Diagnostic value of arterial blood gas lactate concentration in the different forms of mesenteric ischemia
- 320 Downloads
The role of serum lactate measurement in patients with intestinal ischemia still remains unclear. The aim of this study was to prospectively evaluate the diagnostic performance of arterial blood gas lactate concentrations in the patients with acute mesenteric ischemia and its different forms.
All the patients reporting abdominal pain associated with risk factors for mesenteric ischemia underwent arterial blood gas and contrast enhanced abdominal computer tomography (CT).
At CT, 201 patients (70.7%) showed a nonischemic disease (group 1) and 83 patients (29.2%) showed findings of mesenteric ischemia. Out of these, 35 patients (42.1%) showed bowel ischemia secondary to non vascular causes (group 2) and 48 (57.8%) had a vascular intestinal ischemia (group 3). Out of these, 20 showed small bowel arterial occlusion (group 3a), 13 a small bowel nonocclusive ischemia (group 3b), 7 a venous small bowel occlusion (group 3c) and 8 showed isolated colonic ischemia (group 3d). The median lactate serum level was significantly higher in patients with vascular ischemia if compared with patients with nonischemic disease and secondary mesenteric ischemia (p < 0.0001; Kruskal–Wallis test). The areas under ROC curves for the lactate serum levels in the groups 2, 3, 3a, 3b, 3c and 3d were, respectively, 0.61, 0.85, 0.93, 0.93, 0.68 and 0.67.
Arterial blood gas lactate levels seem to show good diagnostic accuracy in diagnosing small bowel arterial and nonocclusive ischemia and poor accuracy in diagnosing secondary mesenteric ischemia, small bowel venous ischemia and ischemic colitis.
KeywordsIntestinal infarction Bowel ischemia Mesenteric ischemia Ischemic colitis Lactate Arterial blood gas
Acute mesenteric ischemia
Arterial blood gas
It may affect the small and/or the colonic bowel and may arise from vascular origin (arterial embolism, arterial thrombosis, venous thrombosis and nonocclusive mesenteric ischemia) or be secondary to non vascular diseases such as intestinal strangulation, incarceration, volvulus and intussusception [3, 5, 6, 7].
Despite the recent advances in surgery, intensive therapy and diagnostic imaging, acute intestinal ischemia continues to be associated with delayed and missed diagnosis and high mortality and morbidity rates. This is related to the atypical, variable and nonspecific clinical symptoms, and lacking of high sensitive and specific diagnostic markers, especially in the early disease stage [3, 4, 8].
Concerning this, the ability of serum lactate levels in predicting acute intestinal ischemia was evaluated in several experimental and clinical studies, with different patients selection criteria, incongruent measurement methods and conflicting results [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25].
Moreover, according with literature, large-scale prospective studies on the alteration of lactate are widely lacking [8, 26, 27] and the role of serum lactate measurement in patients with intestinal ischemia still remains matter of debate.
The aim of this study was to prospectively evaluate the diagnostic performance of arterial blood gas (ABG) lactate concentrations in the patients with acute secondary mesenteric ischemia, vascular mesenteric ischemia and its different forms.
From May 2014 to May 2015, all the patients with acute abdominal pain referred to Emergency Department of the “A. Cardarelli” hospital, were enrolled in this study and inserted in a prospective database, including the results of clinical and instrumental evaluation and any subsequent conservative or surgical treatment.
As the first diagnostic approach, all the patients underwent anamnestic evaluation, vital signs assessment (temperature, respiratory rate, pulse, blood pressure, blood oxygen saturation), routine electrocardiogram and physical examination.
On the basis of this first evaluation, the subjects reporting acute abdominal pain associated with one or more risk factors were considered suspected of acute mesenteric ischemia (AMI). These last comprised risk factors for atherosclerosis (hypertension, smoking, diabetes, obesity, hypercholesterolemia, family history, advanced age), arrhythmias, hypovolemia, congestive heart failure, valvular diseases, previous mesenteric ischemia, previous myocardial ischemia or infarction, cardiomyopathy, ventricular aneurysm, inflammatory bowel diseases, heritable thrombophilias and intra-abdominal malignancy.
All the patients aged at least 18 years and presenting clinical suspicion of AMI were included in this study.
Exclusion criteria included shock, sepsis, need for cardiopulmonary resuscitation, hepatic and renal failure, diabetic ketoacidosis, diabetes under metformin treatment, such as trauma, epileptic attack, muscular exercise, alcohol abuse, and intestinal surgery within the previous 7 days. The patients with allergy to iodine contrast were also excluded.
