SEARCH

SEARCH BY CITATION

Keywords:

  • medical patients;
  • risk prediction;
  • sepsis;
  • venous thromboembolism

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References

Background

Sepsis is prevalent in internal medicine (IM) departments. Elderly patients with sepsis and chronic medical conditions are at an increased risk for venous thromboemolism (VTE). The objective of this study was to assess the rate of VTE and the accuracy of the Padua Prediction Score (PPS) to predict VTE in patient with sepsis admitted to IM departments.

Methods

We prospectively collected data on septic patients admitted to IM departments in a community-based medical center. Additionally, we retrospectively collected VTE risk factors and events throughout a 1-year post hospitalization period. We computed the PPS for every patient, and analyzed the data accordingly.

Results

In total, 1080 patients were included in the study. The mean age was 74.68 ± 16.1 years. The average PPS was 4.86 ± 2.26, and 71.2% of the patients had a positive PPS. Only 17.8% of the patients received anticoagulant prophylaxis during their hospital stay. Seven patients had VTE on admission, 14 (1.29%) acquired in-hospital VTE, and 7 (0.65%) had VTE post discharge throughout 1 year. In all, 21.9% patients died during hospitalization, and the overall survival rate was 64%. PPS was not correlated with anticoagulant administration (P = 0.36), in-hospital VTE (P = 0.23) or 1-year VTE (P = 0.40), but was significantly associated with in-hospital death and survival (P < 0.0001).

Conclusion

The rate of VTE in medical patients with sepsis in IM departments is low, and PPS lacks granularity in detecting patients at risk of acquiring it. In this population, a positive PPS is highly associated with death, and may reflect a more general co-morbidity and disease severity index.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References

Hospitalized patients are at an increased risk for venous thromboemolism (VTE) [1]. About 25% of all cases of VTE are associated with hospitalization [2], and 50–75% of cases of VTE in hospitalized patients occur in those on the medical service [3-5]. In prospective studies of hospitalized patients at high risk who were not receiving prophylaxis, a deep vein thrombosis (DVT), whether symptomatic and asymptomatic, was found in 5.0–14.9% of patients, a proximal DVT in 2.0–4.9% of cases and a pulmonary embolism (PE) occurred in 0.3–1.5% of cases [6-8]. Among others, age, immobility and acute inflammatory states are considered major risk factors in developing VTE during hospitalization [9]. Several risk VTE assessment methods have been proposed for use in hospitalized medical patients [10-12]. Their limitations include a lack of prospective validation, applicability only to high-risk subgroups, inadequate follow-up time and excessive complexity [13]. The Padua Prediction Score (PPS) is a fairly simple score of 11 parameters which was devised from and verified on a cohort of general Internal Medicine patients. In their publication, Barbar et al. [14] suggest that patients with a cumulative PPS of 4 or above are at higher risk of VTE through 90 days, and thus should receive thromboprophylaxis while in the hospital. The PPS has recently been adopted by the American College of Chest Physicians (ACCP) as the preferred risk stratification tool in non-surgical patients [13].

Sepsis is a prevalent, serious and source consuming inflammatory medical condition. The incidence of sepsis has increased by 8.7% annually over the years between 1979 and 2000 [15]. Septicemia is the 10th leading cause of death in the United States [16]. Patients with sepsis are considered to be in a pro-thrombotic state, with clinical presentation ranging from localized thrombotic disease to disseminated intravascular coagulation [17]. As the population is getting older, more elderly patients with sepsis are expected to be cared for by internists in IM departments [18], and are likely to be assessed with VTE risk prediction methods. Moreover, the co-morbidities associated with contemporary hospitalizations [19], particularly in elderly patients admitted to IM departments with sepsis, probably yield a higher risk for VTE during hospital stay.

The aim of the current study was to prospectively assess the rate of symptomatic VTE during hospitalization and up to 1 year post discharge, in patients admitted to IM department with sepsis, and to assess the validity of the PPS in this patient population.

Patients and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References

The study was conducted in a 110-bed department of IM in a 450-bed community-based university affiliated hospital in Haifa, Israel, between 1 February 2008 and 30 April 2009. All enrolled patients were over 18 years old, and had a presumed diagnosis compatible with sepsis. No exclusion criteria were employed.

