Autopsyproven untreated previous pulmonary ... - Wiley Online Library

17 downloads 8930 Views 337KB Size Report
Jan 28, 2011 - *Department of Legal Medicine, School of Medicine, Tokyo Women's Medical University, Tokyo; Tokyo Medical Examiner's Office, Tokyo.
Journal of Thrombosis and Haemostasis, 9: 922–927

DOI: 10.1111/j.1538-7836.2011.04225.x

ORIGINAL ARTICLE

Autopsy-proven untreated previous pulmonary thromboembolism: frequency and distribution in the pulmonary artery and correlation with patientsÕ clinical characteristics A . R O , *   N . K A G E Y A M A ,   T . T A N I F U J I   and M . S A K U M A à *Department of Legal Medicine, School of Medicine, Tokyo WomenÕs Medical University, Tokyo;  Tokyo Medical ExaminerÕs Office, Tokyo Metropolitan Government, Tokyo; and àDivision of Cardiology, Department of Internal Medicine, National Cerebral and Cardiovascular Center, Osaka, Japan

To cite this article: Ro A, Kageyama N, Tanifuji T, Sakuma M. Autopsy-proven untreated previous pulmonary thromboembolism: frequency and distribution in the pulmonary artery and correlation with patientsÕ clinical characteristics. J Thromb Haemost 2011; 9: 922–27.

Introduction Summary. Background: This study aimed to evaluate untreated, previous pulmonary thromboembolism (PE) in patients with acute fatal PE. Patients and Methods: We studied 64 patients diagnosed as having died from acute PE by medicolegal autopsy. Previous PE was histologically confirmed on the basis of organized thrombi (OT). The distributions of OT were analyzed in five different sizes of pulmonary artery branches in each of 18 pulmonary segmental arteries (90 in total). The frequency of OT in each patient was evaluated by determining the percentage of examined sections containing OT. Results: OT were confirmed in 59 of 64 (92%) patients. The mean frequency of OT per patient was 27% of the 90 branches. Among the segmental arteries, the right posterior basal lobe showed the highest frequency of OT; among the five artery branches examined, the subsegmental branch showed the highest frequency of OT. OT were not detected in arterioles. Patients with recent trauma or surgery and inpatients showed significantly lower frequencies of OT than those without these risk factors. The 26 patients with prolonged pre-existing symptoms lasting more than a day showed a higher frequency of OT than the 12 patients who suffered for less than a day and the 26 without preexisting symptoms. Conclusions: Most patients with acute fatal PE have a subclinical history of recurrent PE. The frequency of their untreated PE is suspected to correlate with specific risk factors for venous thromboembolism and their clinical course. Keywords: autopsy, pulmonary thromboembolism, risk factor, venous thromboembolism.

Correspondence: Ayako Ro, Department of Legal Medicine, School of Medicine, Tokyo WomenÕs Medical University, Tokyo 162-8666, Japan. Tel.: +81 3 5269 7300; fax: +81 3 5269 7300. E-mail: [email protected] Received 30 June 2010, accepted 28 January 2011

