Positron emission tomography as a tool for the tailored management of retroperitoneal fibrosis: a nephro-urological experience

Transcription

1 Nephrol Dial Transplant (2010) 25: doi: /ndt/gfq051 Advance Access publication 15 February 2010 Positron emission tomography as a tool for the tailored management of retroperitoneal fibrosis: a nephro-urological experience Giorgina Barbara Piccoli 1, Valentina Consiglio 1, Vincenzo Arena 3, Ettore Pelosi 3, Douroukas Anastasios 3, Francesca Ragni 1, Cristian Fiori 1, Gianfranco Cortese 2, Maria Chiara Deagostini 1, Francesco Porpiglia 1 and Roberto Mario Scarpa 1 1 Urology and Nephrology, San Luigi Gonzaga University Hospital, Regione Gonzole 10, Orbassano Torino, Turin, Italy, 2 Radiology Ospedale Maria Vittoria, via Cibrario 72, Turin, Italy and 3 IRMET S.p.A., via Onorato Vigliani 89/A, Turin, Torino, Italy Correspondence and offprint requests to: Giorgina Barbara Piccoli; gbpiccoli@yahoo.it; gbpiccoli@hotmail.com; giorgina.piccoli@unito.it Abstract Background. Retroperitoneal fibrosis (RF) is a complex clinical entity characterized by a fibro-inflammatory reaction around the abdominal aorta and iliac arteries extended into the retroperitoneum. No biochemical marker correlates with the disease severity and progression, and imaging data fail to discriminate between fibrotic and florid lesions. Positron emission tomography (PET) was recently suggested as a promising tool to detect the disease. Methods. We report on seven consecutive cases of RF managed by tailoring therapeutic interventions to the metabolic activity detected by PET. In , seven patients with RF (five new diagnoses) were referred to the same nephro-urological facility. There were six males and one female aged RF was associated with autoimmune diseases in three patients, with an aortic aneurysm in another three, and was idiopathic in one. The diagnoses were made by imaging techniques [computed tomography (CT) or nuclear magnetic resonance (NMR)]; PET scan was performed in all patients in the same setting at referral and during follow-up. Results. Patients were followed up with tailored interventions (medical therapy: tamoxifen, steroids, and immunosuppressors according to disease activity, side effects and tolerance). Six patients needed ureteral stenting for obstruction. PET imaging was used as a guide for the tapering of immunosuppressors and for stent removal. In this way, stents were safely removed when a negativization of disease activity was revealed by PET. Only one relapse was recorded over 163months of follow-up (median 24months) detected in time by PET. Conclusion. PET is a promising tool for surveillance of disease activity and for planning the removal of ureteral stents in RF. Keywords: medical therapy; positron emission tomography; renal scintigraphy; retroperitoneal fibrosis; ureteral stenting Introduction Retroperitoneal fibrosis (RF) is a complex clinical entity, still incompletely defined, encompassing a wide range of diseases and associations [1 5]. It is characterized by a fibro-inflammatory reaction, usually starting around the abdominal aorta and iliac arteries, hence the alternative name of periaortitis [2 4]. Three major causes/associations are reported for this rare disease: neoplasms, atherosclerotic aneurysms of the abdominal aorta and autoimmune diseases. Idiopathic cases have also been reported [6,7]. Whatever the cause, the inflammatory process extends into the retroperitoneum and tends to envelop neighbouring structures, first of all the ureters [1 8]. In active disease, high concentrations of acute-phase reactants are usually present; there is a close association with autoimmune diseases, leading some authors to conclude that RF is a distinct autoimmune disease [6,7,9 11]. However, no biochemical marker is closely correlated with the disease severity and progression, and imaging data, fundamental for diagnosis, often fail to discriminate between fibrotic scars and florid lesions [12,13]. The diagnosis is mainly made on radiological grounds, despite the difficulty of discriminating between active and fibrotic lesions [12,14,15]. Other authors have underlined the difficulties of differential diagnosis with some forms of retroperitoneal neoplasia and advise a biopsy for diagnosis [15]. Although the matter is still controversial, most series reported thus far are mainly based upon radiological diagnoses, and a biopsy is usually performed only in cases in which neoplasia cannot be excluded on clinical or radiological grounds [16,17]. A consensus on medical or surgical therapy has not been reached. Medical therapy consists of corticosteroids, immunosuppressive agents and the anti-fibrotic agent tamoxifen, at different doses and in various combinations [8,18 24]. The surgical approach is a matter of discussion. The Author Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 2604 G.B. Piccoli et al. Ureteral stenting is a less invasive procedure, and it is possible to remove the stents once a remission is attained. More invasive laparoscopic or open surgical techniques, even up to autotransplantation, are preferred by other specialists, who believe they allow definitive resolution of the obstructive problems [16,17,25 28]. The prognosis strictly depends upon the time of diagnosis and the type (if any) of underlying diseases; the