Data collection

The prospective collection of data through the electronic medical record system has been previously described [20]. In short, we developed a computerized database that was incorporated into our electronic medical record (EMR) system. The computerized system identified patients with presumed sepsis. Thereafter, the physicians were instructed to input pre-determined supplementary data via a mandatory questionnaire that included the structured input of data, alongside automatic data gathering. We prospectively collected demographic and clinical data, as well as rates of death, on all consecutive patients admitted with sepsis. For the purpose of this analysis, we collected additional data from patients’ charts, which included information on acute and chronic VTE risk factors, and rates of in-hospital and post-hospital 1-year VTE.

Assessment of clinical outcomes was performed blinded to the PPS score of the evaluated patients.

Definitions of terms

Sepsis was prospectively captured and defined according to the definition of sepsis provided by the ACCP/SCCM Consensus Conference in 1991 [21], that is any patient admitted with a suspected infection and at least two of the criteria of systemic inflammatory response syndrome (SIRS): (i) a temperature > 38 °C or < 36 °C; (ii) an elevated heart rate > 90 beats per minute; (iii) tachypnea, manifested by a respiratory rate > 20 breaths per minute or hyperventilation, as indicated by a PaCO2 of < 32 mmHg; and (iv) an alteration in the white blood cell count, such as a count > 12 000 per mm2, a count <4000 per mm2; or the presence of more than 10% immature neutrophils.

The VTE events were retrospectively evaluated. The diagnosis of a DVT necessitated a positive finding on Duplex ultrasound or the presence of a venous thrombosis in a computer tomography (CT), and the diagnosis of PE was based on a positive CT angiography (CTA) or a high-probability ventilation perfusion scan. Events were considered to be acquired prior to hospitalization if symptoms, imaging studies, physician orders and considerations as reflected in patients’ charts, were suggestive that the disease was present at admission, and if the diagnosis was performed within 48 h of admission.

The PPS was retrospectively calculated for every patient based on the presence of co-morbidities and clinical presentation. The presence of each medical condition granted cumulative points to the total PPS: the presence of active cancer, previous VTE, reduced mobility, a known thrombophilic condition (three points each); trauma and/or surgery within the last month (two points); and elderly age (> 70 years), heart failure, acute myocardial infarction or ischemic stroke, acute infection and/or rheumatologic disorder, obesity and ongoing hormonal treatment (one point each). All patients had a systemic infection. Reduced mobility was assumed in patients who were bed-ridden prior to hospitalization per history, or were there was no alert on the AVPU (Alert, responding to Voice, responding to Pain, Unconscious) scale on physical examination at presentation. Other conditions were collected from the patient's medical history and list of medications. An overall score of 4 or higher is considered to signify a high risk for VTE.

Follow-up

A 1-year follow-up was available for all patients alive through an innovative system of hospital-community on-line medical records (OFEK), in which diagnoses, prescriptions, laboratory, imaging and hospitalization records are captured [22].

Statistical methods

We described the characteristics of our study group by calculating mean, median, standard deviation and range, and by presenting the percentage of disease occurrences. A comparison of outcomes between patient groups was performed with the Pearson's chi-square and Fisher's exact methods. For the analysis of predictors for in-hospital death we used multivariate logistic regression. Specificity and sensitivity of the PPS to predict various outcomes were calculated using varying cut-off points from 0 to 1. For each decision probability, we recorded the 1-specificity vs. sensitivity on the receiver-operating characteristic (ROC) curve. The estimated AUC was computed by the trapezoid rule. All analyses were performed on IBM SPSS Statistics version 20 (SAS Institute Inc., Cary, NC, USA).

The study was approved by the Carmel Medical Center Institutional Review Board. The need for informed consent was waived.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References

We analyzed data from 1080 patients admitted to the division of internal medicine during a 15-month study period (in 2008–2009), who met the criteria of sepsis on their admission. Baseline characteristics and VTE-related conditions are detailed in Table 1. The mean age of our study group was 74.68 ± 16.15, ranging up to 102 years. Approximately 48% of the patients were female and 20.5% resided in long-term care facilities. Over half of the patients included had a reduced functional status prior to hospitalization, and 34.2% suffered from cognitive decline.

Table 1. Baseline clinical characteristics
  1. Unless otherwise specified numbers are given in n (%). VTE, venous thromboemolism.