Acute pulmonary thromboembolism (PE) is a potentially fatal disease [1–5]. Dalen et al. [2] estimated that approximately 11% of PE patients in the United States died within 1 h of the event. He also stated that the rate of diagnosis of PE in patients having survived at least 1 hour is 26%; however, 92% of these patients received proper therapy with better prognosis. Ota et al. [5] also reported that 43% of the fatal cases of acute PE in Japan died within 1 h of their first clinical onset and that the mortality rate was significantly lower in the early diagnosed group (diagnosed within 24 h; 22%) than in the late diagnosed group (diagnosed after 24 h; 67.6%). Usually, patients experiencing fatal PE do not receive clinical diagnosis or treatment before death. Therefore, little is known about the natural history of acute PE before a lethal attack. In cases of acute PE, it might be considered that a patient suffered from the first and lethal attack based on their clinical features. However, latent existence of previous PE is sometimes detected at autopsy [6–9]. Organized thrombi in the pulmonary artery are pathological evidence of previous PE in autopsy cases. The existence of organized thrombi among sudden death cases due to acute PE without any clinical history of previous PE indicates an episode of untreated previous PE prior to the lethal PE. Dunnill [9] stated that careful autopsy examination of patients with acute PE usually reveals small organized thrombi. Morpurgo [7] reported that autopsy findings of multiple PE and pulmonary infarction are obtained in at least 15% of PE cases. This finding makes clinical verification difficult, because most cases of untreated previous PE are asymptomatic so that they do not receive clinical investigation at that time. Even if angiography could be performed after a lethal PE, already existing organized thrombi, which are usually located in the distal parts of the pulmonary artery, are difficult to detect because of the presence of proximal occlusive fresh thromboemboli. Therefore, autopsy confirmation of organized thrombi could be a clue for estimating the subclinical history of patients with recurrent PE.  2011 International Society on Thrombosis and Haemostasis

Previous untreated PE in cases of acute fatal PE 923

Previous autopsy studies have revealed the possible existence of previous PE in acute PE cases; however, the detailed distributions of previous PE in the pulmonary artery have not been examined. Furthermore, there are few autopsy reports of acute PE that deal with the histopathological characteristics and their correlation with clinical features [4,7,10]. In this report, we analyzed medico-legal autopsy cases of acute PE, focusing on the existence of untreated previous PE confirmed by histopathological methods. The frequency of previous PE in acute PE patients and distributions of previous PE in pulmonary arteries were analyzed. Additionally, the correlations between the patientsÕ frequency of previous PE and their clinical features were evaluated. Materials and methods Sixty-four autopsy cases of patients with acute PE were investigated in this study. The autopsy cases included 27 men and 37 women, showing a slight predilection for women (52.7 years old ± 17.1). The cause of death was ascertained based on results of medicolegal autopsies performed at the Tokyo Medical ExaminerÕs Office between February 1999 and September 2006. During our study period, 10 190 administrative autopsies were performed in our office and PE was detected in 347 cases. Routine lung examination was performed on 5–10 histopathological sections from each patient. Among them, 72 cases could be subjected to the described detailed histopathological lung examinations of 90 sections from each patient. Eight cases out of 72 were excluded from the analysis for the following reasons: five had been under clinical treatment for venous thromboembolism; one had a family history of PE;

A

one had systemic thrombotic disease; and the other suffered from severe pulmonary fibrosis. Acute fatal PE was defined as large fresh thromboemboli that occluded either the pulmonary arterial trunk or more than one main pulmonary artery, as confirmed at autopsy (Fig. 1A). Risk factors for venous thromboembolism (VTE) and each patientÕs clinical course were reviewed retrospectively based on autopsy reports. Medico-legal autopsy deals with cases of unexpected death, most of whom died without undergoing clinical consultations or receiving an exact diagnosis before death [3]. Therefore, most data regarding symptoms and clinical course are gathered in interviews with the bereaved family or police. The autopsy investigation of deep leg veins was performed as follows. The iliofemoral veins were opened with scissors for macroscopic examination. Then, a midline incision of the back side of the leg was made from the knee joint to the ankle. The popliteal vein and the calf vein were extracted with the calf muscles as one block for formalin fixation. Then, all veins were observed in serial 5-mm transverse sections for macroscopic examination. Three to five sections from each venous segment were taken for histopathological examination. This study was approved by the Institutional Review Board Committee of the Tokyo Medical ExaminerÕs Office (approval number: 21-4). Histopathological investigation

The lungs were removed at autopsy and fixed with formalin injected from the main bronchus. All pulmonary segmental arteries in 18 pulmonary lobes (10 from the right lung and eight from the left) were identified. Then, pulmonary segmental arteries were sectioned transversally in serial 5-mm sections