All patients matching the selection criteria were examined with routine biochemical tests, arterial blood gas analysis and computer tomography (CT).
Any intraoperative data was recorded.
Arterial blood gas analysis
A 1 mL blood sample was obtained from the radial, brachial or femoral artery in each patient with a sterile, disposable, heparinized syringe, and measurements were immediately performed after sampling, using a automated, computerized analyzer (GEM® Premier 3600, Instrumentation Laboratory, Milan, Italy).
The following parameters were measured: partial O2 and CO2 pressures, bicarbonate, sodium, potassium, chlorine, calcium, lactate and hemoglobin concentrations, pH value, O2% saturation and bases excess.
According with our analyzer, the measured concentration of lactate was expressed in mmol/L and referred to the total serum level of lactates, inclusive of both stereoisomers l-lactate and d-lactate (reference range: 0.5–1.8 mmol/L).
Thoraco-abdominal computed tomography (CT) were obtained using a 64-detector row (64-r) configuration (VCT, General Electric Healthcare, Milwaukee, Wis, USA).
All patients underwent unenhanced and contrast-enhanced CT, in the arterial phase (using a bolus tracking technique), in the portal venous phase (55 s after the arterial phase) and late phase (120 s) after an intravenous injection of 100–120 mL of nonionic contrast material (Iomeron 400; Bracco Diagnostics, Milan, Italy) at a flow rate of 3 mL/s, followed by 40 mL of saline solution at a flow rate of 2 mL/s through an 18-gauge catheter placed into an antecubital vein using a power injector. Rectal air or rectal contrast material (cm) was not administered.
CT criteria for small bowel arterial ischemia were: filling defect in the lumen of the superior mesenteric artery and/or its branches, lacking or absent enhancement of the intestinal wall associated with bloodless mesentery and thin bowel wall with spastic reflex ileus in the early stage, hypotonic and paralytic ileus in a later stage. Congested mesentery and bowel wall thickening represented signs of subsequent reperfusion damage. Parietal pneumatosis was considered sign of irreversible bowel wall necrosis. The presence of free peritoneal fluid was considered an additional finding [4, 28].
CT criteria for small bowel venous ischemia were detection of thrombus in the lumen of superior mesenteric vein or its branches, mural thickening with target appearance of the involved small bowel segments and mesenteric engorgement associated with peritoneal fluid. The absence of mural enhancement, the presence of gas in the bowel wall, mesenteric and/or portal veins and the presence of associated pneumo or pneumoretroperitoneum, were considered late findings [3, 4, 29].
CT criteria for nonocclusive mesenteric ischemia (NOMI) were the presence of small bowel ischemic damage similar to those of the arterial ischemia, associated with feasible ischemic involvement of colonic bowel (thinned) and/or parenchymatous organs (liver, spleen, kidney), without the evidence of arterial or venous filling defect [3, 4, 30, 31].
CT criteria for ischemic colitis due to AMI thrombosis or embolism were filling defect of inferior mesenteric artery or its branches, pericolic fluid and colonic wall thickening related to the early reperfusion by collateral vessels originated from superior mesenteric artery. In case of failure of reperfusion (usually due to superior mesenteric artery atherosclerosis), the colonic wall appeared thinned. In the nonocclusive forms (NOMI), because of the simultaneous involvement of both superior and inferior mesenteric arteries, the colonic wall appeared thinned in an early stage and thickened in a subsequent stage, if the reperfusion occurred [4, 32, 33, 34].
CT criteria for secondary mesenteric ischemia were the evidence of mechanical bowel obstruction associated with free peritoneal fluid, congested mesentery and thickening of the involved bowel in an early stage and lacking or absent enhancement of the intestinal wall and parietal pneumatosis in a later stage [3, 35, 36].
CT examinations were evaluated in consensus by two radiologists with more than 20 years of experience in the emergency radiology.
Statistical analysis was carried out using the program InStat GraphPad Prism® five (San Diego, California, USA). Values are expressed as means with standard deviation (SD) in parenthesis or medians with range in parenthesis, according with their distribution. Continuous data were compared between each group using the Kruskal–Wallis with Dunn’s Multiple Comparison Test and Tukey’s test, when indicated.
