d -Dimer as a Screening Marker for Venous Thromboembolism After Surgery Among Patients Younger Than 50 With Lower Limb Fractures

Abstract

Objective:

For the present study, the authors hypothesized that the d-dimer levels would be systematically raised in a postoperative population of patients younger than 50 with lower limb fractures and to define a feasible cutoff value for identification of venous thromboembolism (VTE).

Methods:

Doppler ultrasonography of lower limbs was performed pre- and postoperatively to evaluate for deep vein thrombosis in 150 patients who underwent open reduction and internal fixation (ORIF). Plasma d-dimer levels were assessed 2 days before surgery and on the 3rd, 7th, and 10th days after surgery. Statistical analysis was carried out to define a feasible threshold for the d-dimer levels.

Results:

Plasma d-dimer levels were found to be systematically raised postoperatively, and they differed between patients with and without VTE significantly. On the third day after surgery, d-dimer levels of more than 3 mg/L indicated VTE with a sensitivity of 88.37% and a specificity of 96.96%, allowing for the definition of a feasible cutoff value. Duration of surgery, duration of tourniquet, ventilation time, and time of postoperative immobility of lower limbs were identified as highly significant risk factors for the development of VTE.

Conclusion:

Using a threshold of 3 mg/L, the d-dimer levels will screen out VTE with a high degree of sensitivity and specificity in younger patients who have undergone ORIF for lower limb fractures.

Keywords

young patients lower limb fractures deep vein thrombosis venous thromboembolism

Introduction

Venous thromboembolism (VTE) is the major cause of morbidity and mortality in posttraumatic patients who have undergone orthopedic surgeries.¹ Although, the condition is often asymptomatic, life-threatening pulmonary embolism (PE) is the first clinical manifestation of posttraumatic deep vein thrombosis (DVT) of the lower limb in many patients.² Of all trauma admissions with complicated injury, PE has been reported up to 0.55%.³ Thus, the prevention of postoperative DVT and PE must be balanced with the increased bleeding risks of anticoagulation.^4,5

Although low plasma d-dimer levels will rule out VTE with high sensitivity, high d-dimer levels have low specificity for the presence of VTE.^6,7 Many other conditions may also lead to high levels of plasma d-dimer,^8

–11 including infection and tissue injury.^12
–14 Therefore, d-dimer levels are believed to have a low positive predictive value for diagnosis of a thromboembolic phenomenon,¹⁵ which makes a “false positivity” and often leads to an anticoagulant therapy and an increased bleeding risk.

For the present study, we hypothesized that the activation of blood coagulation after surgery among patients younger than 50 years with lower limb fractures would elevate plasma d-dimer levels in a systematic and predictable way in the absence of VTE and that VTE would elevate d-dimer plasma levels beyond this range, implying a threshold with useful values for sensitivity and specificity in postoperative patients.

Methods

Patients

This study was conducted at Shanghai Tenth People’s Hospital, Tongji University, Shanghai, China. After obtaining institutional review board approval (SHSY-IEC-pap3.0/14-11), a prospective study was carried out from October 2012 until March 2014.

Several previous studies have demonstrated that d-dimer levels increase with age.^12,13 Previous study has established an age-adjusted d-dimer cutoff (age × 10 μg/L) as the positive test threshold,¹⁶ and our detecting system considered the d-dimer level of 0.5 mg/L (500 μg/L) as the threshold,⁷ therefore, we selected patients younger than 50 years in our study.

Cases were selected from patients younger than 50 years with lower limb fractures admitted to the Department of orthopedics, Shanghai Tenth People’s Hospital, Tongji University, Shanghai, China. The exclusion criteria were gravidity, age less than 18 years, perioperative administration of blood products, abnormal blood coagulation, administration of anticoagulation therapies, liver disease, and preoperative diagnosis of DVT. A total of 155 consecutive patients met the primary inclusion criteria and signed the informed consent.

Five patients were excluded from the study because of age less than 18 years (n = 2) or administration of aspirin for anticoagulation therapies (n = 3). Of the remaining 150 patients, 58 were females and 92 were males. Their mean age (±standard deviation) was 43 ± 19.1 years (range 18-49 years). All patients were diagnosed by X-ray or computed tomography (CT) of lower limb fractures, including 29 (19.3%) femoral fractures, 33 (22%) patella fractures, 35 (23.3%) tibial plateau fractures, 37 (24.7%) tibia-fibula fractures, and 16 (10.7%) ankle–foot fractures (Table 1). Table 1 summarizes the clinical variables.