crucial problem for follow-up, common to other rare and autoimmune diseases, is the timing of tapering or withdrawal of medical treatment, removal of ureteral stents or the planning of other surgical interventions [8,14 16,25 28]. Positron emission tomography (PET) has recently been described as a very promising tool in the diagnosis of RF, as for other systemic vasculitides [12,13,29 34]. PET sensitivity is very high, allowing detection and quantification of the metabolic activity of the retroperitoneal lesions; however, it lacks specificity and does not allow the distinction between retroperitoneal neoplastic tissue, eventually associated with RF, and metabolically active fibrotic tissue [12,13,15]. In spite of its potential, the use of fluorodeoxyglucose (FDG) PET as a guide for medical and surgical follow-up has not been extensively studied, and only a few reports deal with FDG PET and RF, with a maximum of seven cases reported (Table 1). The present paper reports a series of seven consecutive cases prospectively studied in a setting of close cooperation between urologists and nephrologists, in which tailored clinical management and ureteral stenting/stent removal were based on the disease activity detected by PET performed at referral and during follow-up. Materials and methods Study setting The prospective study gathers all patients with RF referred to the Nephrology or Urology units of the same facility (San Luigi Gonzaga University Hospital, Orbassano, Turin, Italy) since the start of a joint Outpatient Care Unit in Diagnosis was made on clinical and radiological grounds [magnetic resonance (MR) or computed tomography (CT) scans]. Medical therapy and treatment policy All cases were managed by the same physicians with a policy of tailored treatment: active disease. Step 1: Start of therapy: (1a) Start tamoxifen, if no major contraindication to corticosteroids is present, combined with oral steroids, indicatively for 6 12 months, according to clinical and PET response; (1b) if relevant contraindications are present, start with tamoxifen only. Step 2: No response: In patients on steroids, add immunosuppressor (azathioprine or methotrexate) and slowly taper steroids; in the case of tamoxifen only, choose between adding steroids or immunosuppressor. Step 3: Relapse after steroid tapering: In the absence of severe contraindications or side effects to steroids, increase the steroid dose again; in the presence of relevant contraindications, add immunosuppressor. The steroid policy was individualized; while in the absence of severe contraindications oral therapy was preferred (0.5 1 mg/kg/day as starting dose), in selected cases, pulse steroids followed by lower dose oral steroids were considered as steroid-sparing policy, in keeping with the nonhomogeneous indications of the literature [8,18 24]. Table 1. Reports on FDG PET and RF (Medline on Ovid, MESH and free terms, limits: English and Human, January 2000 June 2009; single case reports not included) Author, year (reference) Cases Main results Conclusion Salvarani 2005 [29] 7 Aim: Assessing the metabolic activity in chronic periaortitis by means of PET scan. Results: A consecutive series of seven patients seen over a 3-year period and a control group of 14 patients with malignancy was evaluated with FDG PET. All patients with RF had evidence of grade 2+ or 3+ vascular uptake in the abdominal aorta and/or iliac artery; no controls showed vascular uptake >1+; vascular uptake in the thoracic aorta and/or in its branches in 3/7. Vaglio 2005 [12] 7 Aim: Assessing post-treatment residual tissue in idiopathic RF, in the differential diagnosis between active residual disease and silent scar, by FDG PET. Results: The metabolic activity of residual masses was studied in seven IRF patients; post-treatment CT revealed a considerable reduction in the amount of IRF but all of the patients had a residual retroperitoneal mass; PET revealed slight aortoiliac F-FDG uptake only in one patient; all of the others were negative; no patients relapsed during the follow-up. Nakajo 2007 [13] 6 Aim: Describing the role of PET in six patients with IRF. Results: The test reveals intense FDG uptake in five patients before steroid treatment, no abnormal uptake in one receiving steroid treatment, thus suggesting an important role as diagnostic and prognostic tool. A limit, in the lack of differentiation between FR and neoplasia, is identified. Young 2008 [32] 3 Aim: Correlating hypermetabolic activity on PET scan with clinical data in three patients with active RF. Results: FDG PET is sensitive to disease activity: in two cases soft tissue activity was markedly diminished after immunosuppressive therapy; in one, PET following immunosuppressive therapy showed complete resolution of the retroperitoneal FDG activity. FDG PET is a promising tool in analysing metabolic activity as a sign of active disease, showing sensitivity and specificity. FDG PET is a sensitive mean to evaluate metabolic activity of residual masses in patients in clinical remission with detectable masses at CT scans. FDG PET may be a tool for evaluating extent and activity of RF, but may fail to differentiate between malignancy and RF. FDG PET showed a good correlation with clinical stability and therapeutic approach in RF patients. IRF, idiopathic retroperitoneal fibrosis.