The diagnostic performance of lactate serum value in the different patients groups with CT diagnosis of mesenteric ischemia was evaluated by plotting a receiver operating characteristic (ROC) curve and using, as control values, the lactate serum levels of patients with CT findings of nonischemic disease. The diagnostic value was determined by calculating the area under curve (AUC), sensitivity and specificity.
A probability value of less than 0.05 was considered significant.
Among all the subjects with acute abdominal pain observed during the study period, 284 patients satisfied the selection criteria, were included in the study and constituted the object of analysis.
These patients comprised 113 females (39.8%) and 171 males (60.2%) with a mean age of 56 (9.5). All these patients underwent both ABG and thoraco-abdominal CT.
CT diagnosis in patients with nonischemic disease
Simple bowel obstruction
Abdominal aorta aneurysm
CT diagnosis in patients with secondary mesenteric ischemia
Strangulated bowel obstruction
Among group 3 patients, 40 (83.3%) showed CT findings of small bowel ischemia. Out of these, 20 (50%) showed arterial occlusion (group 3a), 13 (32.5%) a nonocclusive ischemia (group 3b) and 7 (17.5%) a venous occlusion (group 3c). One (5%) group 3a patient and 3 (23%) group 3b patients showed CT findings of combined small and large bowel ischemia. Five (38.4%) group 3b patients showed an ischemic involvement of liver and/or spleen parenchyma.
Among group 3 patients, 8 (16.6%) showed isolated colonic ischemia (group 3d). Out of these, 3 patients showed arterial occlusion and 5 a nonocclusive ischemia. No one case of venous colonic ischemia was found.
Thirty-two patients (17 group 3a, 11 group 3b and 4 group 3c) underwent, after CT diagnosis, surgery and/or therapeutic interventional radiology.
Sixteen patients (3 group 3a, 2 group 3b, 3 group 3c and all the 8 patients with ischemic colitis) showed, at admission, neither clinical nor radiological signs of irreversible ischemia and, therefore, underwent conservative therapy with parental nutrition, intravenous fluids, antibiotics, anticoagulation therapy and close observation. Among the patients under conservative therapy, 3 subjects (1 from group 3a, 1 from group 3b and 1 from group 3d) showed, during hospitalization, worsening general condition and, consequently, underwent surgery.
In all the patients undergone operative treatment, the surgical and/or the angiographic findings confirmed the CT diagnosis.
Arterial blood gas lactate levels
The median lactate serum level in patients with any form of mesenteric ischemia (vascular plus non vascular ischemia) was 1.5 (0.7–5.2), showing a significant difference if compared with patients with nonischemic disease (p < 0.0001; Kruskal–Wallis test).
Mean lactate serum levels in patients with small bowel arterial (group 3a) and nonocclusive ischemia (group 3b) were significantly higher if compared with patients with small bowel venous ischemia (group 3c) and patients with isolate ischemic colitis (group 3d) (p < 0.0001; Tukey’s test). No significative differences were found between group 3a and 3b and between group 3c and 3d patients (p > 0.05; Tukey’s test).
The area under ROC curve for ABG lactate concentration in patients with any form of mesenteric ischemia (vascular plus non vascular ischemia) was 0.65.
The areas under ROC curves for the lactate serum levels in the groups 2 and 3 were, respectively, 0.61 and 0.85, showing poor accuracy in the diagnosis of secondary mesenteric ischemia and good accuracy in the diagnosis of vascular mesenteric ischemia.
The areas under ROC curves for the lactate serum levels in the groups 3a, 3b, 3c and 3d, were, respectively, 0.93, 0.93, 0.68 and 0.67, showing very good accuracy in the diagnosis of small bowel arterial and nonocclusive ischemia and poor accuracy in the diagnosis of small bowel venous ischemia and ischemic colitis.
On the base of ROC curves, the optimal cutoff value for the diagnosis of vascular mesenteric ischemia, arterial and nonocclusive small bowel ischemia was 2.050. A cutoff point of ≥2.050 combines 64% sensitivity with 90% specificity for vascular mesenteric ischemia, 90% sensitivity with 90% specificity for arterial small bowel ischemia and 84% sensitivity with 90% specificity for nonocclusive small bowel ischemia.
The rationale for measuring serum lactate as a diagnostic test in acute mesenteric ischemia relies on the assumption that the hypoperfusion of an organ with high oxygen demand such as the intestine, may amplify the intestinal cells anaerobic metabolism, resulting in increased lactate release from the gut into the portal vein. Therefore, if the amount of released lactate exceeds the hepatic conversion capacity in pyruvate, increased serum lactate level in the general circulation may be found.
