Deep venous thrombosis among disaster shelter inhabitants following the March 2011 earthquake and tsunami in Japan: a descriptive study

Abstract

Objectives

A retrospective analysis of data collected during subject screening following Japan's March 2011 earthquake and tsunami was performed. We aimed to determine the incidence of deep venous thrombosis (DVT) among screened subjects and to identify risk factors associated with the development of DVT as independent variables.

Methods

Calf ultrasonography was undertaken in 269 subjects living in 21 shelters in Miyagi prefecture during the one-month period immediately following the March 2011 disaster. Information regarding the health and risk factors of subjects was collected by questionnaire and assessment of physical signs.

Results

Of the 269 evacuees screened, 65 (24%) met the criteria for calf DVT. We found lower limb trauma, reduced frequency of urination and sleeping in a vehicle to be independent positive predictors of DVT.

Conclusions

Evacuees had an increased risk of developing DVT, associated with tsunami-related lower limb injury, immobility and dehydration.

Keywords

deep vein thrombosis earthquake portable ultrasound machine tsunami

Introduction

The eastern part of Japan experienced a magnitude 9.0 earthquake followed by a tsunami on the afternoon of 11 March 2011. More than 15,000 people died from these twin events, and more than 2500 people remain missing.¹ Of those who survived, a significant number were displaced from their homes and began living in temporary shelters. In Miyagi prefecture, there were 1273 shelters with a maximum of 310,000 evacuees. These shelters were located in schools, gymnasiums and community centres, some of which were damaged and had inadequate facilities for the number of people they housed. Many shelters were overcrowded and lacked electricity, running water and adequate sanitation facilities. Shelter inhabitants did not have access to beds, and instead lay directly on the floor on thin mats (Figure 1). Private occupied areas ranged from 1.6 to 3.3 m², which is less than the 3.5 m² individual shelter space recommended by the Sphere Handbook,² one of the most widely known and internationally recognized sets of common principles and universal minimum standards in life-saving areas of humanitarian response.

Figure 1

Photograph of the living conditions of a typical Japanese shelter following the March 2011 earthquake and tsunami

Venous thromboembolism (VTE) originates as lower limb deep venous thrombosis (DVT) and causes pulmonary embolism (PE). VTE is a serious condition that has the potential to cause significant morbidity and mortality. Early identification and treatment of DVT are vital for the prevention of PE and death.³ The increased risk of sudden death due to PE among individuals living in emergency shelters has been reported following various disasters and emergency situations over the years and was documented as early as 1940 with a study on air-raid shelters. This early study found that of the autopsies performed on cases of sudden death reported to the coroner in various London districts, the number of cases of PE rose from 4 in September and October 1939 to 24 in the same two months of 1940, when London residents’ use of air-raid shelters in response to increased German bombardment of the area.⁴

More recently, following the Mid-Niigata Prefecture Earthquake of 2004, the occurrence of PE markedly increased to nine cases in the four weeks after the earthquake in the eight hospitals that treated most of the patients with acute cardiovascular disease in the disaster-affected area compared with the same hospitals’ treatment of only one case reported in the four weeks before the earthquake. In these hospitals, there were only two cases in the corresponding eight weeks in 2003 and one case in 2002. Six of the nine patients (67%) diagnosed with PE following the 2004 earthquake took refuge in their automobile before the onset of PE.⁵ Systematic ultrasonography of the calf was performed in all residents who took refuge in their cars one week after the earthquake, and 30% of them were found to have calf DVT.⁶ Local health authorities and emergency disaster control services have collaborated extensively on the development of DVT-specific screening and prevention protocols. A protocol to prevent VTE in disaster situations was proposed by the Niigata Medical Association and Niigata prefecture in 2006.⁷ This protocol aimed to identify high-risk evacuees living in disaster shelters and to provide DVT screening using portable ultrasound machines. This ultrasound screening was implemented in all high-risk cases, regardless of whether the evacuees presented with any DVT symptoms. People with calf DVT, which is associated with a low risk of PE, received compression stockings. People with DVT above the calf vein or who had bilateral fresh thrombi, which increase the risk of PE, were transferred to a hospital for immediate anticoagulation treatment. In 2008, another huge earthquake occurred in the northern part of Japan, and this protocol was applied to VTE prevention activity. There was no death caused by PE regardless of the high incidence of DVT (15%).⁸ However, there is an ongoing controversy about the screening of asymptomatic patients for DVT.^9–12 Some clinicians currently consider the use of ultrasound to assess venous thrombosis in asymptomatic patients to be unsatisfactory,^9,10 while others consider that ultrasound may be useful for asymptomatic patients at high risk.^11,12 A reduction in VTE-related morbidity and mortality could be achieved by either preventing the formation of DVT or by screening for DVT once formed, with initiation of treatment in selected cases to prevent progression, or a combination of both.¹²