3 PET in retroperitoneal fibrosis 2605 Surgical therapy Stent placement was planned in the case of ureteral obstruction diagnosed by renal scintigraphy or CT. During a cystoscopic procedure, a retrograde ureteropyelography was performed at the start of procedure to evaluate the excretory system. A heparin-coated double-j stent (6 Fr, 26 cm) Radiance (Cook Ireland) was then placed into the ureter. An antibiotic was used for prophylaxis at Day0, +1, +2 and +3 after the procedure. When retrograde stent placement was not possible, a nephrostomic tube was positioned with ultrasound guidance, the double-j stent was placed with an anterograde approach and the nephrostomic tube was removed. Stent substitution was scheduled within 6 months if PET showed disease activity, whereas the stent was removed when PET showed marked reduction or complete negativization of disease activity. PET PET was performed in the same facility in all cases, both at the time of diagnosis and during follow-up. Scans were routinely scheduled every 3 months in the phase of drug tapering and until complete remission, and subsequently according to individualized schedules, on the average of every 6 12 months after discontinuing therapy. Since, however, this is a rather critical population affected by multiple co-morbid conditions, the intervals were kept as indicative and subject to modification in the case of intercurrent problems. Patients were informed about the procedure and provided written informed consent. PET/CT studies were performed with the same combined PET/CT tomograph: Discovery ST (General Electric Medical Systems, Waukesha, WI, USA). The patients were requested to refrain from food intake for at least 6 h before scanning; at the time of tracer injection, all patients presented a glucose blood level <160 mg/dl. Whole-body emission scans were acquired beginning 60 min after the intravenous injection of FDG (dose range: MBq). The acquisition protocol started with a scout view (a 2D CT projection of the patient), which was used to define the body axial extension (start and end position) over which to acquire the CT and PET data. When the scan range was defined, CT was performed (voltage 140 kv, tube current 60 mas) from the proximal femur to the base of the skull. This scan lasted 1 min and was used for both anatomical localization and attenuation correction of the PET emission data. PET data on the whole-body distribution of the tracer were acquired in 3D mode from the pelvis to the neck [3 min per field of view (FOV); 8 9 FOV]. Coronal, sagittal and transverse data sets were reconstructed. The image reconstruction was performed as a 3D reconstruction algorithm FORE-Iterative, FOV: 50 cm, image matrix size: All viewing of co-registered images was performed with the appropriate software: Advantage 4.2 (GE Healthcare, Chalfont St. Giles, UK). Table 2. Main clinical data and therapeutic approaches Patient Sex Age Overall follow-up (months) Disease association(s), co-morbidity 1 M Hypertension, vasculopathy, aortic aneurysm 2 M Hypertension, psoriasis, diabetes, thyroiditis, The standardized uptake value (SUV) was calculated by drawing a region of interest (ROI) around the fibrotic lesion in PET images, i.e. we found the plane with the hottest tumour voxel, and then, we measured the maximum SUV (SUVmax) for that plane using the formula: SUV = activity (megabecquerel/millilitre) body weight (gram) / injected dose (megabecquerel). We performed the same procedure for the two adjacent planes and then used the average of these measures for the analysis. The analysis of the concordance between CT and PET scans was performed by reviewing all the CT sequences acquired along with the PET scan. The revision was performed by the same operator, unaware of the final PET results, scoring the CT scans as healed, improved or invariate. Results 3 M None Pain, low-grade 4 M Hypertension, vasculopathy, aortic aneurysm 5 M Hypertension, vasculopathy, aortic aneurysm 6 M Hypertension, thrombosis, kidney neoplasia 7 F Hypertension, diabetes, COPD, hypothyroidism, cardiopathy In the study period (January 2006 June 2009), eight cases were referred to the Nephrology or Urology units: six males and two females, median age 66 years. Two patients were re-evaluated, starting specific follow-up, while six cases were newly diagnosed. One case (a female patient) was omitted, as the fibrosis was associated with non- Hodgkin's lymphoma which was subsequently diagnosed. The main clinical data of the seven remaining cases are summarized in Table 2. None of the patients was being treated with any of the drugs known to be associated with RF. The age range is relatively wide (41 79 years), and all but one patient displayed at least one co-morbid condition or