B

C

Fig. 1. A representative case. (A) Fresh thromboemboli occlude the main pulmonary artery. (B) Multiple organized thrombi were macroscopically observed in the peripheral part of the pulmonary segmental artery (arrows and arrowhead). (C) Adjacent histological section to that shown in panel B (arrowhead) stained by the elastica van Gieson method. Magnification · 1. Organized thrombi in the small elastic artery are observed as fibrous septa.  2011 International Society on Thrombosis and Haemostasis

924 A. Ro et al

Statistical analysis

Analyses were performed using SPSS 15.0 (SPSS Inc., Chicago, IL, USA). All continuous variables were analyzed by the Mann–Whitney test or Kruskal–Wallis test, and expressed as medians with interquartile ranges. Non-ordinal categorical data were analyzed using the chi-squared test. The distributions of organized thrombi in the pulmonary artery were assessed by multiple logistic regression using dummy variables and the results are presented as estimated odds ratios (ORs) with the corresponding 95% confidence intervals (CI). All tests of significance were two-tailed. Results Deep vein thrombosis of the leg was detected by autopsy in all except two cases. One of these two had a deep vein thrombosis of the arm, and a leg examination could not be performed on

the other. Cardiac hypertrophy, defined as a heart weighing > 400 g, was detected in 33 patients (52%) by autopsy investigations. Frequency of organized thrombi

Fifty-nine of 64 (92%) patients had at least one lesion of organized thrombi among the 90 examined sections. The mean frequency of OT in the 90 branches per patient (%OT) ranged from 0% to 82% (mean 27 ± 25%) (Fig. 2). The numbers of patients with organized thrombi in each of the five branches of different size were as follows: 47 of 64 (73%) in the segmental branches; 51 (80%) in the subsegmental branches; 54 (84%) in the tertiary branches; 50 (48%) in the small elastic arteries; and 31 (48%) in the small muscular arteries. Organized thrombi were not detected in any of the arterioles from the 64 patients examined. Distribution of organized thrombi in the pulmonary artery

The posterior basal lobe of the right lung showed the highest predominance of organized thrombi among the 18 pulmonary segmental arteries examined. The subsegmental branch showed the highest predominance of organized thrombi among the five sizes of pulmonary artery branches examined. Multivariate logistic regression analysis revealed that the anterior basal lobe of the right lung, the anteromedial basal lobe of the left lung, and the posterior basal lobe of the left lung, which are thick segmental arteries in the lower lobe, also showed high frequencies of organized thrombi. The analysis also revealed that the small muscular artery showed the lowest frequency among the five sizes of pulmonary artery branches (Table 1). Relationship between the frequency of organized thrombi and risk factors for VTE

Fifty-eight patients (91%) had at least one potential risk factor for VTE, as shown in Table 2. Psychiatric disease, obesity, and elderly and inpatient status were frequently observed. Obesity, 25

Number of the patients

from the proximal segmental branch to the peripheral branches for macroscopic observation (Fig. 1B). Five sections representing different sizes of pulmonary artery branches (segmental branch, subsegmental branch, tertiary branch, small elastic artery and small muscular artery), for each of the 18 pulmonary segmental arteries, were picked out for processing into paraffin blocks. The sizes of the branches were classified as follows. Segmental, subsegmental and tertiary branches were identified according to the pulmonary arterial bifurcations. Small elastic arteries, muscular arteries and arterioles were classified on the basis of histological characteristics according to previous definitions [6,9]. Briefly, small elastic arteries were defined as more peripheral lesions than tertiary branches with plentiful elastic tissue arranged in concentric parallel lamina in the media. Muscular arteries were defined as having a thin media defined by internal and external elastic lamina in the range 0.1–1 mm. Histological sections were cut and stained using the hematoxylin and eosin, elastica van Gieson and phosphotungstic acid hematoxylin methods. All cases had histopathologically confirmed fresh thromboemboli in their large pulmonary artery as the cause of their fatal acute PE. The frequency and distribution of organized thrombi were also examined in this study. Organized thrombi were determined in this report as eccentric collagen or elastin elements completely replaced by erythrocytes and fibrin, the main components of acute thromboemboli (Fig. 1C). Arterioles were defined as vessels of < 0.1 mm diameter and composed of endothelium and a single elastic lamina [9]. The incidence of organized thrombi in arterioles was also examined in the above-mentioned histological sections. To assess the distribution of organized thrombi in the pulmonary artery tree, the frequencies of organized thrombi in 18 pulmonary segmental arteries and in five different sizes of pulmonary artery branches were investigated. The frequency of organized thrombi in each patient was evaluated on the basis of %OT, which was defined as the percentage of examined sections (90; five sections in each of 18 pulmonary segmental arteries) with organized thrombi.

