1 3

Radiol medDOI 10.1007/s11547-015-0578-1

EMERGENCY RADIOLOGY

Comparison between MRI and CEUS in the follow‑up of patients with blunt abdominal trauma managed conservatively

Vittorio Miele1 · Claudia Lucia Piccolo1 · Barbara Sessa1 · Margherita Trinci1 · Michele Galluzzo1

Received: 24 May 2015 / Accepted: 30 July 2015 © Italian Society of Medical Radiology 2015

delayed bleeding. The remaining 77 underwent the full fol-low-up protocol. Follow-up protocol included CEUS at 24 and 72 h and CEUS and MRI at 1 month after trauma; only MRI was performed until the complete resolution.Results CEUS at 24-h and at 72-h from trauma showed a very good correlation with onset CE-MDCT in lesions staging. With respect to onset CE-MDCT, CEUS did not identified 2 adrenal injuries and 2 lesions of urinary tract, an intrinsic limit of this technique. CEUS performed at 1 month did not show traumatic lesions in 49/77 of patients. In the remaining 28/77 cases, CEUS demonstrated reduc-tion of the size of the lesions ranging from 25 to 50 %. MRI performed at 1 month from trauma did not show traumatic injuries in 37/77 patients; it demonstrated persistence of organ lesion in 40/77 patients. Therefore, in 12/77 patients MRI performed at 1-month demonstrated the persistence of minimal or moderate organ injury, while CEUS was com-pletely negative. In addition, MRI allowed to enhance the persistence of adrenal lesions in 2 cases and the integrity of urinary tract in 2. In the remaining 28 patients, in which both CEUS and MRI showed disease persistence, MRI, however, allowed a better definition of injury extension with respect to CEUS, in terms of dimensions, edges, and morphological evolution.Discussion and conclusions MRI allowed to make a bet-ter assessment of injuries than CEUS, allowing also a tem-poral stage of lesions. Infact, there are different evolution stages corresponding to accurate imaging findings. To our knowledge, this is the first study that describes the evolu-tion of blood collection in parenchymal abdominal organs. Therefore, in patients who underwent abdominal traumatic injuries conservatively treated, the follow-up at 1 month can be made by MRI, due to its panoramicity and its high contrast resolution, which allow a better morphological and temporal trauma staging respect to the CEUS.

Abstract Introduction Over the past two decades, there has been a shift toward non-operative treatment of patients undergo-ing a solid organ injury, thus requiring an increasing num-ber of imaging studies to monitor the healing of lesions, which were performed by computed tomography (CT). In consideration of the use of ionizing radiation and contrast media, nowadays there is a trend toward the use contrast-enhanced ultrasound (CEUS) in the follow-up of blunt abdominal trauma. However CEUS has some limits, espe-cially in the assessments of small lesions and in the evalu-ation of urinary tract lesions and vascular complications. Magnetic resonance imaging (MRI) is a useful alternative, since its lack of use of ionizing radiation, its panoramicity, the possibility to avoid contrast media and the ability to properly evaluate even small lesions. The aim of this study is to evaluate the usefulness and the feasibility of MRI in the follow-up of patients with low-grade blunt abdominal trauma.Materials and methods We performed a retrospective review of a cohort including 270 consecutive patients with a history of blunt abdominal trauma; among them, 118 underwent a high-energy trauma, and 152 a low-energy trauma. 124 patients had findings of abdominal injuries at the contrast-enhanced multidetector CT (CE-MDCT), including 68 from the group of major trauma and 56 from the group of minor trauma. 39 patients were operated for incoming lesions. The remaining 85 patients were treated conservatively. Eight patients underwent surgery later for

* Vittorio Miele vmiele@sirm.org

1 Department of Emergency Radiology, S. Camillo Hospital, Circonvallazione Gianicolense, 87, 00152 Rome, Italy

Radiol med

1 3

Keywords Blunt abdominal trauma · Solid organ injury · Follow-up · Contrast-enhanced ultrasound (CEUS) · Multidetector computed tomography (MDCT) · Magnetic resonance imaging (MRI) · Emergency radiology

Introduction

Over the past two decades, there has been a shift toward non-operative treatment of patients undergoing a solid organ injury, thus requiring an increasing number of imag-ing studies to monitor the healing of lesions, performed by contrast-enhanced-computed tomography (CE-CT) since the very beginning [1]. However, in consideration of the high number of children involved in blunt abdominal trauma, the use of ionizing radiation and contrast media (potentially related to adverse reactions), and the great resources of contrast-enhanced ultrasonography (CEUS) with respect to CE-CT, nowadays there is a trend toward the use of the latter one in the follow-up of blunt abdominal trauma [2, 3].

