OUP user menu

Prognostic value of clinical and morphologic findings in short-term evolution of aortic intramural haematoma
Therapeutic implications

Arturo Evangelista, Rosa Dominguez, Carmen Sebastia, Armando Salas, Gaieta Permanyer-Miralda, Gustavo Avegliano, Zamira Gomez-Bosh, Teresa Gonzalez-Alujas, Herminio Garcia del Castillo, Jordi Soler-Soler
DOI: http://dx.doi.org/10.1016/j.ehj.2003.10.011 81-87 First published online: 1 January 2004

Abstract

Aims Intramural haematoma (IMH) forms part of the acute aortic syndrome presenting physiopathologic and evolutive patterns different from those of aortic dissection. The aim of this study was to determine the mortality and predictive factors of IMH in the first 3 months of evolution.

Methods and results Sixty-eight consecutive patients diagnosed of IMH (12 type A, 56 type B) were prospectively studied. Ten patients (eight type A, two type B) were surgically treated for clinical or haemodynamic evolution. Mortality rate was 19% (six type A and seven type B): five surgically treated (three type A, two type B) and eight medically treated (three type A, five type B). No relationship was observed between clinical variables and evolution. Maximum aortic diameter was greater in the group of patients who died (65.5±14.4mm vs 46.0±7.6mm; P<0.0001). Mortality rate in patients with aortic diameter >50mm was 50% (P<0.0001). Significant periaortic bleeding was mortality-related (47%; P<0.005). Multivariate analysis showed only a significant relationship between mortality and maximum aortic diameter >50mm (OR=11.33; P<0.005) and ascending aorta involvement (OR=11.18; P<0.05).

Conclusion Intramural haematoma mortality in the first 3 months of evolution is high (19%). Maximum aortic diameter >50mm and ascending aorta involvement are predictive of early mortality.

  • Aortic intramuralhaematoma
  • Aortic dissection
  • Prognosis
  • Magnetic resonanceimaging
  • Tomography
  • Transoesophageal echocardiography

1 Introduction

Aortic intramural haematoma (IMH) is an entity included in the acute aortic syndrome together with aortic dissection and penetrating aortic ulcer. Although it cannot be reliably distinguished from classic dissection on clinical grounds, advances in imaging techniques have significantly facilitated the diagnosis of IMH1–6which is considered to represent 10% to 30% of all cases of acute aortic syndromes.3–8Several studies3,4,7,9have demonstrated that bleeding in the media layer of the aorta evolves dynamically in the short term, and may be reabsorbed or progress to classical dissection or aortic rupture. Mortality in short-term evolution has beenrelated to location of the haematoma.3,4,7,9–11However, while mortality in descending aorta IMH is similar to that of type B dissection, results published in ascending aorta show great disparity, oscillating between 5% and 60%.3,10–16On the other hand, information available on mortality-related clinical and morphologic variables is scant. The aim of the present work was to determine the mortality and predictive factors of IMH in the first 3 months of evolution in a prospective and consecutive series of patients from a single hospital.

2 Methods

2.1 Patients

From January 1990 to December 2000, 302 patients were diagnosed at our centre, a general referral hospital, of acute non-traumatic aortic syndrome by at least two imaging techniques: contrast-enhanced computed tomography (CT), magnetic resonance imaging (MRI) or transoesophageal echocardiography (TOE) and/or anatomical findings at surgery or autopsy. The diagnosis of IMH was made in the presence of regional aortic wall thickness >7mm in circular or crescentic shape with evidence of intramural accumulation of blood3–5and in the absence of intimal disruption. Penetrating atherosclerotic ulcer was considered when an atherosclerotic plaque had a crater-like ulceration with jagged edges, outpouching from the aorta wall.17–19IMH associated with penetrating atherosclerotic ulcer images was not considered a reason for exclusion from the study. The location of IMH was classified according to the Stanford Classification. Cases with intimal flap were specifically excluded.

Clinical baseline findings included demographic data and cardiovascular risk factors. A patient was considered to have atherosclerosis whenever a history of ischaemic heart disease, peripheral artery disease or non-embolic cerebral vascular accident was present.

2.2 Imaging techniques

Transoesophageal echocardiography and MRI or CT were performed in all patients 48h after onset of symptoms. Echocardiographic examination was performed with a GE System Five or HP Sonos 1000 apparatus using a biplane probe up to 1995 and multiplane probe thereafter. Location, maximum thickening, circular or crescentic form and echolucency of IMH were assessed by TOE as reported previously;4,5,16,20the presence of aortic atherosclerosis was also considered.21Computed tomography was performed with Twin Flash (Elscint) and Mx 8000 Quad (Philips) scanners and intravenous boluses with 80 to 150ml of non-ionic contrast medium. Multislice study (MX 8000) begins with an unenhanced helix of thorax and superior abdomen reconstructed at 8mm of thickness and helicoidal study was performed from pulmonary apex to pubic symphisis with 1-mm slices after administration of contrast material. MRI was performed with a superconducting MR 1.5 T and 1.0T image (Magnetom Vision and Impac; Siemens). Transverse, oblique sagittal and coronal images were obtained with gated-ECG. T1-weighted spin-echo and HASTE sequences were performed in all studies. In selected cases, breath-hold gadolinium-enhanced rapid 3D MR angiographic technique was performed to assess flowing blood in thickened aortic wall.