20

15

10

5

0

0–10 11–20 21–30 31–40 41–50 51–60 61–70 71–80 81–90 % OT

Fig. 2. The distribution of the frequencies of organized thrombi in 64 patients with acute massive PE. *%OT: percentage of organized thrombi in the 90 examined pulmonary branches per patient.  2011 International Society on Thrombosis and Haemostasis

Previous untreated PE in cases of acute fatal PE 925

tions, taxi drivers had a high %OT. In contrast, office workers (eight cases) and those in other specific occupations (13 cases) or without occupations (38 cases) showed no significant relationship between occupation and %OT. The occupations of the remaining two patients are unknown.

Table 1 Distribution of organized thrombi in the pulmonary artery Number OR of OT (%) (95% CI) Pulmonary segment (n = 320) Right pulmonary artery Apical lobe 78 (24.4) Posterior lobe 81 (25.3) Anterior lobe 95 (29.7) Lateral lobe 95 (29.7) Medial lobe 88 (27.5) Superior lobe 92 (28.8) Medial basal lobe 62 (19.4) Anterior basal lobe 107 (33.4) Lateral basal lobe 79 (24.7) Posterior basal lobe 136 (42.5) Left pulmonary artery Apicoposterior lobe 79 (24.7) Anterior lobe 76 (23.8) Superior lingular lobe 96 (30.0) Inferior lingular lobe 73 (22.8) Superior lobe 57 (17.8) Anteromedial basal lobe 105 (32.8) Lateral basal lobe 75 (23.4) Posterior basal lobe 109 (34.1) Pulmonary artery branches (n = 1152) Segmental branch 337 (29.3) Subsegmental branch 416 (36.1) Tertiary branch 360 (31.3) Small elastic artery 329 (28.6) Small muscular artery 141 (12.2)

P-value

0.42 0.44 0.55 0.56 0.51 0.53 0.31 0.67 0.44 1

(0.30–0.60) (0.32–0.63) (0.39–0.77) (0.40–0.78) (0.36–0.71) (0.38–0.74) (0.22–0.45) (0.48–0.93) (0.31–0.62)

< 0.0001 < 0.0001 0.0004 0.001 < 0.0001 < 0.0001 < 0.0001 0.016 < 0.0001 –

0.42 0.43 0.58 0.35 0.28 0.65 0.39 0.71

(0.30–0.60) (0.30–0.61) (0.42–0.81) (0.25–0.50) (0.19–0.41) (0.47–0.90) (0.27–0.55) (0.51–0.98)

< 0.0001 < 0.0001 0.001 < 0.0001 < 0.0001 0.01 < 0.0001 0.038

0.91 1.25 1 0.88 0.30

(0.76–1.09) (1.05–1.49)

0.29 0.01 – (0.73–1.05) 0.15 (0.24–0.37) < 0.0001

The right pulmonary artery in the right posterior basal lobe and the tertiary branch were chosen as baseline sites for OR calculation. OT, organized thrombi.

defined as a body mass index over 25, was detected in 19 patients (30%). Fifteen patients (23%) were considered elderly, defined as over 70 years of age. Recent trauma or surgery and inpatient status were all correlated with a low %OT. With regard to patientsÕ occupa-

