Benefits of hybrid methods (PET/CT, PET MRI) in the diagnosis of abdominal aortic pathology

Summary

Introduction: Ultrasound and CT angiography are common diagnostic methods of abdominal aortic pathologies. In the last decade, hybrid methods (PET/CT, PET/MRI) have become more common in this diagnostic algorithm. Originally they were indicated in malignancies or inflammatory processes. Currently, efforts are developed to visualize possible local inflammatory activity in the aortic wall and thus to assess a certain “disease activity” with the goal to anticipate further development of aortic pathology. The aim of our study was to analyze potential benefits of hybrid methods in predicting abdominal aortic pathology progression.

Methods: In this prospective, open-label, observational study we examined 75 patients referred to PET/CT (N=61) or PET/MRI (N=14) due to any aortic pathology in 2015–2017. The patients included those with abdominal aortic aneurysm (AAA) (N=48; 64%), aortitis (N=5; 6.7%), aortic dissection (N=4; 5.3%), patients undergoing EVAR (N=6; 8%), patients with excessive atherosclerosis (N=7; 9.3%), patients with concomitant AAA and retroperitoneal fibrosis (N=4; 5.3%) and patient with an intramural hematoma (N=1; 1.3%). The minimum follow-up period was 6 months (0.5–2.5 years). Clinical symptoms, aortic diameter, growth rate and CRP levels were analyzed during the follow-up and correlation with PET/CT or PET/MRI findings was evaluated.

Results: Increased metabolic activity in the aorta was found in 25 of the 75 examined patients (33.3%). Based on statistical analysis there were no associations between increased activity based on PET/CT or PET/MRI in the aortic wall and disease symptoms or progression.

Conclusion: Our results provide no evidence that hybrid methods can predict further development of pathological findings in the abdominal aorta. PET/CT- or PET/MRI-based activity did not correlate with disease symptoms, AAA progression rate or dissection, either. Our results are also supported by some recent literature data.

Keywords:

dissection – Aorta – PET/CT – PET/MRI – aneurysm

Introduction

At present, ultrasound scanning (USG) and CT angio­graphy (CTA) are fully adequate for the diagnostic algorithm of abdominal aortic pathology in a great majority of cases, and both are routinely used in clinical practice. CTA in particular has high sensitivity and specificity for pathological conditions occurring in the aorta [1].

The question is what is the reason for indicating new hybrid techniques – positron emission tomography combined with CT scanning (PET/CT), and positron emission tomography combined with magnetic reso­nance imaging (PET/MRI). In addition to commonly used indications of these methods for inflammatory or rarely tumorous involvement of the aorta, hybrid methods are increasingly being used for “functional diagnosis” in abdominal aortic aneurysms (AAA) as well as other pathologies [2]. The aim is to try to visualize possible local inflammatory activity and thus certain “disease activity” in the aortic wall with the intention to anticipate further development of AAA.

This hypothesis can also be formulated in other aortic pathologies such as dissection, intramural hematoma, inflammation, etc. The aim of this paper was to try to find a correlation between activity in the aortic wall based on PET/CT or PET/MRI imaging and the behavior, symptoms and progression of potential AAA or non-dilated aorta.

Methods

Patients

In this prospective, open-label, observational study we have included 75 patients referred to PET/CT or PET/MRI at the University Hospital due to any aortic pathologies in 2015–2017. PET/CT was performed in 61 and PET/MRI in 14 patients. All were Caucasians, >18 years old. The indications for the hybrid methods included: AAA (N=48; 64%), suspected aortitis (N=5; 6.7%), aortic dissection (N=4; 5.3%), condition after endovascular AAA treatment (N=6; 8%), excessive atherosclerosis (N=7; 9.3%), concomitant AAA and suspected retroperitoneal fibrosis (N=4; 5.3%) and intramural hematoma (N=1; 1.3%). The study was approved by the institutional ethical review board and the patients signed an informed consent form.