Table 1.

Clinical Parameters of the VTE+ and VTE– Subgroups.^a

Variable	VTE+	VTE–	P Value
Number of patients	34	116
Mean age, years	44 ± 18.5	42 ± 16.1	.442
Sex
Male	20	72
Female	14	44
% Female	41.2%	37.9%	.451
Mean BMI	27.2 ± 4.8	26.5 ± 3.9	.542
Mean duration of operation, in min	95 ± 34	57 ± 26	<.001
Mean duration of tourniquet, in min	81 ± 27	45 ± 19	<.001
Mean ventilation time, in min	104 ± 35	69 ± 28	<.001
Immobility time postoperative, in h	75 ± 19	23 ± 18	<.001
Type of fractures
Femoral	8/34	21/116	.502^b
Patella	10/34	25/116	.405^b
Tibial plateau	7/34	26/116	1.00^b
Tibia-fibula	9/34	28/116	.682^b
Ankle–foot	0/34	16/116	.007^b

^aResults were deemed significant with P < .05 and highly significant with P < .001.

^bFisher exact test.

Abbreviations: VTE+, patients with venous thromboembolism; VTE–, patients without venous thromboembolism; BMI, body mass index.

All patients were treated with open reduction and internal fixation (ORIF) under general anesthesia with ventilation. The duration of surgery, from incision to wound closure, was recorded. All patients received similar prophylactic anticoagulant, with administration of low-molecular-weight (LMW)-heparin 4000 U once a day, from the first postoperative day until discharge. No other prophylactic treatments were performed. When VTE was diagnosed, therapeutic anticoagulation therapy was carried out, with administration of LMW-heparin 4000 U twice a day. All patients were followed up for 3 months or more.

d-Dimer Plasma Levels

d-dimer plasma levels were assessed 2days (−2) preoperatively and on the 3rd, 7th, and 10th days after surgery with the Siemens Innovance (Siemens Healthcare Diagnostics Products GmbH, Marburg, Germany). In order to minimize bias, all qualified laboratory technologists interpreting the results of d-dimer assays were unaware of the patients’ clinical presentation or the results of other objective tests.

Doppler Ultrasonography and Diagnosis of DVT or PE

All examinations were conducted with the same device (Sonoline Antares, Siemens Sector Healthcare, Forchheim, Germany) by the qualified examiners. Examinations were performed 2 days before surgery and between the third and seventh days after surgery. Diagnosis of DVT was based on direct thrombus detection by incomplete compressibility of the vessel, observation of lack of blood flow, or both. All patients were examined on both sides of lower limbs. Pulmonary embolism was diagnosed by CT examinations, which were performed if patients were diagnosed with DVT by Doppler ultrasonography or showed symptoms of PE, including thoracic pain, unexplained loss of consciousness, acutely reduced blood pressure, or acutely reduced oxygen saturation. All radiologists were blind to patients’ d-dimer levels.

Statistical Analysis

All data were entered in Excel database and analyzed with IBM SPSS Statistics 19.0. The mean values of d-dimer plasma levels 2 days before surgery and on the 3rd, 7th, and 10th postoperative days were calculated for patients with and without VTE (VTE+ group vs VTE– group). Between-group comparisons were performed using the Student t test. For dichotomous variables, the Fisher exact test was used. Results were deemed significant with P ≤ .05. For evaluation of plasma d-dimer level discrimination points, a receiver–operating characteristic curve (ROC) analysis was conducted. Youden index (J), sensitivity, specificity, positive and negative predictive values (PPV and NPV), and positive and negative likelihood ratios (LR+ and LR–) were also calculated.