In this study, we performed a retrospective analysis of data collected in Miyagi prefecture during the month immediately following the March 2011 earthquake. We aimed to evaluate the incidence of DVT among screened evacuees and identify factors associated with a high incidence of DVT under these conditions.

Methods

Study design

This was a descriptive study comprising a retrospective analysis of data collected during a mobile DVT screening programme following a major natural disaster. This programme was directed or approved by local emergency disaster control officials. The requirement for written informed consent was waived by the institutional review board of Miyagi Cardiovascular and Respiratory Center based on exemption category 45 CFR 46.101(b)(5).¹³

Setting

DVT screening was undertaken in disaster shelters in Miyagi prefecture for a period of one month starting the day after the earthquake and tsunami. Shelter selection was coordinated with local emergency disaster control officials based on perceived need, according to disaster screening protocols.⁷ A total of 24 DVT screening sessions were undertaken in 21 shelters over the one-month screening period (Figure 2). Each screening team that visited a shelter comprised 2–5 physicians (i.e. cardiologist, thoracic surgeon or neurologist) familiar with ultrasound for venous thrombosis in clinical practice and up to 10 technicians. Teams attempted to screen as many individuals as possible during each shelter visit, but the number of evacuees screened on site was limited by the fact that most shelters did not have electricity, which in turn caused screenings to be subject to limitations in the battery life of the ultrasound machines.

Figure 2

Flow diagram of subject selection. LEDCO, local emergency disaster control official; DVT(+), subjects who were diagnosed with DVT; DVT(–), subjects who did not have DVT

Study subjects

According to the Niigata Medical Association protocol,⁷ subjects staying at each shelter were selected for ultrasound screening if they were perceived to be at high risk for the development of DVT. Individuals known to have low activity levels (i.e. bed rest during the day in the shelter and sleeping in a car), lower limb trauma, lower limb pain or swelling, or a history of VTE were preferentially selected as being at high risk for DVT. In some shelters, public health nurses had selected high-risk subjects for screening in advance. Of a total of 9428 individuals living in the shelters visited by our medical team, 269 subjects were enrolled in the screening programme (Figure 2). Therefore, the correct overall incidence or prevalence cannot be determined.

Prior to ultrasound screening, all subjects completed a questionnaire to gather information about their age, gender and past history (including prior VTE) as well as about trauma, activities and symptoms following the disaster. Recorded patient variables included age, gender, current smoking (yes or no, binary), daily exercise (yes or no, binary), and any history of diabetes, hypertension or dyslipidaemia (yes or no, binary). The questionnaire also asked if the subject was currently taking warfarin for any reason, had used compression stockings since the disaster for any reason, had spent any nights in a vehicle since the disaster, had lower limb trauma, had lower limb symptoms (swelling or pain) or had experienced a reduced frequency of urination since the disaster. Acquired information was self-reported, not based on documents or observation. Health-care providers recorded any signs of lower limb trauma, varices or swelling.