disease association, half of them with aortic aneurysms; autoimmune diseases or positive autoantibodies were present in most cases (three patients with distinct diseases; six of seven with positive autoantibodies). At diagnosis, all patients with active disease displayed an increase in acute-phase reactants. However, the level of C-reactive protein (CRP), usually chosen as the main marker of the disease, was not uniformly related with the severity of the clinical picture nor with the extension of the RF at PET or CT scans (Table 3); this was particularly evident in Case 2 and 6 in which active disease was recorded along with a mild increase of CRP. Symptoms (time to resolution) Auto-Abs Medical therapy Time to stent removal None ENA low grade Tamoxifen, steroids, AZA 7 months Pain, malaise, Antinuclear 1:160 Tamoxifen, 6 months (1 week) steroids, CyA, MTX None Tamoxifen, 7 months fever (2 weeks) steroids, AZA None Antinuclear 1:80 Tamoxifen After restaging Mild back pain Antinuclear 1:80 Tamoxifen, 5 months (3 4 weeks) steroids None LLAC Watch-and-wait None Malaise, back pain (better after stenting) Anti-thyroglobulin Tamoxifen 10 months, removal planned M, male; F, female; Anti-Abs, auto-absorption; AZA, azathioprine; COPD, chronic obstructive pulmonary disease; CyA, cyclosporine A; ENA, extractable nuclear antigens; LLAC, lupus-like anticoagulant; MTX, methotrexate.

4 2606 G.B. Piccoli et al. Table 3. Comparison between PET scan, CT scan and main inflammatory markers at referral and during follow-up Patient Sex Baseline: SUV/CT Baseline: CRP mg/dl/ ESR (mm/h) Baseline: Crs mg/dl (GFR ml/min) Follow-up at first significant change in PET CRP/Crs (mg/dl) at first change SUV/Qualitative CT evaluation at first change Follow-up at negative or last PET CRP/Crs (mg/dl) at negative or last PET SUV/Qualitative CT evaluation negative or last PET 1 M 5.6/Pos (59) 6 months /B 19 months Neg/I 2 M 6/Pos 1 > (55) 5 months <0.5 1 Neg/H a 10 months <0.5 1 Neg/H-I a 3 M 7.7/Pos (70) 6 months Neg/B b 24 months Neg/B b 4 M Neg/Pos (55) As the first PET disclosed no metabolic activity, only first and last controls were considered. 24 months < Neg/I 5 M 6.3/Pos (35) 3 months /B 6 months < Neg/I 6 M 8.5/Pos (50) As no therapy was started and metabolic 13 months /I activity at PETwas stable, only the first and last controls were considered. 7 F 3.5/Pos 5 n.a. 7.3 (<10) As the second PET disclosed no metabolic activity, only the first and second controls were considered. 4 months Neg/B n.a, not available; ESR, erythrocyte sedimentation rate (millimetre per hour); Crs, serum creatinine; CRP, C-reactive protein (milligram per decilitre; normal <0.5); SUV, standard uptake value (<2.5: negative). CT scores: Pos, positive (RF would have been diagnosed at CT scan); B, better (significant reduction of the retroperitoneal fibrotic tissue, with evidence of residual disease); I, invariate. H, healed (minimal or no evidence of retroperitoneal tissue). a PET performed at time of the severe psoriasis relapse, at start of MTX therapy. b Data during relapse: 19 months; SUV: 8.5; CT: worsened. Two months later (clinical remission): SUV: 3.6; CT: better. While, during follow-up, a trend towards normalization of CRP was observed, the association is once more not absolute (Case 7), possibly because of the presence of other inflammatory or autoimmune co-morbidities. Furthermore, while CT scans were sensitive in detecting the presence of the disease, only a complete resolution was observed in one patient, while in the other six, the presence of residual tissue was described. The revision of the CT images, recorded during PET scan, did not allow for the discrimination between residual tissue and active versus non-active disease (Table 3). Detailed descriptions of the patients are reported below. Case 1 A 59-year-old male was referred in January 2006 because of moderate reduction of the glomerular filtration rate (GFR) and left hydronephrosis at routine abdominal ultrasounds, performed in the follow-up of an aortic aneurysm (4 cm). The clinical history was uneventful until age 53 when he experienced an acute myocardial infarction with percutaneous angioplasty and stenting. The diagnosis of RF was suggested by medialization of the ureters with bilateral incomplete obstructive pattern at renal scintigraphy and was confirmed by CT scan; PET imaging showed active disease at the aortic and iliac levels. After a second myocardial ischaemia and angioplasty, he underwent bilateral ureteral stenting (June 2006). Treatment was started with tamoxifen (20mg twice daily) because of the favourable cardiovascular profile. As the PET control was unchanged (October 2006), prednisone was started (1 mg/kg/day) with slow tapering. In February 2007, PET