However, despite serum lactate has been tested as indicator of mesenteric ischemia in many experimental and clinical studies [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26], this assumption is still far from proven and the role of plasma lactate concentration in the diagnosis of mesenteric ischemia remains unclear for several reasons.
First, the value of a biomarker for any disease arises not only from its sensitivity and specificity, also from its ability to allow an early diagnosis. Due to the atypical, not specific and ambiguous symptoms, mesenteric ischemia is rarely identified before irreversible damage occurs, making it difficult to assess the diagnostic performance of any serum marker in early disease stage.
Moreover, the value of lactate serum level as diagnostic marker of mesenteric ischemia depends on lactate metabolism, which in turn depends on adequate tissue perfusion, liver and kidney function. Therefore, an adequate evaluation of the serum lactate concentration in the patients with mesenteric ischemia should contemplate the exclusion of any condition potentially responsible, per se, of increased serum lactate level such as shock, sepsis, impaired liver and kidney function, toxin drugs, diabetic ketoacidosis, diabetes under metformin therapy. This, together with the low incidence of bowel ischemia, makes it difficult to observe a number of patients suitable for an adequate serum lactate level comparative evaluation; consequently, large-scale prospective studies with strict exclusion criteria are lacking. With the expectation of a higher specificity, some authors tried to evaluate the diagnostic value of only d-lactate, a stereoisomer solely originated from bacterial fermentation in the gut lumen, differently from the other stereoisomer, l-lactate, produced by all human body cells glycolysis [22, 23, 24, 25]. However, although d-lactate has been argued by some authors as a more specific marker of mesenteric ischemia, other authors reported increased d-lactate serum levels in other diseases associated with enhanced bacterial fermentation (short bowel syndrome, jejunoileal bypass, carbohydrate malabsorption, ingestion of large amounts of carbohydrates, enterocolitis) [37, 38, 39, 40]. Furthermore, two recent reviews concluded that comparative studies of d-lactate versus l-lactate are lacking and that the current evidence does not support d-lactate as a specific marker of acute mesenteric ischemia [8, 26].
The variable diagnostic performance of serum lactate in diagnosing mesenteric ischemia, as reported in literature, may also depends on the variable diagnostic criteria employed for the recognition of the mesenteric ischemic disease. This last may affect different bowel tracts, may recognize various pathophysiological causes (arterial embolism/thrombosis, venous thrombosis, relative hypoperfusion) and, alternatively, may be secondary to non vascular diseases (intestinal strangulation, incarceration, volvulus and intussusception).
Due to this variable pathophysiology, vague criteria for cases of bowel ischemia were reported in literature and, only recently, some authors compared the serum lactate levels in vascular mesenteric ischemia with those in secondary bowel ischemia, including, as diagnostic criteria, CT finding of small bowel dilation or wall thickening . However, the last findings are not specific of mesenteric ischemia and may also be indicative of inflammatory bowel diseases or infective enterocolitis. Enhanced CT findings are conditioned by the involved tract (some intestinal segments are more sensitive to ischemic injury) by the typology (varying according to the obstructive mechanisms) and by the timing [3, 4]. So, the bowel wall thickening, although early detectable by CT in venous mesenteric ischemia , may only represent a subsequent finding in arterial and nonocclusive ischemia. In these last cases the bowel wall may appear thin, with a typical “paper thin” aspect, in the early stage and thickened in a subsequent stage, if ischemia is followed by reperfusion [4, 28, 30, 31].
Finally, the serum lactate level may be influenced by the type of blood sample. Indeed, although good correlation between venous and arterial lactate values was found and there is good evidence that venous values are equally useful , the venous lactate levels tend to be higher than arterial approximately 0.2 mmol/L  and may be distorted by the prolonged and improper application of a tourniquet . So, the venous blood samples can result in some false positives and the arterial lactate values should still be considered the gold standard.
The aim of this study was to evaluate the value of lactate serum concentration in diagnosing mesenteric ischemia and its various forms.
With this intent, unlike previous studies, the ABG was used as sampling method, strict exclusion criteria for the patients selection were employed and detailed CT criteria for mesenteric ischemia diagnosis were included. To our knowledge, this is the first study evaluating the diagnostic performance of serum lactate in the different forms of vascular mesenteric ischemia.