The incidence of VTE increases markedly with advancing age.¹⁴ The age-adjusted rate of PE and DVT diagnoses were not different in either gender without a bias in PE and DVT diagnoses irrespective of gender.¹⁵ However, in women, the incidence of DVT by calendar year decreased among those aged younger than 55 years and increased among women aged older than 60 years.¹⁴ Earthquake-induced stress can promote a thrombogenic state.¹⁶ In 1995, the number of patients with acute myocardial infarction increased by about 3.5-fold during the first four weeks after the Hanshi-Awaji earthquake.¹⁷ Regarding the changes in cardiovascular risk factors after a major disaster, there are some conflicting results in studies.^17–19 Cardiovascular risk factors were also included in a questionnaire. Dehydration can cause increased blood viscosity. Even in shelters with adequate drinking water, dehydration can prevail. Under the circumstance of limited access to toilets, people tend to refrain from water intake to reduce urination. We noticed that one of the characteristics of tsunami survivors was injury of the lower extremities. Tsunamis carry drifting materials and pieces of broken glass. Survivors suffered bruises and/or cuts on their legs while escaping from tsunami-flooded areas.

Portable ultrasonography

Portable ultrasonography enables efficient screening of an at-risk population who might otherwise have unreliable or no access to medical care. The use of ultrasound machines running on battery power allowed for screening in areas without electricity, but limited the number of study subjects because each battery was only able to screen eight to 12 individuals. In some cases, it was possible to transport up to six machines to a shelter, allowing screening of a larger group. This method has previously been used to detect DVT in disaster shelters.⁸

Screening was undertaken using portable ultrasound machines (Toshiba Viamo, Tokyo, Japan, with a 7.5-MHz linear probe; Hitachi-Aloka Medical MyLabFive, Tokyo, Japan, with a 4–9 MHz linear probe; and SonoSite MicroMaxx, Bothell, WA, USA, with a 5- to 10-MHz linear probe). Ultrasound examination of the calf veins to the popliteal vein was performed in the sitting or supine position, depending on the health of the subject. Because of the suboptimal environment, which was cold and did not allow for adequate security of privacy and space, the proximal leg vein ultrasound scan was limited to subjects with proximal extension (Figure 3). DVT was diagnosed if there were direct ultrasound signs, including thrombotic echoes or a lack of venous compressibility, according to DVT guidelines published by the Terminology and Diagnostic Criteria Committee, Japan Society of Ultrasonics in Medicine.²⁰ Compression ultrasonography of the lower extremities for screening DVT is an established method. The accuracy of sonography in the evaluation of calf DVT among postoperative and symptomatic patients is reported as follows: the sensitivity specificity negative predictive value and positive predictive value were 92%, 100%, 98% and 100%, respectively, in the postoperative group and 86%, 96%, 94% and 90%, respectively, in the symptomatic group.²¹ Regarding portable sonography, there are no data estimating the accuracy of detecting calf DVT. However, for proximal DVT, the sensitivity and specificity of portable sonography in the emergency department were 100% and 99%, respectively.²²

Figure 3

Scene of ultrasound examination

Positive findings were assessed by at least two team members, including at least one physician. All subjects with bilateral leg DVT, multiple calf vein involvement or a progressive thrombus above the popliteal vein were transferred to the hospital or recommended to undergo a hospital visit for anticoagulation therapy and further examination to screen for PE (Figure 2).²³ All other subjects with positive DVT findings received compression stockings after confirming the absence of contraindications, and subjects were followed up to the extent that the disaster conditions allowed. Some subjects were lost to follow-up because of the evacuation conditions and the inherent difficulties in tracking this post disaster population.

Statistical analysis

Continuous variables are reported as mean (standard deviation, SD), and categorical variables are reported as frequencies and percentages. Continuous variables were compared between the groups with and without DVT using Student's f-test, and categorical variables were compared using the chi-squared test. To identify variables associated with DVT risk, multivariate analysis was performed using logistic regression analysis with all variables employed and the presence of DVT as an independent variable. Forward stepwise selection was used to screen for significance among these variables and to calculate their P value and odds ratio (OR). A forced entry method was used to estimate the OR of all variables. All analyses were undertaken using SPSS software version 11 for Windows (SPSS Inc., Chicago, IL, USA). Statistical significance was defined as P < 0.05.