revealed reduction of the metabolic activity; azathioprine 100 mg was added, and tamoxifen was withdrawn, with clearing of the PET activity. Stents were removed in June As of June 2008, the patient is in remission and on azathioprine 50 mg daily. Renal function is stable (creatinine mg/dl, GFR ml/min). Case 2 A 68-year-old male, affected by psoriasis, autoimmune hypothyroidism and type 2 diabetes, was referred in November 2006 to the Urology Unit for renal colicky pain and bilateral hydronephrosis. RF was diagnosed by CT scan; moderate renal insufficiency was present at diagnosis (serum creatinine 1.3 mg/dl, GFR 55 ml/min); PET showed an active disease. Bilateral obstruction was relieved by ureteral stenting, allowing preservation of kidney function. Due to the severe clinical impairment, with weight loss, systemic symptoms and severe abdominal pain, corticosteroids were started (prednisone 0.9 mg/ kg/day) in association with tamoxifen (20 mg twice daily), with prompt clinical response. Rapid negativization of metabolic activity allowed removal of the stents (June 2007). However, in July 2007, the patient experienced a severe psoriatic flare-up, leading to the addition of cyclosporine, which was poorly tolerated, and then methotrexate, with slow remission of the psoriasis. The patient has been free from steroids since June 2008 and is in full clinical remission on tamoxifen 20 mg twice daily, continuing methotrexate 15 mg/week for psoriasis control (Figure 1). Case 3 A 41-year-old male was hospitalized in December 2006 because of abdominal and flank pain and moderate reduction of renal function. RF was detected by CT scan; renal scintigraphy detected significant left obstruction, with no dilatation of the renal pelvis.

5 PET in retroperitoneal fibrosis 2607 Left ureteral stenting restored kidney function (serum creatinine 1, GFR >100 ml/min); corticosteroids (prednisone 50 mg/day) were started in association with tamoxifen (20 mg twice daily) with initial benefit and rapid remission of the abdominal pain, but clinical relapse occurred after slow tapering (25 mg daily). After addition of azathioprine (150 mg/day), PET activity rapidly cleared, and the ureteral stent was successfully removed (July 2007). Steroids were slowly tapered over 12 months, but in July 2008, a further relapse was detected at the scheduled PET (prednisone 5 mg/day). Interestingly, the detection of disease activity at PET preceded the increase in serum creatinine (up to 1.63 mg/dl). Increasing steroid doses to 25 mg/day promptly restored renal function, without the need for ureteral stenting. In June 2009, the patient was still in remission on prednisolone 15/10 mg on alternate days and azathioprine 150 mg/day (Figure 2). Case 4 A 71-year-old male, with RF associated with aortic aneurysm (45 mm), was referred in June 2007 for re-evaluation. He had been treated, with erratic compliance, with immuno- Fig. 1. Case 2. Upper scan: first scan at diagnosis. Lower scan: complete remission. This is the only patient in whom complete resolution of the retroperitoneal mass is observed at PET and CT scans.

6 2608 G.B. Piccoli et al. suppressants (azathioprine, methotrexate and steroids) and had recently self-withdrawn all therapies, except azathioprine 50 mg/day and occasional anti-hypertensives. He was on regular follow-up with periodic substitution of renal stents (every 6 months, since 2003). Since diagnosis, serum creatinine ranged from 1.8 to 2.2 mg/dl (GFR ml/ min). Tamoxifen was added for disease stabilization; PET scan did not show any residual activity, and the patient underwent safe removal of the ureteral stents in November As of June 2009, renal function is stable (serum creatinine mg/dl) on tamoxifen alone. Case 5 A 74-year-old male, referred from the Cardiology Unit in April 2008 because of moderate reduction of renal function and unilateral hydronephrosis, was diagnosed during follow-up of a aortic aneurysm (4 cm). The clinical history showed arterial hypertension for at least 15 years. RF was diagnosed by CT scan, and intense activity was detected by PET. Bilateral signs of obstruction were present at scintigraphy, and bilateral stents were positioned in June Due to the fragile cardiovascular status, bolus steroids were chosen as a steroid-sparing policy (methylprednisolone 475 mg in 5 days, followed by oral prednisone 0.5 mg/kg/day) in association with tamoxifen (20 mg twice daily). The occurrence of deep venous thrombosis, in the context of severe varicose veins, led to tamoxifen withdrawal in July The patient was hospitalized in October 2008 for atrial flutter and pulmonary oedema; in the absence of metabolic activity on PET, the renal stents were successfully removed during hospitalization. At the last control, the