate of reinterventions to treat endoleak, or to maintain patency of the graft.

Placement of a fenestrated Palmaz stent across the renal arteries. Feasibility and outcome in an animal study.

OBJECTIVES: To investigate the feasibility of placing stents across renal arteries. Design we have studied in pigs: (i) the feasibility of accurately placing a fenestrated stent in front of one renal ostium; (ii) the short-term effects on renal arteries and function after the placement of such a fenestrated stent. MATERIALS AND METHODS: Eight fenestrated Palmaz stents were placed over pigs' renal ostia under fluoroscopy. Five weeks later, angiograms were performed and the animals were sacrificed. Proliferation of the healing tissues over the ostia was measured and analysed by microscopy. Serum creatinine was measured prior to all angiograms and at 5 weeks. RESULTS: All eight stents were correctly placed. One stent later migrated and was excluded from the study. One pig died at day 1. Gross examination confirmed the correct placement of the fenestrations in four pigs out of seven (57%). In the six remaining pigs, at 5 weeks, there was no angiographic evidence of stent misplacement and all the kidneys were fully perfused. Nine renal ostia were covered by struts and neointima with a mean area of coverage of 38+/-5% altogether. No tissue proliferation was observed over the three renal ostia located in front of the fenestration. Serum creatinine did not significantly increase at 5 weeks. CONCLUSION: Creating a fenestration in a stent for renal arteries may be worth while in order to avoid neointimal covering of the renal ostia. However, accurate placement of such a fenestrated stent remains a difficult task.

[Radiation-induced aorto-ilio-femoral arterial arteritis. Mediocrity of the long-term results after conventional surgery]. / Artérites radiques aorto-ilio-fémorales. Médiocrité des résultats à long terme après chirurgie conventionnelle.

OBJECTIVES: To assess long-term outcome after surgical cure of radiation-induced aorto-ilio-femoral arterial occlusion and to deduce therapeutic indications. PATIENTS AND METHODS: Over a 20-year period, 15 patients were treated for occlusion of the aorto-ilio-femoral vessels 4 to 28 years after receiving radiotherapy. Primary treatment was medical (n=1), balloon dilatation (n=5), anatomic revascularization (n=5), and extra-anatomic bypass (n=4). Mean follow-up was 6.8 years. RESULTS: Among the 5 balloon dilatations, there was one failure requiring right axillofemoral bypass that was followed shortly by fatal septic shock; the 4 others have remained patent at 2 to 15 years, one having required stent dilatation at 6 months. Among the 5 patients who had anatomic revascularization, excepting on patient who died shortly after surgery from her cancer, all have required revision; 2 patients died of infection at 9 and 10 years; the two others are alive at 5 and 18 years. Among the 4 patients with extra-anatomic bypass (one aortothoracic-bifemoral bypass, two axillofemoral bypasses and one cross-over bypass), 2 died with prosthetic sepsis at 3 and 7 years and 2 others have had a bypass replacement. Overall, at last follow-up, half of the operated patients have died from infection (n=5) or cancer (n=2). DISCUSSION: In reports in the literature, revision and infection after conventional surgery is frequent. There is less risk with the endovascular approach, but it can be only applied for short occlusions. CONCLUSION: Excepting easily accessible occlusions with an apparent minimal risk for percutaneous balloon dilatation, irradiated arteries should be operated on only in case of severe ischemia. Patients with claudication should be treated conservatively because of the high risk of complications for prosthesis implantation with irradiated arteries.

[Surveillance after treatment of differentiated thyroid cancers]. / Surveillance après traitement des cancers thyroïdiens différenciés.

UNLABELLED: The prognosis of differentiated thyroid carcinoma mainly depends on the quality of the initial treatment and on early detection and management of any recurrences. STUDY AIM: The aim of this retrospective study was to assess the modalities and duration of surveillance in these patients according to an original classification based on the initial extent of the tumour. PATIENTS AND METHOD: The modalities of detection of local recurrences and metastases and the date of the latest recurrences were assessed in a series of 509 patients with a mean follow-up of 8.2 years (range: 1 to 25 years). Most patients were treated by total thyroidectomy, followed by a therapeutic dose of radioactive iodine. The other patients with a small localized carcinoma underwent partial thyroidectomy without radioactive iodine. Patients were divided into four groups on the 7th postoperative month after follow-up scintigraphy and thyroglobulin assay: group I: microcancers (n = 117), group II: cancer without lymph node involvement or metastasis and normal thyroglobulin divided into IIA, age < 45 years (n = 100) and IIB, age > 45 (n = 94), group III: cancer with lymph node involvement and normal thyroglobulin (n = 102), group IV: high-risk cancers with metastases or regional extension other than lymph node extension or thyroglobulin > 3 micrograms/L (n = 96). RESULTS: Cancer-dependent actuarial survival rates for groups I, IIA, IIB, III, IV were 100%, 100%, 96%, 100%, and 73% at 10 years and 100%, 100%, 92%, 100%, and 86% at 15 years, respectively. Local or metastatic recurrences were sometimes detected by a single follow-up examination, while the other examinations were negative: cervical palpation, thyroglobulin assay, iodine scintigraphy, chest X-rays. The latest recurrences were observed at 12 years in groups I and IIA and at 16 years in groups IIB, III, and IV with normal thyroglobulin. CONCLUSION: This study confirms the importance of weaning thyroglobulin assays and scintigraphy which must be repeated every 5 years. Cervical palpation, thyroglobulin assay without weaning, chest X-rays may also detect recurrences. Duration of follow-up must be adapted to the initial extension and subsequent course: 15 years in groups I and IIA, 20 years in groups IIB, III, and IV with normal thyroglobulin, for at least 10 years after each recurrence, and life-long in the case of progression and thyroglobulin > 3 micrograms/L. Patients must be informed about the duration of follow-up at the 7th month when the definitive classification can be established and continuity of this follow-up must be documented in a special register.