General
Age
Mean ± SD74.68 ± 16.15
Range19–102
Median79.24
Female gender516 (47.8)
Residency at long-term care facilities221 (20.5)
Functional status
Normal510 (47.2)
Debilitated554 (50.3)
Unknown16 (1.5)
Cognitive status
Normal609 (56.4)
Dementia378 (34.2)
Unknown93 (9.4)
VTE related
Age > 70796 (73.7)
Confined to bed623 (57.7)
Active cancer180 (16.7)
Previous venous thromboembolism59 (5.5%)
CHF NYHA 3 or 4271 (25.1)
Infectious respiratory disease458 (42.4)
Obstructive respiratory disease198 (18.4)
Systemic infection1080 (100.0)
Obesity127 (11.8)
Arthritis73 (6. 8)
New stroke at presentation25 (2.3)
Operation within the last 30 days21 (1.9)
Known thrombophilia2 (0.2)
Varicose veins31 (3.1)
Inflammatory bowel disease6 (0.6)
Long drive1 (0.1)
Hormone therapy11 (1.0)
Padua Prediction Score4.86 ± 2.27
Positive Padua Prediction Score (≥ 4) 769 (71.2%)

Sources of infection were pneumonia in 42.4% (458/1080), urinary tract infection in 26.6% (287/1080), skin and soft tissue infection in 6.2% (67/1080), intra-abdominal infection in 1.5% (16/1080), unknown in 16.5% (178/1080) and specified as ‘other’ in 6.9% (74/1080) of the cases. On admission, 5.61% of our study cohort fulfilled the criteria of septic shock, and 10.50% met criteria of severe sepsis.

Our patient cohort bared multiple VTE-related risk factors. Seventy-three percent were over 70 years of age, 16.7% had active cancer, 5.5% had a previous VTE, 25.1% had advanced heart failure and 18.4% chronic obstructive pulmonary disease. Acute risk factors were also common, with all patients having a systemic infection, 42.4% suffering from an acute infectious respiratory disease and 57.7% confined to bed during the initial course of the hospitalization. We calculated the PPS for every patient. The average score was 4.86 ± 2.26, ranging from 1 to 12 (Fig. 1). Over 71% of the patients had a positive PPS (defined as ≥ 4).

image

Figure 1. Padua Prediction Score distribution (positive ≥ 4).

Download figure to PowerPoint

Two patients were diagnosed with pre-hospital acquired DVT and five with pre-hospital acquired PE, and were thus excluded from further analysis. An additional 14 patients (1.29%) were diagnosed with VTE during the course of their hospitalization. At 1-year follow-up, seven patients were diagnosed with VTE (overall 21 patient during or after hospitalization, 1.94%). The mean duration for VTE diagnosis during the hospital stay was 6.07 ± 0.83 days. Post discharge, the mean duration to VTE diagnosis was 124.57 ± 11.69 days. In a multivariate analysis of the effect of baseline VTE risk factors on the occurrence of VTE throughout 1 year, only varicose veins and known thrombophilia were significantly associated with the outcome (hazard ratios [HR] 4.25 (95% confidence interval [CI] 1.11–16.27) and 13.48 [95% CI 1.49–122.21]).

One hundred and ninety-two patients (17.8%) received anticoagulant prophylaxis during their hospital stay. No correlation between a positive PPS and anticoagulant administration was found (P = 0.36). Anticoagulant administration was not associated with a reduced in-hospital VTE (P = 0.30) or with reduced in-hospital mortality (P = 0.98). Anticoagulant administration did not affect the rate of clinical outcomes in patients with a negative or a positive PPS (Table 2).

Table 2. Clinical outcomes according to anticoagulant administration and PPS status
 Anticoagulant administration: Yes (n = 192)Anticoagulant administration: No (n = 888)P-value
  1. PPS, Padua Prediction Score; VTE, venous thromboembolism. Numbers are given in n (%).