Contrast-enhanced ultrasound has improved US capa-bility to detect and to better depict abdominal traumatic lesions. As many previous studies reported [4–6], CEUS, by using a second-generation contrast agent, can easily highlight the number of detected lesions, with respect to the US alone, enhancing some qualitative findings, such as lesion extension, margins, and its relationship with capsule and vessels.

CEUS demonstrated to be almost as sensitive as CE-CT in the detection of traumatic injuries in patients with low-energy isolated abdominal trauma, with levels of sensitivity and specificity up to 95 % [7, 8].

Its role seems to be really relevant in pediatric patients, as shown by Valentino et al. [9] who compared the sensi-tivity and specificity of US with those of CEUS using CT as the gold standard; in a cohort of 27 patients, CEUS depicted 13 of the 14 lesions in 12 patients with positive CT scans and no lesions in the patients with negative CT scan, demonstrating that CEUS was almost as accurate as CT in the recognition of solid organ injuries.

However CEUS has shown some limits, such as the fact that it is operator-dependent, has a low panoramicity, the small operating window reducing the visibility, and its low capability to give useful information about some compli-cations, such as abscesses, bilomas, lesions to the urinary tract and vascular complications, requiring the use of CE-CT [10–12].

For these reasons, we have advocated the use of mag-netic resonance imaging (MRI) as a possible alternative to CEUS in the follow-up of patients with low-grade iso-lated abdominal trauma. In fact, we think that all its well-known features, such as the lack of ionizing radiation, its

panoramicity, to be less operator-dependent, the possibil-ity to avoid contrast media and the prolonged monitoring can help to better evaluate the lesion site in a follow-up setting.

Therefore, the aim of this study is to evaluate the use-fulness and the feasibility of MRI in the follow-up of patients with low-grade blunt abdominal trauma managed conservatively.

Materials and methods

We performed a retrospective review of a cohort includ-ing 270 consecutive patients admitted to our Emergency Department between January 2012 and December 2014 (176 males, 94 females, age range 8–61 years, mean age 34 years), with a history of blunt abdominal trauma; among them, 118 underwent a high-energy trauma, and 152 a low-energy trauma.

The 118 patients with high-energy trauma underwent at first an E-FAST in the emergency room to evaluate the presence of hemo-pneumothorax and hemoperitoneum; after obtaining the hemodynamic stability (blood pressure >90 mmHg, heart rate <100 beats per minute, respiratory rate <20 breaths per minute) a total body contrast-enhanced multidetector CT (CE-MDCT) was performed.

The 152 patients with a low-energy abdominal trauma underwent, at first, an abdomen US and CEUS. Among these, those positive for traumatic injuries of the abdominal organs underwent a CE-MDCT to obtain a proper staging of the severity of the traumatic injury.

Of the total 270 patients, 124 had findings of abdominal injuries at the CE-MDCT, including 68 from the group of major trauma and 56 from the group of minor trauma.

Among them, 39 were operated for incoming lesions of liver, spleen, kidney, bowel and/or mesentery and for active bleeding (28 of 39 had high-energy trauma, 11 low-energy trauma). The remaining 85 patients were treated conservatively.

In our institution, in a non-operative setting, our proto-col requires CEUS at 24 and 72 h and a new revaluation at a middle-time by CEUS and MRI at 1 month from trauma. Any following examination is performed only by MRI, until the complete resolution.

Among the 85 patients treated conservatively, 8 under-went surgery for delayed bleeding or because it was chosen for a surgical exploration in presence of reducing values of hematocrit and hemoglobin. Therefore, these patients were eliminated from the study. The remaining 77 underwent the full follow-up protocol (among them, 27 were pediatric patients).