Results

Of the 269 subjects screened, 95 were men and 174 were women, with a mean age (SD) of 70.6 (12.7) years. Of a total of 9428 individuals staying in the shelters visited, DVT was diagnosed in 65 of the 269 subjects screened (24%), of whom 34 (52%, 34/65) were asymptomatic. However three subjects had a history of prior VTE and none of them were found to have DVT under anticoagulant therapy. The incidence of DVT was low during the first week following the disaster (1 of 35 subjects screened) and increased from the second week to the end of the study period (Table 1). DVT was detected in the soleal vein in 57 cases, in the popliteal vein in seven cases and in the femoral vein in one case, which also had a fresh thrombus in the bilateral soleal veins and was diagnosed with silent PE by enhanced CT. In 24 cases, the thrombus was fresh, and in 41 cases, the thrombus was organized.

Table 1

Deep vein thrombosis rates related to subjects’ time in shelters

Post-earthquake (weeks)	Number of subjects (n = 269)	DVT (n = 65)	Fresh thrombus (n = 24)	Organized thrombus (n = 41)
1	35	1	0	1
2	58	20	12	8
3	98	20	5	15
4	25	5	1	4
5	53	19	6	13

DVT, number of subjects who were diagnosed with DVT; fresh thrombus, number of subjects with a fresh venous thrombus; organized thrombus, number of subjects with an organized venous thrombus

The baseline characteristics of subjects in the groups with and without DVT were comparable, with no significant differences in age, gender, smoking, regular exercise, co-morbidities, warfarin treatment or use of compression stockings (Table 2). The most common reason for taking warfarin and using compression stockings was the presence of atrial fibrillation and varices, respectively.

Table 2

Characteristics of subjects with or without deep vein thrombosis

	DVT
Variable	Presence [n = 65)	Absence (n = 204)	P value
Age (years)	70.6 (13.6)	70.6 (12.4)	0.221
Gender			0.224
Female	43 (25%)	131 (75%)
Male	22 (23%)	73 (77%)
Lower limb symptoms			0.286
Present	31 (27%)	82 (73%)
Absent	34 (22%)	1 22 (78%)
Lower limb trauma	12 (18%)	9 (4%)	<0.001^**
Reduced urination	34 (52%)	54 (26%)	<0.001^**
Sleeping in a vehicle	16 (25%)	27 (1 3%)	0.029^*
Daily exercise	20 (31%)	77 (38%)	0.308
Current smoking	6 (9%)	22 (11%)	0.279
Hypertension	38 (58%)	104 (51%)	0.293
Dyslipidaemia	17 (26%)	56 (27%)	0.162
Diabetes	14 (22%)	41 (20%)	0.198
Taking warfarin	4 (6%)	26 (1 3%)	0.142
Compression stockings	4 (6%)	10 (5%)	0.251

Data are mean (SD) or n (% of group with or without DVT)

P < 0.05

P < 0.01

Lower limb trauma (P < 0.01), reduced frequency of urination (P < 0.01) and having slept in a vehicle since the disaster (P < 0.05) were all significantly more common in the group with DVT than in the group without DVT.

Multivariate analysis showed that lower limb trauma (OR: 6.763; 95% confidence interval [CI]: 2.570–17.799; P < 0.001), reduced frequency of urination (OR: 3.362; 95% CI: 1.823–6.201; P < 0.001) and sleeping in a vehicle (OR: 2.131; 95% CI: 1.020–4.451; P = 0.044) were independent positive predictors of DVT (Table 3).