Relationship between the frequency of organized thrombi and clinical course

Prodromal symptoms arising from PTE were detected in 59% (n = 38) of cases. Dyspnea (17 cases), syncope or vertigo (10 cases) and chest pain (six cases) were the most frequent complaints. Leg symptoms related to deep vein thrombosis were also detected in four cases. Twelve patients had a short duration of symptoms of < 1 day, while the other 26 cases had suffered for more than 1 day prior to death. Of the patients with prodromal symptoms, 21 patients consulted doctors. PE was suspected in four of 21 patients; however, none of them was given a definitive diagnosis or was under clinical treatment, because they all died soon after clinical inspection. The others were found dead or found to be in cardiopulmonary arrest upon hospital arrival without receiving any clinical treatment. The %OT was significantly correlated with the presence and duration of pre-existing symptoms. Patients with prolonged pre-existing symptoms of more than 1 day had a higher %OT than those with symptoms lasting < 1 day or those without pre-existing symptoms (Table 3). Discussion The present study revealed a significantly high frequency of organized thrombi in cases with acute fatal PE. About half of the patients showed a low frequency of organized thrombi with a %OT £ 10% (Fig. 2). Perhaps these patients were clinically

Table 2 Relationship between risk factors for VTE and the frequency of organized thrombi With factor

Male Psychiatric disease Obesity  Elderlyà Inpatient Recent trauma Prolonged immobilization Physical restraint Recent surgery Hyperlipidemia Taxi driver Obstetrics and gynecology disease Connective tissue disease Past history of malignant tumor Diabetes One or more potential factors§

Without factor

No. of cases (%)

%OT* (%)

No. of cases (%)

%OT* (%)

P-value

27 22 19 15 15 10 6 5 4 4 3 3 2 2 2 58

30.3 25.7 23.2 23.7 17.4 8.1 19.3 12.9 3.3 21.9 64.1 35.9 22.8 26.8 9.4 26.2

37 42 45 49 49 54 58 59 60 60 61 61 62 62 62 6

25.3 28.4 29.3 28.6 30.6 31.1 28.3 28.7 29.1 27.9 25.7 27.1 27.6 50.0 28.1 40.2

0.66 0.93 0.31 0.58 0.03 < 0.001 0.48 0.33 < 0.001 0.42 0.03 0.74 0.95 0.23 0.48 0.16

(42) (34) (30) (23) (23) (16) (9) (8) (6) (6) (5) (5) (3) (3) (3) (91)

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

27.9 23.0 23.0 24.0 25.0 13.4 22.3 9.8 5.2 30.6 18.0 30.7 24.4 25.5 7.1 25.4

(58) (66) (70) (77) (77) (84) (91) (92) (94) (94) (95) (95) (97) (97) (97) (9)

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

23.7 26.9 26.4 26.0 25.0 25.6 25.8 26.0 25.5 25.3 24.5 25.4 25.7 7.9 25.6 23.6

*%OT, percentage of organized thrombi in the 90 examined pulmonary branches per patient.  Body mass index > 25. àOver 70 years of age. §Potential factors include the above-mentioned 14 factors except gender. Data are expressed as medians (interquartile ranges).  2011 International Society on Thrombosis and Haemostasis

926 A. Ro et al Table 3 Patient profiles and pre-existing symptoms Pre-existing symptoms

Male/female Age (years) Heart weight (g) Body mass index %OT* (%)

No. (n = 26)

Less than 1 day (n = 12)

More than 1 day (n = 26)

P-value

13/13 54.0 (36.8–69.3) 378 (300–441) 22.5 (19.4–26.2) 5.0 (1.0–19.5)

3/9 50.5 (36.5–64.0) 371 (335–424) 25.1 (22.4–27.9) 16.0 (2.3–33.0)

11/15 55.5 (41.0–70.8) 362 (320–428) 23.0 (20.7–25.2) 40.5 (17.3–59.0)

0.34 0.73 0.98 0.12 0.001

*%OT, percentage of organized thrombi in the 90 examined pulmonary branches per patient. Continuous variables are expressed as medians (interquartile ranges).