The informed consent was obtained by all the patients or from their relatives in case of minors.

Radiol med

1 3

Examination technique

CEUS

CEUS examinations were performed using a Siemens Acuson Sequoia 512 system (Siemens Medical Systems, Forchleim Germany), equipped with a curved-array 4 MHz multi-frequency probe, after an I.V. bolus injection of a second-generation blood-pool contrast agent (Sonovue, Bracco). It consists of stabilized aqueous suspension of sul-fur hexafluoride microbubbles with a phospholipid shell.

The Sequoia system has a contrast-pulse sequencing (CPS) software which is able to recognize the fundamental non-lin-ear response of the microbubbles; the low-mechanical index (MI 0.15–0.19) real-time tissue harmonic imaging (Cadence), instead, allows a real-time gray scale imaging [4, 5, 13].

In our setting, a total of 4.8 mL of Sonovue, divided into two 2.4 mL doses, were administered through a 18-gage needle in an antecubital vein, followed by 5–10 mL of saline solution. After the first bolus the right sided organs have been explored for 1–3 min. After the administration of the second bolus, the left sided organs were studied for 3–4 min.

All CEUS examinations were performed by radiologists with a high level of experience, with at least 5 years’ expe-rience in emergency radiology and specialized in trauma imaging.

MRI

MRI examinations were performed on a 1.5 T system (Magnetom Avanto, Siemens) equipped with high perform-ing gradients with phased-array coils.

All the patients were required to fast for 4–6 h before the MRI examination, in order to reduce artifacts due to bowel peristalsis.

The examinations were performed on a supine posi-tion and the images were acquired on multiple planes with breath-hold HASTE T2-weighted sequences (TR: 1500 ms; TE: 95 ms; ST: 6 mm; matrix: 384 × 156), T1 in phase and out-of-phase (TR: 120 ms; TE: 2.38 ms; ST: 6 mm; matrix: 384 × 164) and fat-sat VIBE T1-weighted images (TR: 4.75 ms; TE: 2.39 ms; ST: 3 mm; matrix: 352 × 154) before and after I.V. administration of paramagnetic con-trast media. In particular, the latter one was employed on 71/77 patients; the remaining 6 patients refused the I.V injection of contrast media.

Data collection

This study was focused on traumatic injuries of liver, spleen, adrenals, and kidneys; any pancreatic, mesenteric or bowel lesions were analyzed.

On CEUS follow-up examinations, we analyzed lesions extension in terms of dimensions, margins, location, and numbers. A healthy abdominal organ normally appears as homogeneous and hyperechoic in absence of distortions of the echogenicity, with a well definition of vascular struc-tures. The parenchymal changes caused by a trauma are hematomas, bruising, lacerations, bleeding, and artero-venous fistulas, which have different aspects. In fact, hema-tomas can appear as hypoechoic images without perfu-sion; lacerations appear as hypoechoic linear or branched lesions, oriented perpendicular to the organ surface, usually associated with capsular discontinuity; a bruising is usually depicted as an inhomogeneous hypoechoic area with ill-defined contours with a poor definition of vessels within it; subcapsular hematoma appear as lenticular non-enhancing area; an active bleeding is characterized by the spreading of contrast media within the peritoneal or retroperitoneal space.

CEUS at 72 h and at 1 month after trauma evaluated the same parameters. In particular, it was assessed the extent of traumatic injury, any change of the amount of intraperito-neal fluid and of subcapsular hematoma.

In the RMI at 1 month after trauma, we evaluated the size of injury, the extension to the capsule, its signal inten-sity related to temporal evolution (therefore the presence of granulation tissue and peripheral contrast enhance-ment), the persistence of intra- and retro-peritoneal fluid, as well as the development of any complications as bilomas, delayed bleeding, and urinomas.

Results

CEUS at 24-h from trauma showed a very good correlation with CE-MDCT in lesions staging. It confirmed the pres-ence of organ injury in 77/77 patients: 38 hepatic lesions (49 %), 11 splenic injuries (14.3 %), 16 lesions of right kidney (20.7 %), and 22 lesions of left kidney (28.5 %). 16 patients showed double lesions, in particular 10 of them had simultaneous injuries of liver and right kidney and 6 had involvement of spleen and left kidney (Fig. 1a–d). 1 patient underwent simultaneous injuries of liver, right adrenal gland, and left kidney after a road injury (Fig. 2a–d).