Results

Postoperative VTE was diagnosed in 34 (22.7%) of all patients. Deep venous thrombosis was seen in 28 (18.7%) patients. Of the 28 diagnosed cases with DVT, 7 were symptomatic with calf swelling. Of 28 diagnosed cases with DVT, 4, 2, 2, 16, 2, and 2 clots were located in deep femoral vein, superficial femoral vein, distal femoral vein, popliteal vein, peroneal vein, and posterior tibial vein, respectively. Of the 7 symptomatic cases with DVT, 2, 4, and 1 clots were located in distal femoral vein, popliteal vein, and posterior tibial vein, respectively. Pulmonary embolism was detected in 9 (6%) of the patients. Of the 9 diagnosed cases with PE, 6 were diagnosed without DVT because of thoracic pain, and 3 PE was diagnosed by CT examinations, which were performed after patients were diagnosed with DVT. During all the follow-up period, there was no incidence of new VTE diagnoses or major bleeding events.

Risk Factors for VTE

Various clinical variables were assessed and analyzed for their potential impact regarding incidence of VTE (Table 1). A highly significant difference (P < .001) was demonstrated for duration of surgery, tourniquet time, ventilation time, and time of postoperative immobility of lower limbs. In patients with VTE, the mean duration of surgery was 95 minutes, whereas it was 57 minutes in patients without VTE. The mean ventilation time was 104 minutes in patients with VTE and 69 minutes in patients without VTE. The mean tourniquet time was 81 minutes in patients without VTE and 45 minutes in patients with VTE. The mean time of postoperative immobility was 75 hours in patients with VTE and 23 hours in patients without VTE.

There was no significant difference in the other clinical variables, including age, sex, body mass index (BMI), and type of fractures, between the patients with and without VTE (P > .05).

Plasma d-Dimer Level: Comparison of Mean Values

Student t test was used to compare the mean values of plasma d-dimer level. Comparison of mean plasma d-dimer levels in the VTE− group and VTE+ group was performed for values obtained 2 days preoperatively and on the 3rd, 7th, and 10th postoperative days. Significant differences were observed on the 3rd, 7th, and 10th postoperative days (P < .001 each). But there was no significant difference preoperatively (P = .55; Table 2).

Table 2.

Comparison of d-Dimer Levels (in mg/L) in the DVT+ and DVT−Subgroups.

VTE	−2		3		7		10
VTE	+	–	+	–	+	–	+	–
Mean (±SD)	0.73 (±0.23)	0.69 (±0.29)	6.66 (±2.99)	1.66 (±0.90)	3.87 (±2.09)	1.36 (±0.78)	2.53 (±1.51)	1.21 (±0.72)
P value	.55		<.001		<.001		<.001
95% CI	–0.07-0.13		4.38-5.63		2.04-2.91		0.94-1.68

Abbreviations: DVT+, patients with deep vein thrombosis; DVT–, patients without deep vein thrombosis; VTE, venous thromboembolism; SD, standard deviation; 95% CI, 95% confidence interval.

Determination of Discrimination Points: Plasma d-Dimer Level Thresholds

Considering the d-dimer levels were assessed on the 7th and 10th days under condition of therapeutic anticoagulation treatment in patients with VTE, the discrimination points were evaluated for plasma d-dimer levels only on the third postoperative day.

An ROC analysis was used to evaluate plasma d-dimer levels on the third day after surgery. The area under the curve (AUC) was calculated to be 0.924 with P < .01, a standard error of 0.03, and an 95% confidence interval (95% CI) ranging from 0.872 to 0.984 (Figure 2). Sensitivities and specificities of 5 cutoff values (0.5, 2, 3, 4, and 5 mg/L) are summarized in Table 3.

Figure 1.

Boxplot of d-dimer levels over time.The y coordinates indicate d-dimer plasma levels. The x coordinates indicate days, with −2 being preoperative, and 3, 7, and 10 indicating the 3rd, 7th, and 10th days after surgery.

Figure 2.

The receiver operating characteristic (ROC) curve for d-dimer levels on the third day after surgery.* (1) with the cut-off value of 0.5 mg/L, (2) with the cutoff value of 2 mg/L, (3) with the cutoff value of 3 mg/L, (4) with the cut-off value of 4 mg/L, and (5) with the cut-off value of 5 mg/L. * Area under the curve = 0.924.

Table 3.

Sensitivities and Specificities of Different Thresholds on Third Day Postoperation.

Cutoff value, in mg/L	Se, %	Sp, %	J	PPV, % (95% CI)	NPV, % (95% CI)	LR+ (95% CI)	LR– (95% CI)
0.5	95.35	18.84	0.14	25.62 (20.51-29.32)	93.25 (89.78-98.51)	1.22 0.73-2.31)	0.25 (0.11-0.32)
2	95.35	46.09	0.41	34.14 (29.53-30.02)	97.13 (91.10-99.25)	1.77 (0.96-2.41)	0.10 (0.05-0.15)
3	88.37	96.96	0.85	89.54 (83.95-94.12)	96.60 (90.32-99.71)	29.07 (25.10-35.13)	0.12 (0.06-0.15)
4	79.07	97.10	0.76	88.89 (82.16-91.42)	94.06 (90.16-99.18)	27.27 (22.35-30.89)	0.22 (0.16-0.29)
5	72.10	100	0.72	100	72.73 (69.56-16.89)	–	0.28 (0.26-0.33)

Abbreviations: Se, sensitivity; Sp, specificity; J, Youden index; PPV, positive predictive value; NPV, negative predictive value; LR+, positive likelihood ratio; LR–, negative likelihood ratio; 95% CI, 95% confidence interval.

The Youden index (J), PPV, NPV, and LR+, of 0.85, 89.54% (95% CI 83.95-94.12), 96.60% (95% CI 90.32-99.71), 29.07 (95% CI 25.10-35.13), respectively, reach their maximum distance to the diagonal reference line and optimizes the discriminatory power of a test, assuming that sensitivity and specificity are of the same value. Thus, a cutoff value of 3 mg/L was calculated (J = 0.85). The sensitivity was 88.37%, and the specificity was 96.96% (Table 3).

Discussion

Several studies have established that patients with trauma carry a significantly increased risk for the development of venous thromboembolic disease.^17,18 And VTE is the major cause of morbidity and mortality in posttraumatic patients who have undergone orthopedic surgeries.¹ Therefore, the postoperative diagnosis and management of VTE attract wide attention clinically.

d-Dimer is a breakdown product of cross-linked fibrin resulting from fibrinolysis. The breaking down of an intravascular clot leads to the presence of d-dimer in the blood.¹⁹ Therefore, elevated blood plasma d-dimer level has been considered as an indicator for VTE. However, some other conditions, including an increased age, tissue injury, infection, and acute respiratory distress syndrome,^{13,14,19
–21} also lead to the elevation in d-dimer levels.

The data from present study supports the hypothesis that undergoing ORIF for patients younger than 50 with lower limb fractures raises plasma d-dimer levels in a predictable way, and VTE leads to higher levels. Postoperative plasma d-dimer levels were significantly elevated beyond the standard threshold of 0.5 mg/L⁷ and decreased over time during the postoperative period. In patients with VTE, the levels were found to be significantly higher than in those without (P < .001).

For this study, the development of VTE was not influenced by patient’s age, sex, type of fractures, or BMI. This finding is inconsistent with other studies,²² which have demonstrated a higher risk of VTE in patients with older age. We consider that the contradiction in our study may be due to the sample size of 150 patients. However, longer duration of surgery, longer duration of tourniquet, longer ventilation time, and longer time of immobility are proved to be risk factors for the development of VTE in consistence with other studies.

In 1997, Wells et al carried out a prospective cohort trial of outpatients with symptoms and suspected DVT.²³ The DVT was diagnosed by ultrasound or venography. The results showed 3%, 17%, and 75% of the patients with low, moderate, and high pretest probability, respectively, had DVT. According to Wells’ study, young patients undergoing ORIF with lower limb fractures were at moderate pretest probability (recent immobilization of the lower extremities and major surgery). Of the 150 patients included in our study, the overall incidence of VTE and DVT were 22.7% and 18.7%, respectively. We consider that the discrepancy of the incidence between our study and other studies may be due to the small sample size of 150 patients in our study. We hope further studies with larger sample size in this area would be performed.

In 1884, Rudolph Virchow first proposed that thrombosis was the result of at least 1 of 3 underlying etiologic factors: 1) vascular abnormalities and endothelial dysfunction, 2) hypercoagulability and 3) stasis. This theory was widely accepted during the following hundred years. We failed to find the original report from the database. So we cited the articles that support Rudolph Virchow’s theory.^24,25 For patients undergoing ORIF for lower extremities, trauma and operation lead to an injury of soft tissue and then activate tissue and clotting factors, promoting thrombosis; trauma and surgical stress cause platelet adhesion and lead to fibrin deposition and thrombosis; anesthesia cause peripheral venous dilated, tourniquet during operation, and postoperative immobility of fractured limb lead to stasis of venous blood flow.

From the present study, a highly significant deviation from the ROC was found in prediction of plasma d-dimer levels as a marker for the absence or presence of postoperative VTE. The AUC for our overall group on the postoperative third day was 0.924. Therefore, the definition of a clinically feasible threshold becomes a probable goal.

Based on our data, we propose d-dimer screening with a threshold of 3 mg/L. With this cutoff value, sensitivity is somewhat lower (88.37% vs 95.35%), while specificity is clearly higher (96.96% vs 18.84%) than is achieved when the standard value of 0.5 mg/L is applied, and the Youden index, PPV, NPV, and LR+ are also higher. Thus, the value of 3 mg/L appears to be a robust threshold, and we believe this value has the potential to be an applicable cutoff value for postoperative screening.

In 88.37% of all patients with plasma d-dimer levels lower than 3 mg/L, no VTE was present on the third day after surgery, indicating that 4 of 34 patients would be at risk of undiagnosed VTE. Although not perfect, this test seems to be better than clinical suspicion as the sole “screening” method.

Limitations

Our study has several limitations. First, we diagnosed DVT or PE with Doppler ultrasonography or CT, however, venography is considered as the gold standard for diagnosis of DVT or PE because venography is invasive, and there is a risk that the examination itself may cause a PE.²⁶ Several studies have demonstrated the value of Doppler ultrasonography or CT for diagnosis of DVT or PE,^27
–29 and we believe that our results were reliable. Second, our finding partially contradicts other studies²² that have demonstrated a higher risk of VTE in patients with older age, thus further studies are needed. Third, our study was performed among young patients undergoing ORIF for lower limb fracture, and we hope further study is needed and d-dimer, as a marker to screen out DVT or PE, should be studied among those used to diagnose DVT and PE in the nonoperative settings. Fourth, our study sample is rather small, especially the patients diagnosed with DVT. Nevertheless, we provided a feasible threshold and our results have demonstrated our hypotheses. Further studies with larger sample size are needed to confirm or infirm our hypotheses.

Conclusion

Using a threshold of 3 mg/L, plasma d-dimer levels will indicate VTE in patients younger than 50 with lower limb fractures after ORIF with high sensitivity and specificity. Duration of surgery, duration of tourniquet, ventilation time, and time of lower limb immobility are highly significant risk factors for the development of VTE.

Footnotes

Authors’ Note

Yong Yang and Pengfei Zan contributed equally to the work.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Geerts

. Prevention of venous thromboembolism in high-risk patients. Hematology Am Soc Hematol Educ Program. 2006:462–466.

Anand

Wells

Hunt

. Does this patient have deep vein thrombosis? JAMA. 1998;279(14):1094–1099.

Menaker

Stein

Scalea

. Pulmonary embolism after injury: more common than we think? J Trauma. 2009;67(6):1244–1249.

Rogers

Cipolle

Velmahos

Rozycki

Luchette

. Practice management guidelines for the prevention of venous thromboembolism in trauma patients: the EAST practice management guidelines work group. J Trauma. 2002;53(1):142–164.

Selby

Geerts

Ofosu

. Hypercoagulability after trauma: hemostatic changes and relationship to venous thromboembolism. Thromb Res. 2009;124(3):281–287.

Bounameaux

Perrier

Righini

. Diagnosis of venous thromboembolism: an update. Vasc Med. 2010;15(5):399–406.

de Moerloose

Palareti

Aguilar

Legnani

Reber

Peetz

. A multicenter evaluation of a new quantitative highly sensitive D-dimer assay for exclusion of venous thromboembolism. Thromb Haemost. 2008;100(3):505–512.

Knowlson

Bacchu

Paneesha

McManus

Randall

Rose

. Elevated D-dimers are also a marker of underlying malignancy and increased mortality in the absence of venous thromboembolism. J Clin Pathol. 2010;63(9):818–822.