Table 3

Logistic regression analysis for risk factors associated with deep vein thrombosis

Variable	P value	Odds ratio (95% confidence interval)
Forward stepwise selection
Lower limb trauma	<0.001^**	6.763 (2.570–17.799)
Reduced urination	<0.001^**	3.362 (1.823–6.201)
Sleeping in a vehicle	0.044^*	2.131 (1.020–4.451)
Forced entry method
Age	0.855	0.998 (0.972–1.024)
Gender (female)	0.860	0.942 (0.485–1.830)
Lower limb symptoms	0.428	1.294 (0.684–2.449)
Lower limb trauma	<0.001^**	6.438 (2.345–17.675)
Reduced urination	<0.001^**	3.434 (1.804–6.540)
Sleeping in a vehicle	0.052	2.116 (0.995–4.500)
Daily exercise	0.803	0.919 (0.472–1.788)
Smoking habit	0.807	1.142 (0.394–3.304)
Hypertension	0.273	1.458 (0.743–2.862)
Dyslipidaemia	0.448	0.750 (0.357–1.576)
Diabetes	0.988	0.994 (0.463–2.135)
Taking warfarin	0.160	0.432 (0.134–1.394)
Compression stockings	0.575	1.490 (0.370–5.998)

P < 0.05

P < 0.01

Nine subjects (8 of the 65 diagnosed with DVT and 1 of the 204 without DVT) were followed up within the study period; one subject with newly onset DVT and four subjects with DVT that subsequently disappeared following compression stocking therapy were observed 5–14 days after their initial examination (Figure 2).

Discussion

Significant findings

Our study found a high incidence of DVT among evacuees staying in shelters after a major earthquake and tsunami. Lower limb trauma, reduced frequency of urination and sleeping in a vehicle were independently associated with an increased risk of DVT. The DVT incidence of 24% (65/269) is much higher than the annual VTE incidence of approximately 0.1% (117/100,000) and DVT incidence of 0.05% (48/100,000).¹⁴ If the remaining evacuees not enrolled in the target shelters had been diagnosed with an absence of DVT, the prevalence would be approximately 0.7% (65/9,428), which is still higher than the annual incidence of VTE or DVT.

The DVT incidence rate of 24% is much higher than the DVT incidence rate of 4.8% (9/187) among earthquake survivors (mean age = 28.3 ± 12.4 years) with spinal cord injury (SCI) in the 2005 Pakistan earthquake.²⁴ However, although SCI is a relatively frequent occurrence after earthquakes and is an established risk factor for developing DVT, our study population did not contain any subjects with SCI. We speculate that the main cause of this higher DVT incidence rate in our studied population compared with that of the Pakistan earthquake survivor population is the significant difference in the mean age of these two groups (i.e. 70.6 ± 13.6 years in the Japanese population versus 28.3 ± 12.4 years in the Pakistani population).

VTE risk

Evacuees staying in emergency shelters following the March 2011 earthquake and tsunami were exposed to many factors that raised their VTE risk. All three factors of Virchow's triad (i.e. venous injury, immobility and a hypercoagulable state) occurred with increased frequency within this population. Lower limb trauma occurred either at the time of the disaster or in the period soon after, during which evacuees sought alternative living arrangements.

Disaster shelters often exposed evacuees to sub-optimal living conditions, including a lack of electricity and inadequate water supply, toilet facilities and heating. In addition, some shelters had flooded basements. Shelters were crowded and cold and offered evacuees limited personal space, resulting in a decreased daily activity. Some evacuees, especially those who had pets, slept in cramped conditions in their vehicles. Because toilet facilities were sometimes outside and lacked lighting and heat, some evacuees tried to reduce their frequency of urination by decreasing their water intake and subsequently became dehydrated.

Earthquake-induced stress can also promote a thrombogenic state.¹⁶ Abnormal coagulation and fibrinolysis have been reported, indicated by an increase in haematocrit, fibrinogen, D-dimer and von Willebrand factor. In a high-stress group, tissue-type plasminogen activator antigen and plasmin-alpha₂-plasmin inhibitor complex levels were also increased.¹⁹

Our findings that lower limb trauma, decreased frequency of urination and immobility due to sleeping in a vehicle were independently associated with an increased incidence of DVT among the individuals screened are consistent with the three factors of Virchow's triad.