Our data show, first, a significant difference in the median lactate serum level between patients with any form of mesenteric ischemia and those with nonischemic disease. However, the related ROC curve analysis showed a poor diagnostic accuracy of lactate and this is probably due to the wide variability in the pathogenesis of the questioned disease.
Indeed, mesenteric ischemia can be divided in the two main groups: vascular and secondary (non vascular). In this last group, as hypothesized by some authors , the mechanism of ischemia (closed-loop bowel obstruction) may determine a venous impaired outflow earlier than the arterial outflow, preventing a massive entry of lactate in the bloodstream, at least in an early stage. So, this peculiar mechanism of ischemia could explain the poor discriminatory ability of ABG lactate level between patients with secondary mesenteric ischemia and nonischemic disease, as confirmed by our data.
The vascular mesenteric ischemia, in turn, may be divided in, at least, 4 subgroups including arterial small bowel ischemia, venous small bowel ischemia, nonocclusive mesenteric ischemia and ischemic colitis.
The analysis of ABG lactate concentrations in the various forms of mesenteric ischemia showed a significant difference between median lactate level in patients with both small bowel arterial ischemia and nonocclusive mesenteric ischemia if compared with patients with both small bowel venous ischemia and exclusive ischemic colitis. Likewise, the ABG lactate level diagnostic accuracy was high in the first two groups and poor in the last two. This may be related to the mechanism of ischemia and to the anatomic characteristics of gut vascularization.
In the small bowel venous ischemia, such as in the secondary ischemia, the predominant venous impaired outflow could delay the lactate level increase in the general circulation at an early stage. Only in a later stage, if bowel necrosis occurs, large amount of lactate could reach the systemic circulation and be measurable, differently from the arterial and nonocclusive mesenteric ischemia that are characterized by early impaired arterial outflow allowing a more easy transfer of lactate from portal to systemic circulation. Unfortunately, in our series, only 7 patients showed venous small bowel ischemia and other researches with larger series are needed to confirm this hypothesis.
The colon is shielded from ischemia by a collateral blood supply via a system of arcades (Riolan’s arcade) connecting the two major arteries represented by the superior mesenteric artery (which supplies the ascending and transverse colon) and the inferior mesenteric artery (which supplies the descending and sigmoid colon). Therefore, if the collateral vessels system is pervious and well-functioning, the ischemic insult (generally due to diffuse small segmental vessels disease in elderly) may be promptly replaced by collateral vessels reperfusion. For these reasons, the prevalent clinical presentation of ischemic colitis is represented by the non gangrenous form that accounts for 80–85% of cases and is often characterized by mild and transient nature and reversible ischemia. Likewise, unless gangrene occurs, it is unlikely that a massive amount of lactate, produced by the ischemic colonic insult, may exceed the hepatic metabolization capacity and reach the systemic circulation. In support to this hypothesis, in our series, among the 8 patients with ischemic colitis, 7 (87%) were conservatively treated and only one (13%) needed laparotomy for worsening general condition.
Beyond these considerations, our findings seem to indicate that in patients with acute abdominal pain, risk factor for mesenteric ischemia and without cofactors influencing the serum lactate concentrations (e.g. diabetes, alcohol abuse, epilepsy, renal or hepatic failure), an ABG lactate level >2.050 is strongly suggestive of arterial (90% sensitivity and 90% specificity) or nonocclusive small bowel ischemia (84% sensitivity and 90% specificity) whereas lower values do not exclude the venous, the colonic and the secondary mesenteric ischemia. Therefore, the abdominal CT with i.v. contrast still remains the gold standard in the diagnosis of intestinal ischemia even though high ABG lactate values may increase the clinical suspicion in selected cases. From our data, a diagnostic management algorithm for selected patients referring to Emergency Department may result: in patients with abdominal pain, risk factor for mesenteric ischemia and high ABG lactate level, the suspicion of arterial or nonocclusive mesenteric ischemia is high, and an urgent abdominal CT with i.v. contrast is indicated to confirm the diagnosis. Instead, in presence of normal ABG lactate levels, abdominal ultrasonography and plain radiography may be indicated, as first line, to exclude other causes of abdominal pain whereas CT may be performed, as second line, in case of persistent abdominal pain and absence of radiographic or echo-graphic findings of not ischemic disease.