With respect to CE-MDCT performed at the beginning, CEUS did not identified 2 adrenal injuries (1 of left and 1 of the right adrenal gland) (Fig. 2c, d) (Table 1). In 2 cases of kidney injuries CEUS, although having properly iden-tified the lesion of renal parenchyma, did not identified lesions of urinary tract, an intrinsic limit of this technique [5, 13] (Fig. 3a–c).

CEUS at 72-h from trauma confirmed the same find-ings and it researched for any dimensional extention of

Radiol med

1 3

Fig. 1 A 61-year-old patient who sustained a car accident. CE-CT (a, b) performed on axial plane shows a small splenic laceration and a hypodense area of the inferior pole of left kidney. CEUS at 24-h (c, d) confirms CT findings. CEUS performed at 1 month from trauma

(e, f) is normal at all. MRI VIBE fat-sat T1-weighted sequences (g, h) performed after administration of contrast media on axial plane show a complete restoration of splenic lesion and a very small remaining hypointense area of the inferior pole of left kidney (white arrow)

Fig. 2 Patient admitted to our hospital after a road injury. CE-CT on axial (a) and coronal (b) plane shows a small hepatic bruising local-ized in the VI segment, a laceration of the upper pole of left kidney and an adrenal hematoma. CEUS performed at 24-h (c, d) confirmed the hepatic and kidney lesions but doesn’t see the adrenal one. MRI performed at 1 month using the out-of-phase (e), HASTE fat-sat T2

-weighted (f) and VIBE fat-sat T1-weighted (g) sequence after gad-olinium well depicts the hepatic bruising (white arrow), the dimen-sional reduction of renal lesion and, above all, the adrenal hematoma (black arrow), not shown by CEUS, allowing a better definition of disease extent

Radiol med

1 3

trauma, of intra-peritoneal fluid and of subcapsular hema-toma, showing that it was not any dimensional increasing of injury nor of hematoma.

CEUS performed at 1 month did not show traumatic lesions in 49/77 of patients. In the remaining 28/77 cases, CEUS demonstrated reduction of the size of the lesions ranging from 25 % to 50 %.

Both at 72-h and at 1 month, CEUS was not able to give information about adrenal and urinary tract injuries, known on the CE-MDCT performed at the beginning.

MRI performed at 1 month from trauma did not show traumatic injuries in 37/77 patients; it demonstrated persis-tence of organ lesion in 40/77 patients. Therefore, in 12/77 patients MRI performed at 1-month demonstrated the per-sistence of minimal or moderate organ injury, while CEUS

was completely negative (Figs. 1g, h, 5g, h) (Table 2). In addition, MRI allowed to enhance the persistence of adre-nal lesions in 2 cases (Fig. 2e–g) and the integrity of urinary tract in 2 cases in which CEUS did not allow to acquire useful information owing to its intrinsic limit (Fig. 3g).

In the remaining 28 patients, in which both CEUS and MRI showed disease persistence, MRI, however, allowed a better definition of injury extension with respect to CEUS, in terms of dimensions, edges, and morphological evolution.

In fact, in 40/77 patients, MRI showed a thin hypointense edge (for its high content of hemosiderin) with a hypoin-tense center on T2-weighted sequences, a hyperintense edge with hypointense center on T1-weighted images owing to the high content of extracellular metahemoglobin, findings typical of a lesion in a late subacute-chronic stage (Figs. 4d–f, 5d–f, 6d–h, 7e–f). In the remaining 37 cases, MRI showed scars appearing as hypointense linear or branched streaks with capsular retraction (Figs. 4g–i, 7g–i, 8d–f).

In those cases in which it was observed a persistence of lesions at MRI performed 1 month after trauma, the follow-ing examinations were performed only by MRI.

All cases have been followed until the complete resolu-tion or until the comparison of scars.