Locations of DVT

The soleal vein was the most common site of DVT detection in this study. The clinical consequences of asymptomatic distal DVT detected by routine screening are uncertain.^25,26 Instances of asymptomatic postoperative distal DVT following surgery for Achilles tendon rupture have a good prognosis and a favourable clinical outcome. Of 46 patients studied by Rosfors et al.²⁵ only two presented with the general clinical criteria for post-thrombotic syndrome at five years’ follow-up. However, among the crural veins, the soleal vein is especially important as an initial site of DVT resulting in massive PE.²⁶ The soleal muscle possesses few or no functional valves and acts as a storage vein. The soleal muscle functions only at the ankle joint and is not active during prolonged sitting, which can easily cause venous stagnation of the soleal vein.²⁷ Kageyama et al.²⁷ examined 120 legs of 60 autopsy cases with fatal PE and detected DVT in 117 legs, of which 115 contained soleal vein thrombi. Another autopsy series revealed that thrombi originating in the calf were common and that in cases with the thigh and calf thrombi in continuity, the thrombi in the calf were older than those in the thigh.²⁸

Calf DVT is more often associated with transient risk factors (e.g. recent surgery, recent plaster immobilization and recent travel), whereas proximal DVT is significantly associated with chronic states (e.g. active cancer, congestive heart failure, respiratory insufficiency and age of >75 years).²⁹ Transient risk factors applied to the majority of the shelter population in the present study, many of subjects were over 75 years of age, and some subjects had chronic illness.

Prevention of progression VTE at shelters

The current screening initiative found a high DVT incidence of 24% among subjects screened (approximately 0.7% of the total population of the shelters visited). These figures are considerably higher than the usual annual DVT incidence of approximately 0.05%.¹⁴ In our study, 52% of DVT cases were asymptomatic. Therefore, the screening programme was able to identify a significant number of individuals with VTE at an early stage and allowed for immediate initiation of treatment to prevent VTE progression. Scanning all evacuees is not realistic, but ultrasound examination raises evacuees’ awareness of VTE. This programme permitted simple, uncomplicated participation because it was undertaken at the subjects’ shelters.

Screening teams also undertook other VTE prevention measures at the shelters, such as education about VTE risks and how small changes in subjects’ living environments, such as moving their sleeping mats to create aisles and allowing easier access to toilets at night, had the potential to greatly reduce the incidence of VTE in shelter populations. In addition, physical therapist groups provided special physical prophylaxis for VTE at shelters when possible. Rathore and Farooq encouraged physical prophylaxis in the prevention of DVT in patients with SCI for whom it is not possible to provide adequate mechanical and pharmacological thromboprophylaxis or when there are inadequate numbers of care-givers and volunteers. Such prophylaxis is also cost-effective and can be carried out by almost anybody after a training demonstration by a physical therapist.³⁰ If permitted, compression stockings should be provided for all shelters. Some elderly evacuees have severe problems with their feet or mental status. In our study, pharmacological prophylaxis in shelters was not realistic because of the postdisaster conditions of poor sanitation, unreliable or challenging access to hospitals, and an unstable food supply. Patients with serious VTE were transferred to a hospital.

Our DVT detection rate increased five days after the disaster. To accurately assess all postdisaster incidents of DVT, a screening programme, such as the one described herein, should be started within one week of an earthquake or a comparable catastrophic event. Similarly, the local public health authorities should become involved in such preventive health activities as soon as possible post-disaster.

Limitations

Several limitations of this study should be noted. First, this was a retrospective analysis of data collected during a disaster response, and a lack of resources such as electricity, equipment and personnel prevented comprehensive screening of a clearly defined population group. Therefore, the sample was intentionally biased toward the higher-risk group because of the conditions under which the screening took place. In addition, this report only presented short-term follow-up data of a small population size. Finally, the clinical meaning of the asymptomatic subjects with DVT is yet to be determined.