patient was in remission on prednisone 10 mg daily. Case 6 A 53-year-old male was referred in June 2008 because of moderate reduction of renal function. The medical history showed arterial hypertension since age 38, left nephrectomy for papillary renal carcinoma in 2000, contralateral relapse treated by radio frequencies in 2004 and lupus anticoagulant (LAC) positivity with cerebral vasculopathy and ischaemic cardiopathy (acute myocardial infarction and coronary stenting in 2003). Diagnosis of RF dated in 1999, without signs of ureteral involvement. No specific treatment was undertaken for RF. At the start of follow-up in our unit, the clinical conditions were good, there was no upper urinary tract dilatation and the renal scintigraphy displayed a normal excretion phase. Due to the risk of side effects (difficult blood pressure control with corticosteroids; immunosuppressive therapy was not felt to be safe due to the history of neoplasia, and tamoxifen was feared because of the risk of erectile deficit) and due also to the absence of symptoms, a watch-and-wait policy was chosen. Two further PET scans were performed, and the clinical conditions remained stable as of June Case 7 A 79-year-old woman was referred in August 2008 because of acute renal failure superimposed on chronic renal disease (serum creatinine 7.3 mg/dl, GFR <10 ml/min). RF was diagnosed by CT scan; PET scan demonstrated a mild but diffuse disease activity. The clinical history showed type 2 diabetes for 10years, arterial hypertension for 20 years, chronic bronchopneumopathy and autoimmune hypothyroidism. The great thickness of the periaortic reaction suggested chronic long-lasting disease; in keeping with this interpretation, moderate dilatation of the renal pelvis was present for at least 3 years, previously attributed to pyeloureteral junction disease. The patient had been treated during the past 5 years with several cycles of steroids for severe chronic obstructive pulmonary disease, which could have blunted the disease activity. Positioning of renal stents by anterograde approach restored renal function. Due to the presence of mild, albeit diffuse, activity at PET and severe co-morbidity, she was started on tamoxifen only (20 mg twice daily), with significant clearing of the PET activity at the first control. As of June 2009, the patient was in clinical remission; stent removal is planned. Discussion This case series of seven consecutive patients with RF is one of the largest in which therapeutic interventions were tailored to the disease activity detected by PET scans in a Fig. 2. Case 3. PET pattern at referral (top left), first response (bottom left), relapse (centre), current remission (right).

7 PET in retroperitoneal fibrosis 2609 setting of close cooperation between nephrologists and urologists (Table 1) [12,18,31 35]. Because of the uronephrological referral pattern, the disease had led in all but one patient to clinically relevant renal obstruction, assessed by renal scintigraphy and/or confirmed by significant dilatation by ultrasound or CT scan; one patient (Case 7) was hospitalized for acute oliguric renal failure. However, in contrast with other reports [17,18,27] in which acute renal failure was almost the rule, diagnosis in our series occurred relatively early, and all but one patient displayed a GFR >30 ml/min at diagnosis (Table 2). Rapid relief of the obstruction by renal stenting, chosen in the context of a minimally invasive policy, was the first step in five out of seven cases (Table 1). In all but one case, in which a wait-and-watch policy was chosen, an individualized stepwise approach, according to a policy of tailored interventions, guided the choice of medical therapy; the therapeutic choices were modulated on the basis of the expected side effects, starting with tamoxifen in the most fragile patients, on account of its safe cardiovascular profile, and adding steroids, as a second step, followed by immunosuppressors in the case of a lack of response upon PET scan [17 23,35,36]. In agreement with the literature, remission was attained in all treated cases, even if the timing of the response was different (3 months to >1 year) [17 23]. While the followup is still relatively short (median 24months) and the case series is relatively small (limits shared by most reports on this rare disease), an important finding is the lack of unexpected relapses when the tapering policy is controlled by the metabolic activity detected upon PET scan. In our series, the disease relapsed in one patient only (Case 4), who experienced two relapses; interestingly, the second one was documented by a positive PET scan before clinically relevant obstruction appeared and before a rise in acute-phase reactants [PCR and erythrocyte