considered as non-specific, because some autopsy investigations have revealed a not insignificant incidence of organized thrombi in patients that died of other diseases [11–13]. However, we consider that the existence of organized thrombi, especially in acute PE cases, suggests the previous occurrence of PE. Additionally, previous investigations have not evaluated the frequency of organized thrombi in the pulmonary artery. The frequency of organized thrombi ranged widely in this study (Fig. 2). We suspected that the frequency of organized thrombi might reflect the severity of recurrent PE. Fresh thromboemboli were more frequently found in the right lung than in the left lung, and more frequently found in the lower lobes than in the upper lobes. This is obviously related to the flow distribution, which favors the right lung and the lower lobes [14]. The distribution of organized thrombi in the present study was similar to that of fresh thromboemboli. The right pulmonary artery in the posterior basal lobe, which is the main branch of the lower lobe, showed the highest frequency of organized thrombi. Other dominant arteries located in the lower lobe, the right anterior basal lobe, the left anteromedial basal lobe, and the left posterior basal lobe, were also favorite sites of organized thrombi. However, organized thrombi were also detected less frequently in the upper and middle lobes. It is considered that, during a recurrent history of PE, second or third thromboemboli are brought to the upper or middle lobar artery because organized thrombi occluded in the lower lobes change the flow distribution from lower to upper lobes. There was a significant preference for the subsegmental branch inherent in the distribution pattern of organized thrombi. Thromboemboli are usually lodged and organized at the bifurcation of the elastic artery [9,11]. These previous studies indicated that the subsegmental branch is a preferred site for PE, both trapped and organized. They also showed comparable high frequencies of organized thrombi at the segmental tertiary branch, and small elastic arteries; however, the muscular artery showed a lower frequency. The muscular artery scarcely becomes a primary lesion of PE from deep vein thrombosis because of the difference in size between thromboemboli and the vessel [9]. Organized thrombi in the muscular artery are considered to arise as a consequence of fragmentation of previous PEs lodged in the proximal elastic artery [9,14]. By contrast, the muscular artery is known as a main lesion site in chronic thromboembolic pulmonary hypertension

(CTEPH) [15]. Organized thrombi in the muscular artery were detected in about 50% of cases with CTEPH [15]. Genetic variations are considered to have a relationship to CTEPH [16]; however, the detailed etiology of CTEPH and its relationship to acute PE is still unclear. It is estimated that 3.8–8.8% of patients with PE develop CTEPH [17–19]. The histopathological characteristics of muscular lesions among individuals with acute PE might provide a clue to their pathological relationship. VTE is recognized as a multicausal disease, involving environmental exposure, genetic factors and their interactions [20]. Many risk factors related to VTE have been discovered [21– 26]. The majority of patients in this study had at least one risk factor. Well-known environmental factors for acute PE, such as obesity, inpatient status and advanced age, were also frequently found in this study. Psychiatric disease is known to be a specific risk factor for VTE found by medico-legal autopsy [3,4,23,24]. This factor was highly detected among the cases in this study. The results of our study suggested that some risk factors are related to the frequency of organized thrombi. Clinical investigations have revealed that trauma, surgery and inpatient status are strong risk factors for acute PE [1,21,22]. In this study, the patients with these factors showed significantly lower frequencies of organized thrombi. This result suggests that sudden vessel wall injury caused by trauma or surgery and consecutive immobilization by hospital admission could lead to the formation of broad deep vein thrombosis resulting in lethal acute PE. By contrast, patients working as taxi drivers showed high frequencies of organized thrombi. This result suggests that their prolonged leg vein stasis is an environmental risk factor that leads to repeated, untreated PE. Therefore, the type of risk factor might affect the clinical features and severity of PE. These results need further verification by both clinical and pathological investigations, because the numbers of study subjects with each risk factor were small. The interactions between factors should also be considered. However, the correlation between the pathological findings of PE and clinical features has scarcely been analyzed in any previous study. Our results shed light on the correlation between clinical and pathological characteristics. We have to take into account the reliability of data gathered in interviews with the bereaved family or police, regarding the symptoms and clinical course. However, despite this possible limitation, more than half of our patients had pre-existing  2011 International Society on Thrombosis and Haemostasis