Maximum diameter of the aorta segment with IMH was measured by CT or MRI using transverse planes. When the aorta was tortuous, maximum diameter was measured by MRI and the diameter most perpendicular to the aorta wall on coronal or oblique sagittal views was chosen. If MRI was counterindicated, the measurement obtained by helical CT was used. IMH was defined as extensive when the longitudinal extension was over 15cm by MRI or CT. The presence of periaortic haematoma was assessed in all cases with the presence of hyperdense collection in the mediastinum surrounding the thoracic aorta in unenhanced CT with Hounsfield units greater than 40,22or presence of hyperintense mediastinal collection on T1 sequences in MRI23as compared with mediastinal fat. Periaortic haematoma and more than mild pleural or pericardial effusion were considered as complications in all imaging tests. Significant periaortic bleeding was conventionally considered when periaortic haematoma and/or more than mild pleural or pericardial effusion were concomitant with a haematocrit value <35% or a decrease >5% during hospitalization in absence of other manifest sources of bleeding.

2.3 Management protocol

Patients diagnosed of IMH were admitted to the coronary unit and medical treatment to control blood pressure started with betablockers and/or other hypertensive drugs; controls with imaging techniques were made every 2–5 days depending on evolution until stabilization of the IMH without complications was verified. Surgical treatment was indicated in cases of poor evolution due to haemodynamic instability, persistent pain or rebleeding of the aorta wall determined by MRI.

Transoesophageal echocardiography and MRI or CT were performed in all patients prior to discharge and a clinical follow-up was made at 3 months.

2.4 Statistical analysis

All values are expressed as mean±SD. Univariate analysis was performed on all clinical and morphologic variables, with the chi-square test and Fisher's exact test used for categorical variables and the Student's t-test for continuous variables. Logistic regression models to identify independent predictors of mortality were constructed. Significance of the models was assessed using Likelihood Ratio chi square test between the model with constant and the full model, and significance of each coefficient was assessed using the Wald test. Estimates of risk (odd ratios) were calculated based on coefficients from the logistic models. Variables showing a relationship with the outcome at the significance level of <0.1 in crosstabulations for dichotomic variables and univariate logistic regression for continuous variables were included in the model. A value of P<0.05 was considered significant in all tests.

3 Results

Sixty-eight patients were diagnosed of IMH, 12 type A and 56 type B, according to previously-established criteria. Clinical characteristics of the patients are specified in Table 1. Maximum IMH thickness oscillated between 7 and 25mm (mean: 13.5±4.2mm). Atherosclerotic plaques in thoracic aorta were observed in 36 cases (53%), 10 of which had small penetrating atherosclerotic ulcers (15%). Maximum aortic diameter in the IMH site ranged from 34 to 100mm (mean: 49.8±11.9mm). Periaortic haematoma was observed in 18 patients (seven type A and 11 type B), pleural effusion in 21 (five type A and 16 type B) and pericardial effusion in eight (five type A and three type B); 17 of these patients had criteria of significant periaortic bleeding.

View this table:
Table 1

Clinical characteristics of patients with IMH (n: 68)

Age (years)67±9
Men/women60/8
Systemic hypertension55 (81%)
Smoking31 (46%)
Hyperlipidaemia18 (26%)
Diabetes10 (15%)
Atherosclerotic disease23 (34%)
Abdominal aortic aneurysm15 (22%)

3.1 Clinical evolution, mortality rate and morphologic changes in IMH

Three-month follow-up data were available in all cases. Six IMH type A and two type B were treated surgically during hospitalization owing to poor clinical or haemodynamic evolution. A further two patients with type A IMH underwent surgery after discharge within the first 3 months, one because of progression of type A dissection (Fig. 1) and the other for progression to fusiform aneurysm after rebleeding in the aorta wall. During the first 3 months of evolution, 13 patients died (19%; 95% confidence interval [CI], 11–30%), six type A (50%; 95% CI, 21–79%) and seven type B (13%; 95% CI, 5–24%). Eight patients died during acute phase, four treated surgically (two type A and two type B) and four treated medically (one type A and three type B) (Fig. 2). After the acute phase, five patients died: three with type A IMH died after the acute phase, two treated medically and the other surgically, and two type B treated medically(Table 2). Therefore, the mortality rate in type A IMH group treated medically was 75% (three of four cases) and in those treated surgically 37% (three of eight cases). Patients with type A who received medical treatment had a higher mortality rate than those in group B who received medical treatment (75% vs 9%, P<0.01).