In-hospital death42 (21.9)195 (22.0)0.98
Positive PPS41 (21.4)177 (19.9)0.56
Negative PPS1 (0.6)18 (2.0)0.34
In-hospital VTE4 (2.1)10 (1.1)0.30
Positive PPS4 (2.1)8 (0.9)0.25
Negative PPS0 (0)2 (0.2)0.60
One-year VTE5 (2.6)16 (1.8)0.47
Positive PPS5 (2.6)13 (1.5)0.41
Negative PPS0 (0)3 (0.3)0.52

A positive PPS was not associated with in-hospital or 1-year VTE (P = 0.23 and P = 0.40, respectively). This was not affected by anticoagulant administration. We computed the ROC curves to assess the accuracy of the PPS in predicting in-hospital and 1-year VTE (Fig. 2A,B). The areas under the curve (AUC) were 0.58 (95% CI 0.43–0.73) and 0.63 (95% CI 0.43–0.83). In patients without anti-coagulant administration, the AUCs for in-hospital and 1-year VTE were 0.54 (95% CI 0.37–0.71) and 0.64 (95% CI 0.41–0.87) (Fig. 2C,D).

image

Figure 2. (A) Receiver-operating characteristic (ROC) curve for the Padua Prediction Score (PPS) and hospital venous thromboembolism (VTE) (area under the curve [AUC] = 0.58, likelihood ratio for a positive PPS 1.21). (B) ROC curve for the PPS and VTE through 1 year (AUC = 0.63, likelihood ratio for a positive PPS 1.21). (C) ROC curve for PPS and in-hospital VTE in patients without anticoagulant prophylaxis (AUC = 0.54, likelihood ratio for a positive PPS 1.16). (D) ROC curve for PPS and VTE through 1 year in patients without anticoagulant prophylaxis (AUC = 0.64, likelihood ratio for a positive PPS 1.18).

Download figure to PowerPoint

Two hundred and thirty-seven (21.9%) patients died during the course of their hospitalization. The overall survival rate for the full duration of the follow-up was 64%. We found no correlation between in-hospital VTE and in-hospital mortality or overall survival (P = 0.96 and P = 0.59, respectively). However, a positive PPS was highly correlated with these two events (P < 0.0001, Fig. 3A). The odds ratio (OR) for in-hospital death in a patient with a positive PPS was 6.08 (95% CI 3.73–9.92). The OR did not change significantly when accounting for anticoagulant treatment. The AUC of the PPS for in-hospital death was 0.68 (95% CI 0.65–0.72, Fig. 3B).

image

Figure 3. (A) In-hospital mortality according to the Padua Prediction Score (PPS) (P < 0.0001). (B) Receiver-operating characteristic (ROC) curve for the PPS and in-hospital mortality (AUC = 0.68, likelihood ratio for a positive PPS 1.41).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References

In this study, we analyzed data from a prospective cohort of patients with sepsis admitted to general IM departments in a community-based medical center. We performed a post-hoc analysis of the rate of VTE and the effectiveness of the PPS to predict VTE events and guide prophylaxis treatment. In spite of the fact that our cohort was largely at a high risk of developing VTE owing to old age, co-morbidities and a diagnosis of sepsis at admission, we have found in-hospital overt VTE to occur in only 1.29% of the patients. The rate of VTE through 1 year was 1.94%. In this patient cohort PPS was inefficient to detect patients likely to sustain VTE, but was accurate in detecting patients at risk for mortality. VTE prophylaxis did not alter overall mortality, which is in line with previous studies that showed no effect on autopsy-proven PE [23] or overall mortality [24]. The nature of our analysis cannot lead to a conclusion that pharmacological prophylaxis in patients with sepsis does not alter their chances to acquire VTE within or post hospitalization, but our results can be considered as hypothesis generating and should facilitate more in-depth research.

The rate of symptomatic VTE in patients admitted to IM departments is in the range of 0.69% and 3.7% for short- and long-term follow-up periods, respectively. These were calculated in a meta-analysis of studies that prospectively followed cohorts of medical patients representing the case mix of IM patients, and assessed the rates of VTE [25]. Particularly, the presence of an infectious disease is associated with an increased risk for VTE. In one self-controlled case-series study, the likelihood of experiencing VTE doubled in the first 2 weeks after an acute urinary or respiratory infectious disease [26]. In another case-control study of medical outpatients, the OR for DVT associated with the presence of infectious disease was 1.95 [27]. A post-hoc analysis from the MEDENOX trial showed that the OR for VTE associated with acute infectious disease was 1.74 [28]. These observations are supported by studies in which anticoagulant treatment reduced the risk of VTE in patients with an acute infectious disease [29, 30].

The patients represented in our cohort reflect in many ways the current patient mix in IM departments in community-based hospitals: they are elderly, carry multiple co-morbidities, reside in long-term care facilities, and many have mental and physical deterioration. Thus, according to clinical guidelines, these patients are considered to be at an increased risk for VTE and physicians are encouraged to utilize prophylaxis measures. Most of our patients had multiple risk factors for VTE and over 70% of them had a positive PPS score. In spite of being severely ill and at a high risk for VTE, the VTE event rate observed in our study is similar to the rate of clinically overt DVT and PE observed in previous publications, in which the presentation of IM patients was not confined to a single diagnosis [25]. The current analysis confirms the relatively low rate of these events even in a highly morbid population, in which anticoagulation was administered in a very conservative manner (17.8%). The mortality rate in our cohort was very high, and one can argue that some of these events may be attributable to fatal PE. However, the results from the Heparin Prophylaxis Study Group trial show that necropsy-confirmed PE is rare in patients with acute infectious disease (0.27%) [30], and cannot explain the mortality rate observed in our study. Yet, the possibility of some under diagnosis of VTE in our cohort cannot be ruled out.

The PPS was ineffective in predicting anticoagulant administration. At the time of enrollment to this study no formal risk assessment tool was available to the treating physicians. We suspect that in spite of a general awareness of the risk of VTE, the fear of mal-effect of this intervention may have guided physician's choices, but under treatment is also a plausible explanation. In spite of the lack of adherence to pharmaceutical preventive strategies, PPS remained inefficient in predicting VTE in our cohort of patients. However, it was accurate in detecting death during and after the hospitalization. It is likely that in this highly morbid population, the PPS functions as a comorbid index rather than a specific VTE predictor, and thus is associated with mortality rather than with VTE. Moreover, characteristics such as immobility which have a high impact on the PPS final score also reflect the severity of presentation, which probably affects survival. In patients with sepsis, the AUC of the PPS to predict mortality was in the range of the AUC of other validated clinical predicting scales used to assess prognosis in medical patients (modified early warning score, simple clinical score, mortality in emergency department sepsis score and the rapid emergency medicine score), as assessed on our patient population and reported previously [20].

Our study has some limitations. In spite of being prospective in design some of the data for the current post-hoc analysis were retrospectively collected from electronic medical records. This limitation is mitigated by the use of OFEK system [22], which has the strength of covering full hospital and community patient level clinical, laboratory and imaging data, and by the fact that most of our patients are likely to be readmitted to our services if needed. We focus on clinically driven events, rather than on DVT observed on routine examinations and on necropsy-confirmed PEs. The debate on the significance of asymptomatic events is ongoing [25, 31], but current analyses of VTE events favor a focus on the former [32]. The current study did not employ an event adjudication committee for affirmation of the diagnosis of VTE. However, all events were evaluated and validated by a study researcher and by an expert internist. We also did not capture rates of bleeding events, and are thus reluctant to comment on adverse effects of pharmacological prophylaxis in our patient cohort. It is important to emphasize the unique characteristics of this patient cohort, which represents an old and largely debilitated population. Some may argue that our cohort may not accurately reflect current IM admittance characteristics. We believe that the data represent the demographics of patients in a growing number of community-based medical centers, and in view of longevity trends, are likely to become even more universally relevant in the near future.

The data from studies assessing the rate of adherence to VTE prevention strategies show without doubt that internists defy treatment in many clinical scenarios, in spite of wide spread publicity of the disease burden [1]. This is not surprising when comparing the numbers needed to treat to prevent one overt VTE (=292), and the numbers needed to treat one major bleeding (=336) [25]. In a previous survey among internists in Europe, the risk of bleeding was highlighted as the first and most important reason for not prescribing pharmacological prophylaxis [33]. Although under prescription may be a reason for the lack of adherence to societies’ guidelines, it is not unlikely that Internists are concerned about the direct and overt consequences of anticoagulation administration in their patients. The data presented above sheds additional light on presumed reasons for physicians’ skepticism related to the true rate and impact of VTE in their admitted patients. It is suggested that the PPS lacks the required precision to be endorsed in every clinical scenario. We believe that further studies are warranted to better define stratification tools to aid in the selection of patients that may truly benefit from prophylaxis measures, without undue risk.

Disclosure of Conflict of Interest

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References

The authors state that they have no conflict of interest.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. Disclosure of Conflict of Interest
  8. References
  • 1
    Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, Huang W, Zayaruzny M, Emery L, Anderson FAJ, Investigators E. Venous thromboembolism risk and prophylaxis in the acute care setting (ENDORSE study): a multinational cross-sectional study. Lancet 2008; 371: 38794.
  • 2
    Antiplatelet Trialists’ Collaboration: Collaborative overview of randomised trials of antiplatelet therapy. III. Reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. BMJ 1994; 308: 23546.
  • 3
    Anderson FA Jr, Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, Forcier A, Dalen JE. A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism: the Worcester DVT Study. Arch Intern Med 1991; 151: 9338.
  • 4
    Goldhaber SZ, Dunn K, MacDougall RC. New onset of venous thromboembolism among hospitalized patients at Brigham and Women's Hospital is caused more often by prophylaxis failure than by withholding treatment. Chest 2000; 118: 16804.
  • 5
    Goldhaber SZ, Tapson VF. A prospective registry of 5,451 patients with ultrasound-confirmed deep vein thrombosis. Am J Cardiol 2004; 93: 25962.
  • 6
    Cohen AT, Davidson BL, Gallus AS, Lassen MR, Prins MH, Tomkowski W, Turpie AG, Egberts JF, Lensing AW; ARTEMIS Investigators. Efficacy and safety of fondaparinux for the prevention of venous thromboembolism in older acute medical patients: randomised placebo controlled trial. BMJ 2006; 332: 3259.
  • 7
    Leizorovicz A, Cohen AT, Turpie AG, Olsson CG, Vaitkus PT, Goldhaber SZ. Randomized, placebo-controlled trial of dalteparin for the prevention of venous thromboembolism in acutely ill medical patients. Circulation 2004; 110: 8749.
  • 8
    Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, Leizorovicz A, Nguyen H, Olsson CG, Turpie AG, Weisslinger N. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. N Engl J Med 1999; 341: 793800.
  • 9
    Francis CW. Clinical practice. Prophylaxis for thromboembolism in hospitalized medical patients. N Engl J Med 2007; 356: 143844.
  • 10
    Cohen AT, Alikhan R, Arcelus JI, Bergmann JF, Haas S, Merli GJ, Spyropoulos AC, Tapson VF, Turpie AG. Assessment of venous thromboembolism risk and the benefits of thromboprophylaxis in medical patients.. Thromb Haemost 2005; 94: 7509.
  • 11
    Kucher N, Koo S, Quiroz R, Cooper JM, Paterno MD, Soukonnikov B, Goldhaber SZ. Electronic alerts to prevent venous thromboembolism among hospitalized patients. N Engl J Med 2005; 352: 96977.
  • 12
    Lecumberri R, Marqués M, Díaz-Navarlaz MT, Panizo E, Toledo J, García-Mouriz A, Páramo JA. Maintained effectiveness of an electronic alert system to prevent venous thromboembolism among hospitalized patients. Thromb Haemost 2008; 100: 699704.
  • 13
    Kahn SR, Lim W, Dunn AS, Cushman M, Dentali F, Akl EA, Cook DJ, Balekian AA, Klein RC, Le H, Schulman S, Murad MH. Prevention of VTE in nonsurgical patients prevention of VTE in nonsurgical patients Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141: e195S226S.
  • 14
    Barbar S, Noventa F, Rossetto V, Ferrari A, Brandolin B, Perlati M, De Bon E, Tormene D, Pagnan A, Prandoni P. A risk assessment model for the identification of hospitalized medical patients at risk for venous thromboembolism: the Padua Prediction Score. J Thromb Haemost 2010; 8: 24507.
  • 15
    Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med 2003; 348: 154664.
  • 16
    Heron M, Hoyert DL, Murphy SL, Xu J, Kochanek KD, Tejada-Vera B. Deaths: final data for 2006. Natl Vital Stat Rep 2009; 57: 1136.
  • 17
    Semeraro N, Ammollo CT, Semeraro F, Colucci M. Sepsis, thrombosis and organ dysfunction. Thromb Res 2012; 129: 2905.
  • 18
    Sundararajan V, MacIsaac CM, Presneill JJ, Cade JF, Visvanathan K. Epidemiology of sepsis in Victoria, Australia. Crit Care Med 2005; 33: 7180.
  • 19
    Duckit R, Palsson R, Bosanka L, Dagna L, Durusu Tanriover M, vardi M, CDIME group. Common diagnoses in internal medicine in Europe 2009: a pan-European, multi-centre survey. Eur J Intern Med 2010; 21: 44952.
  • 20
    Ghanem-Zoubi NO, Vardi M, Laor A, Weber G, Bitterman H. Assessment of disease-severity scoring systems for patients with sepsis in general internal medicine departments. Crit Care 2011; 15: R95.
  • 21
    Bone RC, Cerra FB, Dellinger P, Fein AM, Knaus WA, Schein RMH, Sibbald WJ. Committee. ASCC. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis – ACCP/SCCM Consensus Conference – American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101: 164455.
  • 22
    Nirel N, Rosen B, Sharon A, Blondheim O, Sherf M, Samuel H, Cohen AD. The impact of an integrated hospital-community medical information system on quality and service utilization in hospital departments. Int J Med Informatics 2010; 79: 64957.
  • 23
    Mahé I, Bergmann JF, d'Azémar P, Vaissie JJ, Caulin C. Lack of effect of a low-molecular-weight heparin (nadroparin) on mortality in bedridden medical in-patients: a prospective randomised double-blind study. Eur J Clin Pharmacol 2005; 31: 34751.
  • 24
    Kakkar AK, Cimminiello C, Goldhaber SZ, Parakh R, Wang C, Bergmann JF, LIFENOX Investigators. Low-molecular-weight heparin and mortality in acutely ill medical patients. N Engl J Med 2011; 365: 246372.
  • 25
    Vardi M, Steinberg M, Haran M, Cohen S. Benefits versus risks of pharmacological prophylaxis to prevent symptomatic venous thromboembolism in unselected medical patients revisited. Meta-analysis of the medical literature. J Thromb Thrombolysis 2012; 34: 119.
  • 26
    Smeeth L, Cook C, Thomas S, Hall AJ, Hubbard R, Vallance P. Risk of deep vein thrombosis and pulmonary embolism after acute infection in a community setting. Lancet 2006; 367: 10759.
  • 27
    Samama MM. An epidemiologic study of risk factors for deep vein thrombosis in medical outpatients: the Sirius study. Arch Intern Med 2000; 160: 341520.
  • 28
    Alikhan R, Cohen AT, Combe S, Samama MM, Desjardins L, Eldor A, Janbon C, Leizorovicz A, Olsson CG, Turpie AG. Risk factors for venous thromboembolism in hospitalized patients with acute medical illness risk factors for venous thromboembolism in hospitalized patients with acute medical illness. Arch Intern Med 2004; 164: 9638.
  • 29
    Alikhan R, Cohen AT, Combe S, Samama MM, Desjardins L, Eldor A, Janbon C, Leizorovicz A, Olsson CG, Turpie AG. Prevention of venous thromboembolism in medical patients with enoxaparin: a subgroup analysis of the MEDENOX study. Blood Coagul Fibrinolysis 2003; 14: 3416.
  • 30
    Gardlund B. Randomised, controlled trial of low-dose heparin for prevention of fatal pulmonary embolism in patients with infectious diseases. The Heparin Prophylaxis Study Group. Lancet 1996; 347: 135761.
  • 31
    Vaitkus PT, Leizorovicz A, Cohen AT, Turpie AG, Olsson CG, Goldhaber SZ, Group PMTS. Mortality rates and risk factors for asymptomatic deep vein thrombosis in medical patients. Thromb Haemost 2005; 93: 769.
  • 32
    Lederle FA, Zylla D, MacDonald R, Wilt TJ. Venous thromboembolism prophylaxis in hospitalized medical patients and those with stroke: a background review for an American College of Physicians Clinical Practice Guideline. Ann Intern Med 2011; 155: 60215.
  • 33
    Vardi M, Dagna L, Haran M, Duckit R. Attitudes towards and practice of venous thromboembolism prevention in general internal medicine wards. A multinational survey from member countries of the European Federation of Internal Medicine. Thromb Res 2011; 129: 5736.