However, our results are interesting because they show that there is a high incidence of DVT in shelters and that portable equipment can be used to screen and diagnose individuals at high risk. These findings support further evaluation and expansion of post-disaster screening and treatment programmes. Further research should be undertaken to more clearly define the specific population groups that could benefit from such screening.

Footnotes

Acknowledgements

No grant or other financial support was received for this study. We thank the following: the city of Kurihara, Japan, for supporting our activities since the 2008 earthquake by providing gasoline and accommodations; Miyagi prefecture for coordinating local headquarters for emergency disaster control; the Miyagi Cardiovascular and Respiratory Center for providing diesel fuel; Toshiba and Hitachi-Aloka for providing portable ultrasound machines; Alcare, Covidien, Dr Scholl's, Kendall and Kobayashi Medical for providing compression stockings; Niigata University Hospital, Tohoku University Hospital, Kohnan Hospital, Japanese Red Cross Ishinomaki Hospital and Kurihara Central Hospital volunteer healthcare providers; the Miyagi Cardiovascular and Respiratory Center for supporting our literature search; and Dr Marguerite Elgin Ms Emily Bass and Dr Angela Morben of Edanz Group Ltd. for assistance with manuscript preparation.

Conflicts of interest: The authors declare that they have no conflicts of interest.

References

National Police Agency. Damage situation and police countermeasures associated with 2011 Tohoku district - off the Pacific Ocean earthquake (cited 5 December 2012). See: http://www.npa.go.jp/archive/keibi/biki/index_e.htm

The Sphere Project 2011. The Sphere Handbook [Internet]. Northampton, UK: Belmont Press Ltd, 2011 (cited 5 December 2012). See: http://www.sphereproject.org/silo/files/the-sphere-handbook.zip

Galson

. Prevention of deep vein thrombosis and pulmonary embolism. Public Health Rep 2008; 123: 420–1.

Simpson

. Shelter deaths from pulmonary embolism. Lancet 1940; 2: 744

Watanabe

, Kodama

, Tanabe

. Impact of earthquakes on risk for pulmonary embolism. Int J Cardiol 2008; 129: 152–4.

Hanzawa

, Hayashi

, Fuse

, Aizawa

. Prevalence of calf deep vein thrombosis in residents in rural Japan, and diameter of soleus vein indicates a risk of calf DVT in Mid Niigata prefecture Earthquake 2004. Phlebolymphology 2007; 14: 149–50.

Fuse

, Aizawa

, Hayashi

, Hanzawa

. Guidelines for the diagnosis, treatment and prevention of pulmonary thromboembolism and deep vein thrombosis for the residents in the disaster area of Mid Niigata Prefecture Earthquake 2004. J Niigata Med Assoc 2006; 675: 2–12 (in Japanese)

Shibata

, Kikuta

, Sumiyoshi

. Team Kurihara - venous thromboembolism prevention activities. Shinzo 2010; 42: 473–80 (in Japanese)

Schellong

, Beyer

, Kakkar

. Ultrasound screening for asymptomatic deep vein thrombosis after major orthopaedic surgery: the VENUS study. J Thromb Haemost 2007; 5: 1431–7.

10.

Haut

, Schneider

, Patel

. Duplex ultrasound screening for deep vein thrombosis in asymptomatic trauma patients: a survey of individual trauma surgeon opinions and current trauma center practices. J Trauma 2011; 70: 27–34.

11.

Kawaguchi

, Kumabe

, Kanamori

. Early detection of venous thromboembolism in patients with neuroepithelial tumor: efficacy of screening with serum D-dimer measurements and Doppler ultrasonography J Neurooncol 2011; 101: 495–504.

12.

Kelly

, Rudd

, Lewis

, Hunt

. Screening for subclinical deep-vein thrombosis. QJM 2001; 94: 511–9.

13.