Fig. 1

(A) Evolution of type A intramural haematoma (IMH). Transoesophageal echocardiography (TOE) showed an IMH (arrows) in ascending aorta wall. (B) After 1 month, type A aortic dissection was observed on TOE; arrows indicate intimal flap.

Fig. 2

Magnetic resonance imaging (MRI) showing an intramural haematoma (IMH) in ascending aorta (arrows). Maximum aortic diameter was 59mm and IMH thickening 16mm. The patient died after surgical treatment (patient 1).

View this table:
Table 2

Mortality data. Characteristics of patients and IMH complications

PatientsTypeAge yearsTreatmentInterval daysDiameter (mm)Periaortic bleeding
1A74medical159mediastinal
2A75medical3161mediastinal
pericardial
3A59surgical2170mediastinal
pericardial
pleural
4A74medical8052
5A63surgical860mediastinal
pleural
6A68surgical6061
7B49medical479pleural
8B73medical1061pleural
9B58medical2866mediastinal
pleural
10B74medical8042pleural
11B71medical766mediastinal
pleural
12B61surgical378mediastinal
pleural
13B67surgical464mediastinal
pleural

Transoesophageal echocardiography performed prior to discharge showed a decrease in IMH thickness in three patients (one type A, two type B) and the focal disruption of the intima in six IMH type B. No relationship between mortality rate and morphologic changes was observed.

3.2 Prognostic variables of IMH evolution

No significant relationship was found between clinical variables and outcome (mortality or mortality and surgical treatment). The relationship between imaging technique findings obtained in acute phase and mortality is shown in Table 3. No significant differences were observed in IMH thickness between mortality and survivor groups (13.2±4.0mm vs 13.3±5.1mm). Significant periaortic bleeding (Fig. 3) was mortality-related (eight of 17, 47%; P<0.005). Maximum aortic diameter was significantly greater in the group of patients who died than in the rest (65.5±14.4mm vs 46.0±7.6mm; P<0.00001); using univariate logistic regression analysis, OR was 1.18 (1.08–1.28, P<0.0001.). Twenty-four patients had a diameter >50mm; of these, six (25%) underwent surgery. Patients with maximum aortic diameter >50mm had a higher mortality rate than those with aortic diameter <50mm (12 of 24, 50% vs one of 44, 2%; P<0.001). Multivariate analysis showed only a significant effect on mortality of increased maximum aortic diameter (OR=1.16; 1.06–1.27; P<0.005) and ascending aorta involvement (OR=11.18; 1.40–89.25; P<0.05). The adjusted odds ratio for a maximum aortic diameter >50mm was 11.33 (1.87–68.57; P<0.005). Maximum aortic diameter in the group of deceased patients with type A IMH was 60.5±6.0mm and with type B IMH was 68.4±17.8mm.

View this table:
Table 3

Univariate analysis between IMH morphologic findings and 3-month mortality

Mortality (%)
Variablen (%)PresentaAbsentaP
Type A12 (18%)6 (50%)7 (13%)0.001
Diameter >50mm24 (35%)12 (50%)1 (2%)0.0001
Echolucent areas29 (43%)3 (10%)10 (26%)0.02
Circular IMH21 (31%)4 (18%)9 (19%)0.47
Extensive IMH47 (69%)9 (19%)4 (19%)0.29
Atherosclerotic ulcer10 (15%)0 (0%)13 (22%)0.06
Mediastinal haematoma18 (26%)8 (44%)5 (10%)0.02
Pleural effusion21 (31%)9 (43%)4 (9%)0.01
Pericardial effusion8 (12%)3 (38%)10 (17%)0.24
Significant periaortic bleeding17 (25%)8 (47%)5 (10%)0.002
  • a The columns labelled ‘present’ and ‘absent’ indicate the number of patients with and without the corresponding variable (in absolute and percentage terms) that died within the first 3 months.

Fig. 3

Evolution of type B intramural haematoma (IMH) by computed tomography (CT). (A) IMH type B (large arrow) with severe periaortic haemorrhage, pleural and mediastinal (small arrows); maximum aortic diameter was 43mm. (B) Complete reabsorption of IMH and absence of periaortic bleeding after 3 months of acute phase.

Similar results were obtained when the relationship between basal variables and the group of patients who died or required surgical treatment was considered. Multivariate analysis showed an independent association between this evolution and the presence of ascending aorta involvement (OR=53, 3–904; P<0.01) and increased maximum aortic diameter (OR=1.14; 1.10–1.27; P<0.01). The adjusted odds ratio for a maximum aortic diameter >50mm was 6.0 (1.1–32; P< 0.05).