Assessment and follow-up

All patients were examined repeatedly; the minimum follow-up period was 6 months (0.5–2.5 years). Assessments performed at the first visit included PET/CT or PET/MRI, physical examination and evaluation of the circulating biomarker C-reactive protein (CRP). Elevated CRP levels were defined as >20 mg/L. Additional examinations included any imagine technique to compare the aortic size (ultrasound, CTA, PET/CT, PET/MRI), and also physical examination.

Imaging methods

The PET/CT was performed using an instrument with a 128-line CT subsystem and a four-ring PET subsystem (Biograph mCT 128, Siemens Healthineers, Knoxville, TN, USA) following administration of 18F-FDG (fluorodeoxyglucose, UJV, Rez u Prahy, Czech Republic) at 2.5 MBq/kg. CT data were acquired using a 100 kV protocol with automatic voltage reduction down to 80 kV in sthenic patients. Data were reconstructed in diagnostic images using iterative SAFIRE data reconstruction at a 0.75mm image width with 0.6mm reconstruction increment and a soft tissue algorithm. PET data were reconstructed in a 400x400 point matrix in a 46cm diameter transaxial field-of-view using the ultraHD algorithm that combines the TOF (time-of-flight) and PSF (point spread function) algorithms with 2mm FWHM. Data acquisition was performed by the step-and-shoot technique using 5 to 7 positions, with the data acquisition time of 1.5 minutes per position. The CT scan was performed after intravenous administration of iodinated contrast agent (Ultravist 370, Bayer Pharma, Berlin, Germany) at a quantity of 60–100 mL according to the weight of the examined person. The examination was performed in the early arterial phase throughout the body and in the late venous phase in the abdomen and pelvis.

PET/MRI scanning was performed using an integrated PET/MRI system (Biograph mMR, Siemens Healthineers, Erlangen, Germany) with the administration of ¹⁸F-FDG at 2.5 MBq/kg. As part of the examination, whole-body scanning was performed using T1 VIBE Dixon two-point, TIRM T2 STIR and DWI sequences following administration of gadolinium contrast agent. The reconstruction of PET images was done using the VIBE sequence to correct attenuation; OSEM reconstruction algorithm was used. The dose of 4 mL of 1-molar gadobutrol (Gadovist, Bayer Pharma, Berlin Germany) was used. PET data acquisition was performed over 4 minutes in each scanning position of 5 to 7 positions.

Evaluation of imaging methods

The evaluation was performed using the dedicated syngo.via MM software (Siemens Healthineers, Forchheim, Germany). The analysis was carried out by consensus of two experienced specialists double boarded in radiology and nuclear medicine who used ROI (region of interest) analysis to determine the maximum accumulation of FDG in the vascular wall and, at the same time, performed an overall assessment of changes in the abdominal aortic wall.

Terminology

In consistence with recent literature, we use the term “mycotic AAA” for AAA clearly caused by infections, in most cases with confirmation of infectious agents in the aortic wall or blood culture [3,4]. On the other hand, we save the term “inflammatory AAA” for AAA without clear infectious etiology where an autoimmune component is likely to be important [5].

Statistical analysis:

The log-rank test was used for the analysis, and associated p values <0.05 were defined as significant.

Results

Increased metabolic activity in the non-dilated aorta either in the AAA wall, atherosclerotic plaque, or an intramural hematoma on PET/CT or PET/MRI was found in 25 of the 75 examined patients (33.3%).

Overall, in terms of increased activity in PET/CT or PET/MRI, we found an atypical positive finding in these patients (15 cases of AAA, 1 after EVAR), 3 cases of aortitis, 6 cases of atherosclerotic changes, 1 case of intramural hematoma). In the remaining 50 patients (66.7%), we did not observe increased metabolic activity in the aortic area.

Data on patients with increased activity in the aortic wall are shown in Tab. 1.

A summary of all patients examined, their findings in the aorta and potential correlation of symptoms with positive PET/CT or PET/MRI findings and correlation of CRP levels with positive PET is presented in Tab. 2.

There were no associations between increased activity during PET/CT or PET/MRI in the AAA wall and its symptoms (pain, fever…) or growth rate (growth rate was analyzed as a increase diameter of AAA in the follow-up period).

Similarly, we found no association between atherosclerotic aortic wall activity and clinical manifestations or further progression of these findings.