U.S. Department of Health and Human Services (DHHS). OPKR Guidance on 45 CFR 46.101(b)(5): Exemption for Research and Demonstration Projects on Public Benefit and Service Programs (cited 5 December 2012). See http://www.hhs.gov/ohrp/policy/exmpt-pb.html

14.

Silverstein

, Heit

, Mohr

, Petterson

, O'Fallon

, Melton

III . Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population-based study. Arch Intern Med 1998; 158: 585–93.

15.

Stein

, Hull

, Patel

. Venous thromboembolic disease: comparison of the diagnostic process in men and women. Arch Intern Med 2003; 163: 1689–94.

16.

Matsuo

, Suzuki

, Kodama

, Kario

. Hemostatic activation and cardiac events after the 1995 Hanshin-Awaji earthquake. Int J Hematol 1998; 67: 123–9.

17.

Suzuki

, Sakamoto

, Koide

. Hanshin-Awaji earthquake as a trigger for acute myocardial infarction. Am Heart J 1997; 134: 974–7.

18.

Trevisan

, Celentano

, Meucci

. Short-term effect of natural disasters on coronary heart disease risk factors. Arteriosclerosis 1986; 6: 491–4.

19.

Kario

, Matsuo

, Kobayashi

, Yamamoto

, Shimada

. Earthquake-induced potentiation of acute risk factors in hypertensive elderly patients: possible triggering of cardiovascular events after a major earthquake. J Am Coll Cardiol 1997; 29: 926–33.

20.

Tanaka

, Nishigami

, Taniguchi

. Members of The Subcommittee for Preparing Guidelines for Ultrasound Diagnosis of Venous Thrombosis of Lower Extremities of The Terminology and Diagnostic Criteria Committee, Japan Society of Ultrasonics in Medicine. Criteria for ultrasound diagnosis of deep venous thrombosis of lower extremities. J Med Ultrason 2008; 35: 33–6.

21.

Atri

, Herba

, Reinhold

. Accuracy of sonography in the evaluation of calf deep vein thrombosis in both postoperative surveillance and symptomatic patients. Am J Roentgenol 1996; 166: 1361–7.

22.

Crisp

, Lovato

, Jang

. Compression ultrasonography of the lower extremity with portable vascular ultrasonography can accurately detect deep venous thrombosis in the emergency department. Ann Emerg Med 2010; 56: 601–10.

23.

Masuda

, Kistner

, Musikasinthorn

, Liquido

, Geling

, He

. The controversy of managing calf vein thrombosis. J Vasc Surg 2012; 55: 550–61.

24.

Rathore

, Hanif

, New

, Butt

, Aasi

, Khan

. The prevalence of deep vein thrombosis in a cohort of patients with spinal cord injury following the Pakistan earthquake of October 2005. Spinal Cord 2008; 46: 523–6.

25.

Rosfors

, Persson

, Lärfars

, Lapidus

. A follow-up study of the fate of small asymptomatic deep venous thromboses. Thromb J 2010; 8: 4

26.

, Kageyama

, Tanifuji

, Fukunaga

. Pulmonary thromboembolism: overview and update from medicolegal aspects. Legal Med 2008; 10: 57–71.

27.

Kageyama

, Ro

, Tanifuji

, Fukunaga

. Significance of the soleal vein and its drainage veins in case of massive pulmonary thrombosis. Ann Vasc Dis 2008; 1: 35–9.

28.

Stein

, Evans

. An autopsy study of leg vein thrombosis. Circulation 1967; 35: 671–81.

29.

Galanaud

, Sevestre-Pietri

, Bosson

. OPTIMEV-SFMV Investigators. Comparative study on risk factors and early outcome of symptomatic distal versus proximal deep vein thrombosis: results from the OPTIMEV study. Thromb Haemost 2009; 102: 493–500.

30.

Rathore

, Farooq

. Physical therapy versus heparin for prevention of deep venous thrombosis. Spinal Cord 2010; 47: 769–72.