4 Discussion

This prospective study represents the largest series of early evolution of IMH from one centre. The study shows IMH mortality in the first 3 months of evolution to be high (19%). Predictive IMH mortality factors are basically maximum aortic diameter and ascending aorta involvement. Mortality of patients with aortic diameter over 50mm was 50% and only 2% in those with diameter less than 50mm. Patients with ascending aorta involvement had 50% mortality and 12% when only the descending aorta or arch were involved.

In a meta-analysis12of 143 IMH cases reported before 2000, mortality was 21%, and mortality of type A IMH has been considered to be similar to that of aortic dissection. Recently, von Kodolitsch et al.24published a multicentre study of 66 IMH with 30-day mortality of 20%. In this study, a high incidence of type A IMH (38/66, 58%) was observed; in 84% of them surgery was indicated, and early mortality was 8% with swift surgery versus 55% without surgery. These results, like those of the present study, suggest that the short-term prognosis is serious in IMH involving the ascending aorta, and surgical repair should be considered. On the other hand, studies from Japanese and Korean groups13,25,26showed that medically-treated patients with IMH have low mortality regardless of whether they are type A or B. No definitive explanation is available for the considerable discrepancy among the results published; such a discrepancy could be related to differences in the population studied, co-morbidity, racial factors, early diagnosis on imaging techniques used or maximum aortic diameter.27–29

The relationship between maximum aortic diameter and mortality has not been established, although some studies have shown IMH reabsorption to be related to smaller aortic diameter.15,16,24In 22 patients with type A IMH, Kaji et al.30showed that the group in which IMH regressed had a significantly smaller aortic diameter than in the group in which IMH progressed or evolved to dissection or aortic rupture (47±3 vs 55±6mm), which suggested an optimum cutoff value of 50mm, identical to that used in the present study. Ide et al.31observed that evolution to classical dissection was only seen in IMH with aorta diameter >50mm. There is a paucity of information regarding the evolution of type B IMH and maximum aortic diameter. In a recent study, Sueyoshi et al.32reported that in the follow-up of 35 type B IMH maximum aortic diameter higher than 40mm predicts IMH progression. Although normal descending aorta diameter is smaller than that of ascending aorta, the lower risk of descending aorta rupture could justify 50mm being an adequate cutoff value in our study for type A and type B. A possible explanation for the prognostic value of maximum aortic diameter in short-term IMH evolution is that when intramural bleeding weakens the aorta wall, the greater stress on the wall of dilated aorta implies a greater risk of rupture than that of non-dilated aorta.

Several published series showed that periaortic bleeding is more frequent in IMH (21–38%) than in classical dissection.13,26,30Although this bleeding is not necessarily the equivalent of aortic rupture and most tend to reabsorb without complications,30the mortality in our patients with these complications was high, particularly in the subgroup of patients who presented anaemia on biochemical analysis (47%).

Ganaha et al.33suggested the presence of penetrating atherosclerotic ulcer in acute phase as the factor predicting IMH progression, with maximum aortic diameter lacking prognostic value. The high incidence of aortic ulcers (52%) in this retrospective series could be explained by the inclusion of IMH cases in subacute or chronic phase with ulcer-like images due to localised dissections during IMH evolution.34Thus, these localised dissections would constitute a complication rather than a prognostic variable in acute phase. In our series, only TOE showed small atherosclerotic ulcers in 15% of cases and 11% of patients presented localised dissections with ulcer-like images in the study conducted prior to discharge.

4.1 Limitations

The present study has several limitations: (a) normalized aortic diameter values with body surface area or age were not considered; (b) the number of cases with ascending aorta involvement was low given the greater prevalence of IMH in descending aorta; (c) type A mortality was the same whether treated surgically or medically; however, these results could have been influenced by the fact that only patients with complications were treated surgically; and (d) in our series, 35% of IMH were located in aorta dilated >50mm, which could imply worse prognosis than in other published series with smaller aortic diameter.

5 Conclusions

Mortality in the first 3 months of IMH evolution is high (19%). Our results suggest that not only IMH location but, even more so, maximum aortic diameter (>50mm) should be considered in the IMH prognosis. According toour data, patients with large aortic diameter might be better managed non-conservatively: those with ascending aorta IMH might benefit from surgery and in those with location in descending aorta endovascular treatment should be considered. Therefore, our findings might be helpful in the decision-making process regarding the most appropriate therapeutic management of these patients which should be validated in wider studies

Acknowledgments

The authors thank Aida Ribera for statistical analysis and Christine O'Hara for help with the English version of the manuscript.

References

View Abstract