sedimentation rate (ESR)] and serum creatinine (Figure 1). On both occasions, the timely diagnosis allowed for modulation of the therapy: steroids were increased, with prompt reduction of the metabolic activity and without the need to reinsert ureteral stents. The control policy is the Achilles heel of RF therapy. Indeed, the disease tends to respond well to almost all therapies proposed in the literature, and does so in a stepwise fashion [17 24]. As a first step, the local and systemic clinical signs of the disease remit rapidly. This finding is reported in several case series and is confirmed in ours, which included only four symptomatic cases, probably because of the early diagnosis; in all four cases, the main clinical complaints remitted within days weeks after the start of therapy (Tables 2 and 3; [17 24]). The second step involves the decrease or normalization of the non-specific signs and symptoms of inflammation, mainly C-reactive protein. According to the literature, they display a rapid but often incomplete response within a few weeks [15 19,29]. While these data correlate with the clinical response, their practical interest as a guide for stent removal and for drug tapering is limited, as normal values may co-exist with active disease shown at PET [12,29,31]. The reverse may also be true; in several series, the levels of CRP had not fully returned to normal at the time of stent removal and normalized PET scan. In our series, both discrepancies are present, thus confirming the limited value of the usual markers of active-phase reactants in guiding delicate therapeutic decisions in RF (Table 3). The close association with autoimmune or neoplastic diseases and the concomitance with chronic infections, as described in our series, may offer alternative grounds for the increase in acute-phase reactants, whose pattern, however, is unrelated to RF activity (Table 2). Likewise, the re-evaluation of the CT scans acquired contextually with PET scans confirmed the diagnostic value of CT scans in identifying the fibrotic tissue, but at the same time, there was a lack of discrimination between active and residual fibrotic tissue (Table 3). In this context, in which complex and potentially toxic therapies have to be modulated or ureteral stenting/removal has to be planned, FDG PET scan appears to be the most reliable therapeutic guide. The absence of unexpected relapses in our series may further support this policy that is already advocated by others [12,13,29 34]. Several questions remain: the overall duration of therapy, the policy of controls, the length and modality of drug tapering, and the frequency of PET scans during therapy and after disease remission. Our policy was one of the frequent controls, on average every 3 months, until remission was attained, according to the hypothesis that the morbidity due to PET was minor with respect to the drug- and stent-related morbidity. This favoured timely stent removal and slow steroid tapering, in the context of a minimally invasive surgical approach. As PET is a rather expensive and not readily available technique, one of the goals of further analysis will be to define a favourable cost/benefit policy in this regard. This goal will, however, probably be attained only in larger, eventually multicentre experiences. Conclusions Our data support the use of PET, not only for the diagnostic workup but also in the follow-up of patients with RF, in order to tailor medical and surgical approaches. In particular, the comparison between the PET study acquired during the diagnostic workup and during follow-up may be of use to identify the best time for safe removal of the renal stents and the schedule for tapering of steroids and immunosuppressors, crucial issues to reduce morbidity and improve the quality of life of RF patients. Acknowledgements. revision. We thank Dr. Peter Christie for his careful language Conflict of interest statement. None declared. References 1. Parums DV. The spectrum of chronic periaortitis. Histopathology 1990; 16: Gilkeson GS, Allen NB. Retroperitoneal fibrosis: a true connective tissue disease. Rheum Dis Clin North Am 1996; 22: Mitchinson MJ. Chronic periaortitis and periarteritis. Histopathology 1984; 8:

8 2610 G.B. Piccoli et al. 4. Vaglio A, Buzio C. Chronic periaortitis: a spectrum of diseases. Curr Opin Rheumatol 2005; 17: Vaglio A, Salvarani C, Buzio C. Retroperitoneal fibrosis. Lancet 2006; 367: Haug ES, Skomsvoll JF, Jacobsen G et al. Inflammatory aortic aneurysm is associated with increased incidence of autoimmune disease. J Vasc Surg 2003; 38: Vaglio A, Corradi D, Manenti L et al. Evidence of autoimmunity in chronic periaortitis: a prospective study. Am J Med 2003; 114: Moroni G, Gallelli B, Banfi G et al. Long-term outcome of idiopathic retroperitoneal fibrosis treated with surgical and/or medical approaches. Nephrol Dial Transplant 2006; 21: Littlejohn GO, Keystone EC. The association of retroperitoneal fibrosis with systemic vasculitis and HLA-B27: a case report and review of the literature. J Rheumatol 1981; 8: Okada H, Takahira S, Sugahara S et al. Retroperitoneal fibrosis and systemic lupus erythematosus. Nephrol Dial Transplant 1999; 14: Moroni G, Farricciotti A, Cappelletti M et al. Retroperitoneal fibrosis and membranous nephropathy: improvement of both diseases after treatment with steroids and immunosuppressive agents. Nephrol Dial Transplant 1999; 14: Vaglio A, Greco P, Versari A et al. Post-treatment residual tissue in idiopathic retroperitoneal fibrosis: active residual disease or silent scar? A study using 18 F-fluorodeoxyglucose positron emission tomography. Clin Exp Rheum 2005; 23: Nakajo M, Jinnouchi S, Tanabe H et al. 18F-Fluorodeoxyglucose positron emission tomography features of idiopathic retroperitoneal fibrosis. J Comput Assist Tomogr 2007; 31: Vivas I, Nicolas AI, Velazquez P et al. Retroperitoneal fibrosis: typical and atypical manifestations. Br J Radiol. 2000; 73: Cronin CG, Lohan DG, Blake MA et al. Retroperitoneal fibrosis: a review of clinical features and imaging findings. Am J Roentgenol 2008; 191: Ilie CP, Pemberton RJ, Tolley DA. Idiopathic retroperitoneal fibrosis: the case for nonsurgical treatment. BJU Int 2006; 98: Fry AC, Singh S, Gunda SS et al. Successful use of steroids and ureteric stents in 24 patients with idiopathic retroperitoneal fibrosis: a retrospective study. Nephron Clin Pract 2008; 108: c213 c Marcolongo R, Tavolini IM, Lavender F et al. Immunosuppressive therapy for idiopathic retroperitoneal fibrosis: a retrospective analysis of 26 cases. Am J Med 2004; 116: Warnatz K, Keskin AG, Uhl M et al. Immunosuppressive treatment of chronic periaortitis: a retrospective study of 20 patients with chronic periaortitis and a review of the literature. Ann Rheum Dis 2005; 64: Maillart E, Laueriere L, Kassis S et al. Is there an evidencebased management of idiopathic retroperitoneal fibrosis? Revue de Medecine Interne 2006; 27: Kardar AH, Kattan S, Lindstedt E et al. Steroid therapy for idiopathic retroperitoneal fibrosis: dose and duration. J Urol 2002; 168: Scheel PJ Jr, Piccini J, Rahman MH et al. Combined prednisone and mycophenolate mofetil treatment for retroperitoneal fibrosis. JUrol 2007; 178: discussion van Bommel EF, Hendrisksz TR, Huiskes AW et al. Brief communication: tamoxifen therapy for nonmalignant retroperitoneal fibrosis. Ann Intern Med 2006; 144: Vega J, Goecke H, Tapia H et al. Treatment of idiopathic retroperitoneal fibrosis with colchicine and steroids: a case series. Am J Kidney Dis 2009; 53: Mufarrij PV, Lipkin ME, Stifelman MD. Robot-assisted ureterolysis, retroperitoneal biopsy, and omental wrap: pilot series for the treatment of idiopathic retroperitoneal fibrosis. J Endourol 2008; 22: Srinivasan AK, Richstone L, Permpongkosol S et al. Comparison of laparoscopic with open approach for ureterolysis in patients with retroperitoneal fibrosis. JUrol2008; 179: Stifelman MD, Shah O, Mufarrij P et al. Minimally invasive management of retroperitoneal fibrosis. Urology 2008; 71: Duchene DA, Winfield HN, Cadeddu JA et al. Multi-institutional survey of laparoscopic ureterolysis for retroperitoneal fibrosis. Urology 2007; 69: Salvarani C, Pipitone N, Versari A et al. Positron emission tomography (PET): evaluation of chronic periaortitis. Arthritis Rheum 2005; 53: Wenger M, Calamia KT, Salvarani C et al. Do we need 18F-FDGpositron emission tomography as a functional imaging technique for diagnosing large vessel arteritis? Clin Exp Rheumatol 2003; 21: S1 S2 31. Pipitone N, Ghinoi A, Versari A et al. Images in cardiovascular medicine. Chronic periaortitis. Circulation 2008; 118: Young PM, Peterson JJ, Calamia KT. Hypermetabolic activity in patients with active retroperitoneal fibrosis on F-18 FDG PET: report of three cases. Ann Nuclear Med 2008; 22: Quante M, Appenrodt B, Randerath S et al. Atypical Ormond's disease associated with bile duct stricture mimicking cholangiocarcinoma. Scand J Gastroenterol 2009; 44: Vaglio A, Versari A, Fraternali A et al. 18 F-Fluorodeoxyglucose positron emission tomography in the diagnosis and follow-up of idiopathic retroperitoneal fibrosis. Arthritis Rheum 2005; 53: Grainger DJ, Shofield PM. Tamoxifen for the prevention of myocardial infarction in humans: preclinical and early clinical evidence. Circulation 2005; 112: Wie L, MacDonald TM, Walker BR. Taking glucocorticoids by prescriptions is associated with subsequent cardiovascular disease. Ann Intern Med 2004; 141: Received for publication: ; Accepted in revised form: