Venous thrombosis following lower-leg cast immobilization and knee arthroscopy: From a population-based approach to individualized therapy

Osteoarthritis (OA) is the most common joint disease associated with pain and disability. OA patients are at a high risk for venous thrombosis (VTE). Here, we developed an interpretable machine learning (ML)-based model to predict VTE risk in patients with OA. To establish a prediction model, we used six ML algorithms, of which 35 variables were employed. Recursive feature elimination (RFE) was used to screen the most related clinical variables associated with VTE. SHapley additive exPlanations (SHAP) were applied to interpret the ML mode and determine the importance of the selected features. Overall, 3169 patients with OA (average age: 66.52 ± 7.28 years) were recruited from Xi’an Honghui Hospital. Of these, 352 and 2817 patients were diagnosed with and without VTE, respectively. The XGBoost algorithm showed the best performance. According to the RFE algorithms, 15 variables were retained for further modeling with the XGBoost algorithm. The top three predictors were Kellgren–Lawrence grade, age, and hypertension. Our study showed that the XGBoost model with 15 variables has a high potential to predict VTE risk in patients with OA.

Download Full-text

Geographical variation in incidence of knee arthroscopy for patients with osteoarthritis: a population-based analysis of Victorian hospital separations data

Internal Medicine Journal ◽

10.1111/imj.12438 ◽

2014 ◽

Vol 44 (6) ◽

pp. 537-545 ◽

Cited By ~ 2

Author(s):

M. Bohensky ◽

A. Barker ◽

R. Morello ◽

R. N. De Steiger ◽

A. Gorelik ◽

...

Keyword(s):

Geographical Variation ◽

Knee Arthroscopy ◽

Population Based

Download Full-text

A common genetic variation in the 3'-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis

Blood ◽

10.1182/blood.v88.10.3698.bloodjournal88103698 ◽

1996 ◽

Vol 88 (10) ◽

pp. 3698-3703 ◽

Cited By ~ 1991

Author(s):

SR Poort ◽

FR Rosendaal ◽

PH Reitsma ◽

RM Bertina

Keyword(s):

Confidence Interval ◽

Venous Thrombosis ◽

Case Control Study ◽

Direct Sequencing ◽

Population Based ◽

Coding Regions ◽

History Of ◽

Prothrombin Gene ◽

Polymerase Chain ◽

Control Study

We have examined the prothrombin gene as a candidate gene for venous thrombosis in selected patients with a documented familial history of venous thrombophilia. All the exons and the 5′- and 3′-UT region of the prothrombin gene were analyzed by polymerase chain reaction and direct sequencing in 28 probands. Except for known polymorphic sites, no deviations were found in the coding regions and the 5′-UT region. Only one nucleotide change (a G to A transition) at position 20210 was identified in the sequence of the 3′-UT region. Eighteen percent of the patients had the 20210 AG genotype, as compared with 1% of a group of healthy controls (100 subjects). In a population-based case-control study, the 20210 A allele was identified as a common allele (allele frequency, 1.2%; 95% confidence interval, 0.5% to 1.8%), which increased the risk of venous thrombosis almost threefold odds ratio, 2.8; 95% confidence interval, 1.4 to 5.6. The risk of thrombosis increased for all ages and both sexes. An association was found between the presence of the 20210 A allele and elevated prothrombin levels. Most individuals (87%) with the 20210 A allele are in the highest quartile of plasma prothrombin levels (> 1.15 U/mL). Elevated prothrombin itself also was found to be a risk factor for venous thrombosis.

Download Full-text

Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden Thrombophilia Study) [see comments]

Blood ◽

10.1182/blood.v85.10.2756.bloodjournal85102756 ◽

1995 ◽

Vol 85 (10) ◽

pp. 2756-2761 ◽

Cited By ~ 226

Author(s):

T Koster ◽

FR Rosendaal ◽

E Briet ◽

FJ van der Meer ◽

LP Colly ◽

...

Keyword(s):

Relative Risk ◽

Venous Thrombosis ◽

Protein C ◽

Protein S ◽

Population Based ◽

Repeated Measurements ◽

Protein C Deficiency ◽

Antithrombin Deficiency ◽

Vein Thrombosis ◽

Thrombosis Risk

A deficiency of protein C (PC), antithrombin, or protein S is strongly associated with deep-vein thrombosis in selected patients and their families. However, the strength of the association with venous thrombosis in the general population is unknown. This study was a population-based, patient-control study of 474 consecutive outpatients, aged less than 70 years, with a first, objectively diagnosed, episode of venous thrombosis and without an underlying malignant disease, and 474 healthy controls who matched for age and sex. Relative risks were estimated as matched odds ratios. Based on a single measurement, there were 22 (4.6%) patients with a PC deficiency (PC activity, less than 0.67 U/mL or PC antigen, less than 0.33 U/mL when using coumarins). Among the controls, the frequency was 1.5% (seven subjects). Thus, there is a threefold increase in risk of thrombosis in subjects with PC levels below 0.67 or 0.33 U/mL [matched odds ratio, 3.1; 95% confidence interval (CI), 1.4 to 7.0]. When a PC deficiency was based on two repeated measurements, the relative risk for thrombosis increased to 3.8 (95% CI, 1.3 to 10); when it was based on DNA-confirmation, the relative risk increased further to 6.5 (95% CI, 1.8 to 24). In addition, there was a gradient in thrombosis risk, according to PC levels. The results for antithrombin are similar to those for PC, although less pronounced (relative risk, 2.2; 95% CI, 1.0 to 4.7). We could not find an association between reduced total protein S (relative risk, 0.7; 95% CI, 0.3 to 1.8) or free protein S levels (relative risk, 1.6; 95% CI, 0.6 to 4.0) and thrombosis risk. Although not very frequent, PC and antithrombin deficiency are clearly associated with an increase in thrombosis risk.

Download Full-text