Table 1 Number of lesions detected on CEUS and CE-CT at 24-h from trauma

CEUS CE-CT

Liver 38 38

Spleen 11 11

Adrenals 0 2

Kidneys 38 38

Fig. 3 A 8-year-old patient who sustained a sport trauma. CE-CT on axial (a) and coronal (b) plane demonstrates a right kidney fracture with wide perirenal fluid. CEUS at 24-h (c) confirms CE-CT findings showing a hypoechogen streak with perirenal fluid. CEUS performed at 1 month (d) is normal. MRI VIBE fat-sat T1-weighted sequences

after gadolinium on axial (e) and coronal (f) plane show complete restoration of renal pelvis with a soft hypoperfusion of the inferior pole (white arrows). MIP reconstruction (g) on coronal plane con-firms the latter finding

Radiol med

1 3

Discussion

To our knowledge, until now there is not a standardized protocol for the follow-up of blunt abdominal trauma con-servatively treated.

Whereas for the detection of parenchymal lesions, there is a wide range of imaging techniques available, such as CE-MDCT and CEUS, instead it’s not the same for the follow-up management, especially when pediatric patients are involved; in fact, in these cases, a systematic use of CT would imply an excessive exposure to ionizing radiation.

In our Department the protocol for the follow-up of blunt abdominal trauma conservatively treated includes a CEUS at 24 and 72 h after trauma, CEUS and MRI at 1 month and only MRI until the complete resolution or the demonstration of stabilized residual scarring. Only in case of clinical worsening a CE-MDCT is performed.

This approach is related to the fact that previous studies, such as that made by Miele et al. [5], showed the ability

Table 2 Number of lesions detected on CEUS and MR at 1 month from trauma

CEUS MR

Liver 12 15

Spleen 5 9

Adrenals 0 2

Kidneys 11 14

Fig. 4 A 40-year-old women admitted to hospital after a road injury. Axial (a) and coronal (b) contrast-enhanced CT show a wide lesion involving the VI and VII hepatic segment, with capsular involve-ment. It is also associated a conspicuous perihepatic hematoma; no evidence of active bleeding. CEUS performed at 24-h (c) confirms CT findings. The MRI out-of-phase sequence (d) performed on axial plane at 1 month from trauma shows dimensional reduction of the injury, appearing as an organizing hematoma because of a hyperin-

tense edge, expression of extracellular metahemoglobin content. The MRI images performed after administration of contrast media on axial (e) and coronal (f) plane don’t show any lesion enhance-ment. MRI performed after 1 year from injury, using the following sequences: out-of-phase (g), after contrast media on axial plane (h) and HASTE T2-weighted on coronal plane (i) shows the presence of a hypointense linear streak with capsular retraction (white arrows)

Radiol med

1 3

of CEUS to identify parenchymal lesions and the capsular involvement with an accuracy comparable to that of CT. Furthermore, as also demonstrated by Valentino et al. [9] and Menichini, et al. [12] CEUS demonstrated to be a very effective technique in the pediatric field, showing a high correlation with CT findings.

A further study by Sessa et al. [4] showed that CEUS can be used as a technique of first approach to identify and to stage traumatic abdominal injuries conservatively treated, compared with ultrasound and CE-MDCT as a gold standard. However, CEUS proved to be ineffective to recognize 4 cases of active bleeding, and 1 case of injury of the renal pelvis.

In another study conducted by Manetta et al. [3], CEUS was employed in the follow-up of hepatic and splenic inju-ries, showing an excellent correlation with CT as regards as

number, location, and extension of the lesions, highlight-ing thus its use at this stage instead of the CT, especially in young patients.

However, in many experiences, CEUS showed some lim-its because, besides being operator-dependent, it has a low panoramicity and does not provide important information about possible complications, such as the presence of bilo-mas, abscesses, urinary tract, and vascular lesions [10–12].

In our study, in fact, CEUS did not identified 2 adrenal lesions that were already known by the onset CT, and was not able to show the state of the urinary tract in 2 patients in whom it was previously recognized a lesion of the uri-nary tract.