We found an association between CRP levels and positive findings in PET/CT o PET/MRI in aortitis but the number of patients was insufficient for statistical analysis.

Discussion

For years, many research teams have been trying to develop a method that could predict the “behavior” of AAA in clinical practice. This includes in particular a prediction of progression of the aneurysm size, which may serve as an indication for surgical or endovascular intervention. Rutherford postulated the relationship between the AAA size and the risk of its rupture many years ago [6]. In recent years, the aneurysm volume has been reported, in addition to the maximum dimension, as an additional marker of rupture risk [7]. On the other hand, a number of papers describe rupture of small aneurysms (smaller than 5 cm), while large AAAs can be stable [8]. It is clear that the AAA size may not be the only factor playing a role in the risk of rupture. The ideal goal of research in this field would be to obtain a method predicting the potential progression of AAA. This would be especially useful for small AAAs where open or endovascular management is not yet indicated or for patients who have large aneurysms but a high operation risk.

Many teams have been involved in modeling AAA rupture risk based on numerical simulation [9–11]. Another potential method studied aneurysm wall “distensibility”, performing an evaluation of a certain “stretching capacity” of the aneurysm wall during the pulse wave [12,13]. All of these methods bring promise for the future but are not routinely used in clinical practice.

A lot of evidence supports the key role of inflammation in the AAA etiopathogenesis [14–17]. Options for monitoring inflammatory activity in vessel walls are limited and include measurements of circulating markers such as MMPs, inflammatory cell activity markers and cytokine levels. Despite extensive research in this field, none of these markers have not integrated into clinical practice as their circulating levels do not necessarily reflect their levels in the aortic wall. Indeed, a clear correlation between any of the circulating cytokines and AAA symptoms has not been confirmed by extensive studies.

In the last decade, the benefits of hybrid imaging techniques have been considered a promising option. These imaging techniques, which were primarily developed for the detection of inflammation or malignancy, may be promising, according to a number of papers, in detecting this inflammatory process in the aortic wall to predict the behavior of aneurysms, as well as atherosclerotic ulcers or aortic dissection [18–20]. The anticipated benefit of PET/CT scanning was the detection of an active ongoing inflammatory process in the aortic wall or even aortic blebs, described many years ago, which are local deformities often infiltrated by polymorphonuclear cells [21].

A number of clinical trials have demonstrated selective binding of transporters used in hybrid techniques to microcalcifications in coronary or carotid arteries, visualizing a “vulnerable” plaque with a higher risk of rupture [22]. Further work has shown a correlation between a higher uptake of 18F-FDG and higher AAA wall stress [23]. This is probably also related to a correlation between 18F-FDG activity and higher long-term growth of AAA [24,25]. The ability of hybrid techniques to detect inflammatory activity in the AAA wall, in particular the activity of macrophages, has been demonstrated [26]. In general, a number of papers show a higher activity in the AAA wall in PET/CT scanning as a marker of disease activity [27]. On the other hand, some studies have rejected this hypothesis [26,28]. One of the largest meta-analyzes dealing with this issue was published in 2015 with contradictory conclusions of individual studies (29).

The basic question we asked in our study is whether PET/CT or PET/MRI can be effectively used to monitor the development, progression and potential symptoms of AAA and other pathological conditions in the aorta. In our group of patients, we found an increased FDG uptake on PET/CT or PET/MRI in 14 of 48 cases of AAA (29.2%), but only 7 patients had clear symptoms (pain) or a fast-growing, mycotic or inflammatory AAA. On the other hand, hybrid imaging did not reveal higher activity in the remaining 34 cases of AAA in our population, which also included patients with rapidly growing or symptomatic aneurysms (N=8; 23.5%). Therefore, we have not found a clear correlation between higher activity in PET/CT (PET/MRI) and the symptoms or growth rates in patients with AAA. Similarly, we found no correlation between positive findings in excessive aortic atherosclerosis and potential symptoms or further progression of the aortic pathology. The numbers of patients with aortic dissection or aortitis are too small for any statistical evaluation.

In our study we did not confirm certain conclusions of other authors about the contribution of hybrid techniques as a functional examination able to predict potential development of a degenerative aortic disease [29]. Our conclusions are supported by other recent publications (40).