Another point of question is the ability of ABG lactate level to predict mesenteric ischemia in early disease stage. On the base of rationale for serum lactate measurement, the total amount of lactate in the general circulation should reflect the onset, the duration and the extension of bowel ischemia. Unfortunately, the onset and duration of symptoms does not necessarily correlate with the severity and the extension of intestinal ischemia that vary depending on the mechanism of ischemia (vessel occlusion, nonocclusive ischemia, venous occlusion), the number and the caliber of occluded vessels, the effectiveness of collateral vessels reperfusion, the patients comorbidity. This, together with the limited number of patients in our population study did not allow an adequate analysis of correlation between lactate serum level and severity such as duration of ischemia. Likewise, in our study 72.5% of patients with mesenteric ischemia underwent operative treatment (90% of patients with arterial small bowel ischemia, 92.5% of patients with nonocclusive ischemia, 42% of patients with venous small bowel ischemia and 12.5% of patients with ischemic colitis), suggesting that increased serum lactate level may represent mainly a late marker of irreversible mesenteric ischemia.
The weaknesses of this study were, first, the limited number of patients with venous small bowel ischemia and ischemic colitis and, second, the impossibility to evaluate the correlation between the duration and extension of ischemia and ABG lactate level. Furthermore, the strict inclusion criteria could not have allowed the evaluation of all the cases of mesenteric ischemia admitted to our emergency department.
In conclusion, the results of this study seem to indicate that ABG lactate level shows good diagnostic accuracy in diagnosing small bowel arterial and nonocclusive ischemia and poor accuracy in diagnosing secondary mesenteric ischemia, small bowel venous ischemia and ischemic colitis. This variability of diagnostic accuracy, probably related to the specific ischemic mechanism and to the gut vascularization properties, could be a reason of the contrasting results reported in literature. However, other researchers with larger series are needed to confirm these data.
AB and FI participated in the conception of study, carried out the statistical analysis and drafted the manuscript. AR, RN and LS participated in the acquisition data and helped in drafting the manuscript. MG, MC and MDP participated in the data acquisition and in the study design. LR, MS and RG contributed in data interpretation. AC, NDM and FP participated in study design and coordination and reviewed the final version of the manuscript.
Compliance with ethical standards
Conflict of interest
Antonio Brillantino, Francesca Iacobellis, Adolfo Renzi, Rodolfo Nasti, Lavinia Saldamarco, Maurizio Grillo, Luigia Romano, Maurizio Castriconi, Antonio Cittadini, Maurizio De Palma, Mariano Scaglione, Natale Di Martino, Roberto Grassi and Fiorella Paladino declare that they have no conflict of interest regarding the publication of this article.
Research involving human participants and/or animals
The study involves Human Participants; the ethical committee of our Institution approved the study protocol.
All the patients gave informed written consent.
- 30.Saba L, Berritto D, Iacobellis F, Scaglione M, Castaldo S, Cozzolino S, Mazzei MA, Di Mizio V, Grassi R. Acute arterial mesenteric ischemia and reperfusion: macroscopic and MRI findings, preliminary report. World J Gastroenterol. 2013;19(40):6825–33. doi: 10.3748/wjg.v19.i40.6825.CrossRefPubMedPubMedCentralGoogle Scholar
- 31.Mazzei MA, Guerrini S, Cioffi Squitieri N, Vindigni C, Imbriaco G, Gentili F, Berritto D, Mazzei FG, Grassi R, Volterrani L. Reperfusion in non-occlusive mesenteric ischaemia (NOMI): effectiveness of CT in an emergency setting. Br J Radiol. 1061;2016(89):20150956. doi: 10.1259/bjr.20150956.Google Scholar
- 34.Iacobellis F, Berritto D, Somma F, Cavaliere C, Corona M, Cozzolino S, Fulciniti F, Cappabianca S, Rotondo A, Grassi R. Magnetic resonance imaging: a new tool for diagnosis of acute ischemic colitis? World J Gastroenterol. 2012;18(13):1496–501. doi: 10.3748/wjg.v18.i13.1496.CrossRefPubMedPubMedCentralGoogle Scholar
- 35.Iacobellis F, Berritto D, Belfiore MP, Di Lanno I, Maiorino M, Saba L, Grassi R. Meaning of free intraperitoneal fluid in small-bowel obstruction: preliminary results using high-frequency microsonography in a rat model. J Ultrasound Med. 2014;33(5):887–93. doi: 10.7863/ultra.33.5.887.CrossRefPubMedGoogle Scholar