MRI allowed to make a better assessment of injuries, because of the improved contrast and soft tissue resolution, allowing also a temporal stage of lesions, as several studies

Fig. 5 A 23-year-old patient sustained a trauma during a sport activ-ity. CE-CT on axial plane (a, b) shows a wide laceration of the left kidney with a conspicuous perirenal hematoma. CEUS at 24-h (c) confirms CT findings. MRI performed at 1 month (d–f) from injury using the out-of-phase (d) sequence demonstrates the perirenal

organizing hematoma with a hyperintense edge; after administration of gadolinium on axial (e) and coronal plane (f) there is no lesion enhancement. MRI performed at 3 months (g, h) from injury shows the almost complete recovery of the perirenal hematoma (white arrows)

Radiol med

1 3

showed [14–17]. Infact, the interpretation of traumatic lesions in MRI requires a huge knowledge of the differ-ent stages of blood in the various sequences, responsible of specific imaging findings. All these data are earned prima-rly from intracerebral hemorrhage Infact, as for the patho-physiology, four stages of hematoma evolution are known: hyperacute stage (0–3 h), acute stage (4 h–3 days), suba-cute stage (4 days–4 weeks), and chronic stage (>1 month).

At MRI, the signal of the hematoma changes in rela-tion with two parameters: the state of hemoglobin oxy-genation (which in turn influences the relaxation proper-ties and magnetic susceptibility) and the state of red blood cells membrane (intact or lysed). During the hyperacute, the acute and the early subacute stage red blood cells are

intact and the hemoglobin is transformed respectively in oxyhemoglobin, deoxyhemoglobin, and methemoglobin. Subsequently, cell membranes lyse and the methemoglobin becomes extracellular (late subacute stage). Finally, in the chronic stage, the wall of the hematoma contains mac-rophages rich in hemosiderin.

With SE sequences, in the hyperacute phase, the sig-nal of the hematoma appears isointense or hypointense on T1 and hyperintense on T2-weighted images (oxyhemo-globin) because of the diamagnetic effect of blood rich in water and proteins. In the acute stage, the presence of deoxyhemoglobin is responsible for a shortening of T2 relaxation times causing a signal reduction at T2-weighted sequences. Thus the hematoma will be isointense in T1 and

Fig. 6 A 28-year-old patient admitted to our hospital after a sport trauma. CE-CT on axial plane (a, b) shows a hypodense laceration of left kidney upper pole with a wide perirenal hematoma. CEUS at 24 h (c) confirmed CT findings. CEUS at 1 month from trauma (d) demonstrates lesion persistence without perirenal free fluid. MRI performed at 1 month (e, f) using the out-of phase sequence on axial plane (e) shows a renal hematoma with a hyperintense edge, owing to its metahemoglobin content; VIBE fat-sat T1-weighted sequence

(f) on axial plane doesn’t show any mass enhancement. MRI per-formed at 3 months after trauma with HASTE T2-weighted sequence (g) on axial plane shows dimensional reduction of injury, with a very hypointense rim, because of hemosiderin content, as observed in chronic lesions; out-of-phase sequence (h) on axial plane shows a hyperintense edge with a hypointense center, findings typical of organizing hematoma (white arrows)

Radiol med

1 3

will be hypointense at the center in T2-weighted images (deoxyhemoglobin) and hyperintense at the periphery (oxyhemoglobin). In the early subacute stage, the intra-cellular methemoglobin will be responsible for a short-ening of T2 relaxation times, causing a hyposignal in T2-weighted images and a hypersignal at the periphery in the T1-weighted images. During the late subacute phase, methemoglobin becomes extracellular and, being strongly paramagnetic, it will be responsible for a hyperintense sig-nal in both T1 and hypersignal in T2. In the chronic stage, the signal in T1 and T2-weighted images is highly variable, and may be hyper, hypo, or isointense.

MRI, compared with CE-CT, is also able to better depict subcapsular hematoma, which is an indicator of traumatic

injury, appearing as a hyperintense layer on T2-weighted sequences; it can also demonstrate vascular lesions, such as pseudoaneurysms, as a focular, nodular hyperintense lesion in arterial phase [17]; a splenic infarction, appearing as a hypointense image with a hyperintense edge without contrast enhancement; finally, it can easily demonstrate the presence of scars as hypointense linear or branched streaks with capsular retraction, as showed in our study.

To our knowledge, this is the first study that describes MRI aspects of the evolution of blood collection in paren-chymal organs. MRI is an excellent imaging technique for diagnosis, evaluation, and characterization of abdominal organ lesions, as liver, spleen, adrenals, and kidneys, which are parenchymal organs that we evaluate in this work.

Fig. 7 CE-CT on axial plane (a) shows a traumatic lesion of splenic upper pole with an artero-venous fistula. the patient underwent a hembolization procedure (b). CEUS performed at 24-h (c), after the hembolization, shows an avascular, triangular area in the splenic upper pole. CEUS performed at 1-month (d) demonstrates a triangu-lar area with a fluid component inside. MRI performed at 1 month using the out-of-phase sequence (e) and VIBE fat-sat T1-weighted (f) after contrast media, on axial plane, demonstrates the presence of

an organizing hematoma without enhancement. MRI performed at 1 year from injury with VIBE fat-sat T1-weighted (g), VIBE fat-sat T1-weighted after contrast media on axial (h) and coronal plane (i) show the resorption of ischemic area which appears really hypoin-tense because of superparamagnetic substances like hemosiderin and ferritin which cause a clear reduction of T1 and T2 relaxation times (white arrows)

Radiol med

1 3

During the phase of trauma staging, CEUS at 24-h showed a high accuracy, with a good correlation with the CE-MDCT. In fact, the use of contrast agents in ultrasound has greatly increased the detection of lesions with a better assessment of the findings, better defining the extent of the lesion, its margins, and its relations with the capsule and vessels.

CEUS at 1 month from trauma did not show the pres-ence of traumatic lesions in 49/77 cases; In the remaining 28/77 cases, CEUS showed a reduction of the size of the injury ranging from 25 to 50 %.

MRI, instead, showed persistent disease in 12 of these (Figs. 1g, h, 2e–g, 5g, h).

In 40 cases MRI showed a thin hypointense edge (for its high content of hemosiderin) with a hypointense center on T2-weighted sequences, a hyperintense edge with hypoin-tense center on T1-weighted images owing to the high con-tent of extracellular methemoglobin, findings typical of a lesion in a late subacute-chronic stage (Figs. 4d–f, 5d–f, 6d–h, 7e, f). In the remaining 37 cases, MRI showed scars appearing as hypointense linear or branched streaks with capsular retraction (Figs. 4g–i, 7g–i, 8d–f).

Another advantage is the possibility of MRI to study more accurately the urinary tract, not explorable by CEUS, since the contrast medium used is not metabolized from the

urinary tracts (Fig. 3e, f), as well as the ability to monitor 2 adrenal lesions (Fig. 2e–g).

However, MR has some limits, such as the need to have compliant patients able to follow breath-hold commands during images acquisition, which is not always simple with pediatric patients; the relatively long exam duration, related with the latter point; the impossibility to be performed in cases of pacemaker, cochlea implants, and patient’s claustrophobia.

Conclusions

Therefore, in patients who underwent abdominal traumatic injuries conservatively treated, the follow-up at 1 month can be made by MRI, since MRI, due to its panoramicity, and its high contrast resolution, allows a better morphological and temporal trauma staging [16] with respect to the CEUS, and allows to exclude all negative prognostic factors, such as late bleeding, rupture of the urinary tract, or additional injuries.

The use of MRI is even more important considering that many trauma patients are in childhood; for this reason the use of this technique in the follow-up, which does not use ionizing radiation, makes it extremely profitable compared to CT.

Fig. 8 A 8-year-old patient admitted to our hospital because of a fall from a bicycle. Axial (a) and coronal (b) contrast-enhanced CT performed at the beginning show a wide hepatic injury of VI and V segment involving the capsule. CEUS performed at 24-h from the trauma (c) demonstrates a large non-henancing area of the right lobe,

confirming the CT findings. MRI performed after 1 month from the injury before (d) and after contrast injection (e, f) using VIBE fat-sat T1-weighted sequences on axial plane, show a complete resorption of parenchymal injury with a linear non-henancing hypointense streak with capsular retraction (white arrows)

Radiol med

1 3

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflict of interest.

Ethical standards All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent Informed consent was obtained from all individ-ual participants included in the study.

References

1. Poletti PA, Wintermark M, Schnyder P, Becker CD (2002) Traumatic injuries: role of imaging in the management of the polytrauma victim (conservative expectation). Eur Radiol 12:969–978

2. Mizzi A, Shabani A, Watt A (2002) The role of follow-up imaging in paediatric blunt abdominal trauma. Clin Radiol 57:908–912

3. Manetta R, Pistoia ML, Bultrini C, Stavroulis E, Di Cesare E, Masciocchi C (2009) Ultrasound enhanced with sulphur-hex-afluoride-filled microbubbles agent (SonoVue) in the follow-up of mild liver and spleen trauma. Radiol Med 114:771–779

4. Sessa B, Trinci M, Ianniello S, Menichini G, Galluzzo M, Miele V (2015) Blunt abdominal trauma: role of contrast enhanced ultrasound (CEUS) in the detection and staging of abdominal traumatic lesions compared to US and CE-MDCT. Radiol Med 120:180–189

5. Miele V, Buffa V, Stasolla A, Regine G, Atzori M, Ialongo P, Adami L (2004) Contrast enhanced ultrasound with second generation contrast agent in traumatic liver lesions. Radiol Med 108:82–91

6. Regine G, Atzori M, Miele V, Buffa V, Galluzzo M, Luzietti M, Adami L (2007) Second-generation sonographic contrast agents in the evaluation of renal trauma. Radiol Med 112:581–587

7. Valentino M, Ansaloni L, Catena F, Pavlica P, Pinna AD, Barozzi L (2009) Contrast enhanced ultrasonography in blunt abdominal trauma: considerations after 5 years of experience. Radiol Med 114:1080–1093

8. Clevert DA, Wechbach S, Minaifar N, Clevert DA, Stickel M, Reiser M (2008) Contrast enhanced ultrasound versus MS-CT in blunt abdominal trauma. Clin Hemorheol Microcirc 39:155–169

9. Valentino M, Serra C, Pavlica P, Labate AM, Lima M, Baroncini S, Barozzi L (2008) Blunt abdominal trauma: diagnostic perfor-mance of contrast-enhanced US in children—initial experience. Radiology 246:903–909

10. Pinto F, Miele V, Scaglione M, Pinto A (2014) The use of con-trast-enhanced ultrasound in blunt abdominal trauma: advantages and limitations. Acta Radiol 55:776–784

11. Pinto F, Valentino M, Romanini L, Basilico R, Miele V (2015) The role of CEUS in the assessment of haemodinamically stable patients with blunt abdominal trauma. Radiol Med 120:3–11

12. Menichini G, Sessa B, Trinci M, Galluzzo M, Miele V (2015) Accuracy of contrast-enhanced ultrasound (CEUS) in the iden-tification and characterization of traumatic solid organ lesions in children: a retrospective comparison with baseline US and CE-MDCT. Radiol Med. doi:10.1007/s11547-015-0535-z

13. Valentino M, Serra C, Zironi G, De Luca C, Pavlica P, Barozzi L (2006) Blunt abdominal trauma: emergency contrast-enhanced sonography for detection of solid organ injuries. AJR 186:1361–1367

14. Hamon M, Leclerc X, Oppenheim C, Guavrit JY, Meder JF, Pruvo JP (2005) Neuroimaging characteristics of intracerebral haematomas. Rev Neurol 161:997–1006 French

15. Duhem R, Vinchon M, Tonnelle V, Soto-Ares G, Leclerc X (2006) Main temporal aspects of MRI signal of subdural hema-tomas and practical contribution to dating head injury. Neurochi-rurgie 52:93–104 French

16. Elsayes KM, Narra VR, Mukundan G, Lewis GS Jr, Menias CO, Heiken JP (2005) MR Imaging of the spleen: spectrum of abnor-malities. Radiographics 25:967–972

17. Gordic S, Alkadhi H, Simmen HP, Wanner G, Cadosh D (2014) Characterization of indeterminate spleen lesions in primary CT after blunt abdominal trauma: potential role of MRI imaging. Emerg Radiol 21:491–498