Raimondi

Bongard

de Moerloose

Reber

Waldvogel

Bounameaux

. D-Dimer plasma concentration in various clinical conditions: implication for the use of this test in the diagnostic approach of venous thromboembolism. Thromb Res. 1993;69(1):125–130.

10.

Taylor

Jr Toh

Hoots

Wada

Levi

. Towards definition, clinical and laboratory criteria, and a scoring system for disseminated intravascular coagulation. Thromb Haemost. 2001;86(5):1327–1330.

11.

Wells

. The role of qualitative D-dimer assays, clinical probability, and noninvasive imaging tests for the diagnosis of deep vein thrombosis and pulmonary embolism. Semin Vasc Med. 2005;5(4):340–350.

12.

Righini

Goehring

Bounameaux

Perrier

. Effects of age on the performance of common diagnostic tests for pulmonary embolism. Am J Med. 2000;109(5):357–361.

13.

Tardy

Tardy-Poncet

Viallon

. Evaluation of D-dimer ELISA test in elderly patients with suspected pulmonary embolism. Thromb Haemost. 1998;79(1):38–41.

14.

Johna

Cemaj

O’Callaghan

Catalano

. Effect of tissue injury on D-Dimer levels: a prospective study in trauma patients. Med Sci Monit. 2002;8(1):Cr5–Cr8.

15.

Janssen

Heebels

de Metz

. Reliability of five rapid D-dimer assays compared to ELISA in the exclusion of deep venous thrombosis. Thromb Haemost. 1997;77(2):262–266.

16.

Douma

le Gal

Sohne

. Potential of an age adjusted D-dimer cut-off value to improve the exclusion of pulmonary embolism in older patients: a retrospective analysis of three large cohorts. BMJ. 2010;340:c1475.

17.

Knudson

Collins

Goodman

McCrory

. Thromboembolism following multiple trauma. J Trauma. 1992;32(1):2–11.

18.

Rogers

. Venous thromboembolism in trauma patients: a review. Surgery. 2001;130(1):1–12.

19.

Enderson

Chen

Robinson

Maull

. Fibrinolysis in multisystem trauma patients. J Trauma. 1991;31(9):1240–1246.

20.

Idell

James

Levin

Schwartz

. Local abnormalities in coagulation and fibrinolytic pathways predispose to alveolar fibrin deposition in the adult respiratory distress syndrome. J Clin Invest. 1989;84(2):695–705.

21.

Carvalho

DeMarinis

Scott

Silver

Schmaier

Colman

. Activation of the contact system of plasma proteolysis in the adult respiratory distress syndrome. J Lab Clin Med. 1988;112(2):270–277.

22.

Kearon

. Natural history of venous thromboembolism. Circulation. 2003;107(23 Suppl 1):I22–I30.

23.

Wells

Anderson

Bormanis

. Value of assessment of pretest probability of deep-vein thrombosis in clinical management. Lancet. 1997;350(9094):1795–1798.

24.

Anderson

Jr Spencer

. Risk factors for venous thromboembolism. Circulation. 2003;107(23 Suppl 1):I9–I16.

25.

Ballard

Marques

. Pathology consultation on the laboratory evaluation of thrombophilia: when, how, and why. Am J Clin Pathol. 2012;137(4):553–560.

26.

Terao

Ozaki

Sato

. Diagnosis of deep vein thrombosis after operation for fracture of the proximal femur: comparative study of ultrasonography and venography. J Orthop Sci. 2006;11(2):146–153.

27.

Borris

Christiansen

Lassen

Olsen

Schott

. Comparison of real-time B-mode ultrasonography and bilateral ascending phlebography for detection of postoperative deep vein thrombosis following elective hip surgery. The Venous Thrombosis Group. Thromb Haemost. 1989;61(3):363–365.

28.

Lensing

Prandoni

Brandjes

. Detection of deep-vein thrombosis by real-time B-mode ultrasonography. N Engl J Med. 1989;320(6):342–345.

29.

Agnelli

Volpato

Radicchia

. Detection of asymptomatic deep vein thrombosis by real-time B-mode ultrasonography in hip surgery patients. Thromb Haemost. 1992;68(3):257–260.