Many situations or diagnoses have been reported in the abdominal aorta where hybrid techniques have played an important role. In case of aneurysms or undilated aorta where we assume an inflammatory etiology (whether infectious or non-infectious), there is usually no doubt about the contribution of hybrid techniques. PET/CT and PET/MRI are able to reliably detect active infectious inflammation in the aortic wall (Fig. 1) [30]. In our study we observed this in 6 patients. It is also possible to recognize the extent of the inflammatory process, its location and potential transfer from another primary source (e.g., spondylodiscitis, pancreatitis), and the effect of potential conservative antibiotic therapy may be monitored (e.g., in Salmonella aortitis or Salmonella-induced AAA). Likewise, PET/CT may trigger the suspicion of inflammatory AAA, an aneurysm with wall inflammation and thickening which lacks infectious etiology (Fig. 2). The etiopathogenesis of these aneurysms has not yet been fully clarified. Knowing about the presence of this type of aneurysm before surgery is very beneficial as open surgery is quite difficult and very risky, sometimes impossible. The endovascular approach is therefore strictly preferred in these patients. Hybrid techniques are very well able to reveal retroperitoneal fibrosis concomitantly present with AAA, which again has an impact on treatment strategy decisions (Fig. 3). PET/CT (PET/MRI) is also indicated for suspected infectious complications both after open surgery and after EVAR. A hybrid technique with high sensitivity and specificity reveals this serious complication [31]. However, it cannot be assumed that increased activity in the AAA sac wall after EVAR or in the left sac after resection will automatically suggest an infectious complication. In our study, we also observed a case (after EVAR) of increased activity in the aortic wall, but neither clinical nor laboratory findings suggested infectious complications. Similarly, we have repeatedly observed increased activity in the atherosclerotic aorta at the site of the plaque, but no progression occurred during longer follow-up periods (Fig. 4). Some papers address the benefits of hybrid techniques in the diagnosis of late endoleak after EVAR, also without clearly convincing results [32].

The above mentioned indications, especially in cases of a suspected infectious process in the aortic wall, are usually indisputable, and there are no doubts about the benefits of hybrid techniques in these cases, either. According to recent papers, the incidence of aortitis of various etiologies has been increasing, probably with an increasing number of differently immunosuppressed individuals (33), so we can expect more frequent indications of hybrid techniques in these patients. A discussion is still ongoing over the advantages and disadvantages of PET/CT and PET/MRI, and in which situations to choose a particular method. Due to the slightly different course of the scanning, a clear consensus is needed as to the indications of the respective techniques between angiolo­gists or vascular surgeons and radiologists. Due to the relatively low experience with PET/MRI scanning in vascular surgery, it cannot be ruled out that different conclusions may be reached in larger patient populations.

Our study has some limitations, including the small number of patients and a relatively short observational time. However, compared to other available studies, our group of patients is comparable. Heterogeneity of the patient population is another limitation – which at the same time is an advantage, though, as it gives an opportunity to compare different diseases. A benefit of our study consists in using not only PET/CT but also PET/MRI.

Conclusion

From our point of view, the indications for PET/CT and PET/MRI remain questionable in small aneurysms, in high-risk patients with AAA and in patients with dissection or excessive atherosclerosis. Of course, unless there is another reason for a hybrid imaging technique. Our hypothesis expected to find a benefit in terms of prediction of further behavior of the aortic pathology. We were unable to demonstrate the intended benefit, i.e. obtaining information about “disease activity” and therefore predicting the progression of the findings in our study population. We are aware of the many study limitations, yet we tend to agree in our conclusions with those authors who have also failed to demonstrate the benefits of hybrid imaging techniques in the above situations.

This article was supported by the project UK Progres Q39 and AZV 15-32727A

Conflict of interests

The authors declare that they have no conflict of interest in connection with this paper and that the article has not been published in any other journal.

doc. MUDr. Jiří Moláček, Ph.D.

Chirurgická klinika FN a LF v Plzni

Alej Svobody 80

304 60 Plzeň

e-mail: molacek@fnplzen.cz