Previous untreated PE in cases of acute fatal PE 927

symptoms related to VTE. Their frequent complaints, including dyspnea, syncope and chest pain, were similar to those of patients with clinically-proven PE [22,25]. Leg symptoms related to deep vein thrombosis were also detected [26]. Onethird of the symptomatic patients suffered for < 1 day, which might reflect their fatal PE. However, most patients had preexisting symptoms lasting for more than 1 day prior to death, and these patients had a high frequency of organized thrombi. These clinical and pathological features indicated that a not insignificant proportion of patients with acute PE subclinically suffer from recurrent PE. Clinical investigation revealed that delayed diagnosis of PE was observed in 16% of cases, and that multiple symptoms at presentation and the presence of transient risk factors were significantly associated with earlier diagnosis [25]. Dunnill [9] proposed that autopsy-proven previous PE be termed ÔheraldÕ emboli. The results of our study emphasize that clinical awareness of such subclinical PE at an early stage is an effective way to prevent sudden death by acute PE. Conclusion Autopsy-proven previous untreated PE was detected in 92% of sudden death cases with acute PE. Their previous PEs predominantly occluded the thick segmental artery in the lower lobe and the subsegmental arterial branch. Inpatient status and recent trauma or surgery were significantly associated with a low frequency of organized thrombi, and taxi drivers had a high frequency of organized thrombi. The frequency of organized thrombi was high in patients with pre-existing symptoms from VTE. These correlations between clinical and histopathological characteristics provide suggestive information for physicians to predict acute PE at a preventable stage. Acknowledgement This work was supported in part by a Grant-in-Aid for Scientific Research for Young Scientists to A. Ro (17790413). Disclosure of Conflict of Interests The authors state that they have no conflict of interest. References 1 Giuntini C, Di Ricco G, Marini C, Melillo E, Palla A. Pulmonary embolism: epidemiology. Chest 1995; 107: 3–9s. 2 Dalen JE, Alpert JS. Natural history of pulmonary embolism. Prog Cardiovasc Dis 1975; 17: 259–70. 3 Murai T, Baba M, Ro A, Murai N, Matsuo Y, Takada A, Saito K. Sudden death due to cardiovascular disorders: a review of the studies on the medico-legal cases in Tokyo. Keio J Med 2001; 50: 175–81. 4 Ro A, Kageyama N, Tanifuji T, Fukunaga T. Pulmonary thromboembolism: overview and update from medicolegal aspects. Leg Med (Tokyo) 2008; 10: 57–71. 5 Ota M, Nakamura M, Yamada N, Yazu T, Ishikura K, Hiraoka N, Tanaka H, Fujioka H, Isaka N, Nakano T. Prognostic significance of early diagnosis in acute pulmonary thromboembolism with circulatory failure. Heart Vessels 2002; 17: 7–11.

 2011 International Society on Thrombosis and Haemostasis

6 SmithGT,DamminGJ,DexterL.Postmortem arteriographic studies of the human lung in pulmonary embolization. JAMA 1964; 188: 143–51. 7 Morpurgo M, Schmid C. Clinico-pathological correlations in pulmonary embolism: a posterior evaluation. Pulmonary Embolism (Progress in Respiration Research vol 13), 1st edn. Karger, Basel: Switzerland, 1980: 8–15. 8 Ro A, Kageyama N, Tanifuji T, Kobayashi M, Takada A, Saito K, Murai T. Histopathological study of pulmonary arteries in 14 autopsy cases with massive pulmonary thromboembolism. Leg Med (Tokyo) 2003; 5: s315–7. 9 Dunnill MS. Pulmonary Pathology, 2nd edn. Churchill, NY: Livingstone, 1987: 283–331. 10 Morpurgo M, Schmid C. The spectrum of pulmonary embolism. Clinicopathologic correlations. Chest 1995; 107: 18–20S. 11 Morrell MT, Dunnill MS. Fibrous bands in conducting pulmonary arteries. J Clin Pathol 1967; 20: 139–44. 12 Morrell MT, Dunnill MS. The post-mortem incidence of pulmonary embolism in a hospital population. Br J Surg 1968; 55: 347–52. 13 Freiman DG, Suyemoto J, Wessler S. Frequency of pulmonary thromboembolism in man. N Engl J Med 1965; 272: 1278–80. 14 Wagenvoort CA. Pathology of pulmonary thromboembolism. Chest 1995; 107: 10S–17S. 15 Moser KM, Bloor CM. Pulmonary vascular lesions occurring in patients with chronic major vessel thromboembolic pulmonary hypertension. Chest 1993; 103: 685–92. 16 Tanabe N, Kimura A, Amano S, Okada O, Kasahara Y, Tatsumi K, Takahashi M, Shibata H, Yasunami M, Kuriyama T. Association of clinical features with HLA in chronic pulmonary thromboembolism. Eur Respir J 2005; 25: 131–8. 17 Pengo V, Lensing AW, Prins MH, Marchiori A, Davidson BL, Tiozzo F, Albanese P, Biasiolo A, Pegoraro C, Iliceto S, Prandoni P; Thromboembolic Pulmonary Hypertension Study Group. Incidence of chronic thromboembolic pulmonary hypertension after pulmonary embolism. N Engl J Med 2004;350:2257–64. 18 Becattini C, Agnelli G, Pesavento R, Silingardi M, Poggio R, Taliani MR, Ageno W. Incidence of chronic thromboembolic pulmonary hypertension after a first episode of pulmonary embolism. Chest 2006; 130: 172–5. 19 Dentali F, Donadini M, Gianni M, Bertolini A, Squizzato A, Venco A, Ageno W. Incidence of chronic pulmonary hypertension in patients with previous pulmonary embolism. Thromb Res 2009; 124: 256–8. 20 Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999; 353: 1167–73. 21 Heit JA, OÕFallon WM, Petterson TM, Lohse CM, Silverstein MD, Mohr DN, Melton LJ 3rd. Relative impact of risk factors for deep vein thrombosis and pulmonary embolism. Arch Intern Med 2002; 10: 1245–8. 22 Sakuma M, Nakamura M, Nakanishi N, Miyahara Y, Tanabe N, Yamada N, Kuriyama T, Kunieda T, Sugimoto T, Nakano T, Shirato K. Clinical characteristics, diagnosis and management of patients with pulmonary thromboembolism who are not diagnosed in the acute phase and not classified as chronic thromboembolic pulmonary hypertension. Jpn Circ J 2005; 69: 1009–15. 23 Kamijo Y, Soma K, Nagai T, Kurihara K, Ohwada T. Acute massive pulmonary thromboembolism associated with risperidone and conventional phenothiazines. Jpn Circ J 2003; 67: 46–8. 24 Hamanaka S, Kamijo Y, Nagai T, Kurihara K, Tanaka K, Soma K, Miyaoka H. Massive pulmonary thromboembolism demonstrated at necropsy in Japanese psychiatric patients treated with neuroleptics including atypical antipsychotics. Jpn Circ J 2004; 68: 850–2. 25 Ageno W, Agnelli G, Imberti D, Moia M, Palareti G, Pistelli R, Rossi R, Verso M; MASTER Investigators. Factors associated with the timing of diagnosis of venous thromboembolism: results from the MASTER registry. Thromb Res 2008; 121: 751–6. 26 Stein PD, Gottschalk A, Saltzman HA, Terrin ML. Diagnosis of acute pulmonary embolism in the elderly. J Am Coll Cardiol 1991; 18: 1452–7.