| | Prevalence and Risk Factors of Asymptomatic Lower Extremity Deep Venous Thrombosis in Asian Neurorehabilitation Admissions in SingaporeAbstract Chua K, Kong KH, Chan SP. Prevalence and risk factors of asymptomatic lower extremity deep venous thrombosis in Asian neurorehabilitation admissions in Singapore. ObjectiveTo determine the local prevalence and risk factors of asymptomatic lower limb deep venous thrombosis (DVT) among neurorehabilitation admissions. DesignA prospective observational single-center study. SettingTertiary rehabilitation center affiliated to a public hospital. ParticipantsA total of 419 Asian neurorehabilitation admissions with a mean of 26 days to rehabilitation. InterventionAdmission screening protocol included quantitative D-dimer assay within 24 to 48 hours of rehabilitation admission and targeted hemiplegic/weaker lower-extremity venous duplex ultrasonography was performed if D-dimer assay level was elevated at 0.34 μg/mL or higher. Main Outcome MeasuresThere were 251 (59.9%) men, and the subjects were predominantly Chinese (76.6%). Subjects had a mean age of 59±15 years. Admitting diagnoses included ischemic stroke (212), hemorrhagic stroke and subarachnoid hemorrhage (129), traumatic brain injury (59), and nontraumatic brain injury (19). The screening protocol included a quantitative D-dimer assay within 24 to 48 hours of rehabilitation admission, and hemiplegic/weaker lower-extremity venous duplex ultrasonography was performed if D-dimer assay level was elevated at 0.34 μg/mL or higher. ResultsAltogether, 247 (58.9%) patients had an elevated D-dimer assay, and all underwent venous duplex ultrasonography. The incidence of lower-limb DVT was 5.01% (21), including 11 proximal and 10 distal DVT. No patients had clinical pulmonary embolism. Using 247 venous duplex ultrasonography results for analyses of correlates with logistic regression analyses, no significant demographic or clinical predictive factors for DVT were found. ConclusionsThis study confirms that asymptomatic lower limb DVT is indeed uncommon in Asian neurorehabilitation admissions. Possible reasons include genetic or ethnic protective factors, early walking initiated at rehabilitation, and timing of the admission protocol (median of 14 days postevent) when the maximal thrombotic risk was on the decline. List of Abbreviations: BI, brain injury, CI, confidence interval, DVT, deep venous thrombosis, GCS, Glasgow Coma Scale, LOS, length of stay, OR, odds ratio, PE, pulmonary embolism, PTA, posttraumatic amnesia, SAH, subarachnoid hemorrhage, TBI, traumatic brain injury PATIENTS ADMITTED TO inpatient neurorehabilitation programs for stroke or BI are at increased risk of developing DVT. Factors contributing to this increased risk include relative immobilization, stroke or BI-related hemiplegia or hemiparesis, trauma-related vascular complications, long bone fractures, cancer, and severe neurotrauma, all of which predispose to thrombus formation. It is well known that the individual clinical features of DVT and PE are notoriously nonspecific. Hence, underdiagnoses or misdiagnoses are well documented in this population, particularly in the elderly.1 The diagnostic dilemma is pertinent in neurorehabilitation patients who may underreport symptoms of limb pain or swelling or dyspnea because of aphasia, neglect, cognitive impairment, or altered conscious states.2 Clinically apparent DVT is considered to be a major cause of PE, with mortality from PE of up to 37%. The risk of fatal PE associated with untreated subclinical DVT is lower, although it remains significant. Few studies address the incidence and risk factors of lower-limb DVT and PE after neurologic events such as stroke and acquired BI. In acute hemiplegic stroke, studies with 125 I fibrinogen screening have shown an incidence of DVT of approximately 50% within 2 weeks in the absence of heparin prophylaxis; most affect the paralyzed leg and are asymptomatic. Approximately two thirds of these are below-knee DVTs, in contrast with nonstroke or surgical patients presenting with symptomatic DVT, in whom most are proximal.1, 2, 3, 4 DVTs develop as early as the second day, with the peak incidence between days 2 and 7. Contrast venography and 125 I fibrinogen are no longer favored because of their invasive and potentially infective complications and radiation risk. Real-time venous duplex ultrasonography is regarded as the most sensitive and specific noninvasive test for the detection of proximal DVT of the legs, with a positive predictive value of an abnormal ultrasound greater than 90% and a sensitivity of 97% for symptomatic patients for lower-extremity DVT. Doppler ultrasound is a powerful technique for detecting symptomatic proximal DVT. It also has the advantage of noninvasiveness, low cost, high sensitivity, availability, portability, and lack of radiation exposure.5, 6 Unselective imaging of the lower limbs at rehabilitation is cost-prohibitive and would have major resource implications. Plasma D-dimer assay derived from cross-linked fibrin breakdown products generated from the degradation of the fibrin matrix of fresh venous thromboemboli is a safe and cost-effective exclusionary test for DVT when used within defined diagnostic strategies; absolute levels correlate with thrombus load and are significantly higher in patients with proximal DVT than below-knee DVT.1, 7 However, optimal cut points for predicting DVT are variable depending on the desired sensitivity and specificity.8 In the early days of acute strokes, D-dimers may be falsely elevated because acute nonlacunar ischemic strokes can increase D-dimer levels, which may decay to baseline over the next few days. In addition, healthy ranges differ in acute and rehabilitation settings. The sensitivity and negative predictive values of D-dimer latex assay in diagnosing or excluding DVT and PE in a rehabilitation inpatient population were noted in previous studies to be high at 95.2% and 96.2%, respectively; however, specificity and positive predictive value were low at 55.3% and 48.7%, respectively.9 Hence, the D-dimer assay may be a reliable and useful screening test for DVT and may allow the identification of a subgroup of patients to undergo targeted limb imaging as the sensitivity and specificity approach 100% and 66%, respectively, in patients greater than 3 weeks poststroke. Thus, such a selective protocol may reduce the need for expensive and time-consuming diagnostic investigation for at-risk rehabilitation inpatients.1, 7, 8, 9 These considerations were taken into account in determining the optimal screening protocol in our patient population. There is longstanding belief that venous thromboembolism in Asia is extremely rare and clinically insignificant, but recent surveys suggest that this may no longer be true. In Asia, the incidence of venous thrombosis and PE in some high-risk clinical settings is increasing and may approach that seen in the West. There is also a suspicion that predisposing factors may differ in Asia and the West. For instance, there appears to be a higher prevalence of inherited antithrombin, protein C, and protein S deficiency among Chinese patients presenting in Hong Kong or Taiwan with venous thromboembolism. In hospitalized patients with acute stroke in mainland China, the incidence of occult DVT was 21.7% using a protocol of serial D-dimer assay and lower limb ultrasonography at 7 to 10 days poststroke, with older men and more dependent patients at higher risk of DVT.10, 11, 12, 13, 14, 15, 16 In Singapore, the rate of acute DVT in the hospitalized population appears to be on the rise from 7.9 per 10,000 admissions in 1990 to 15.1 per 10,000 admissions in 2002.17 In a locally selected group of 106 patients with brain tumors undergoing neurosurgery locally without routine heparin prophylaxis, the incidence of lower-limb DVT using ultrasound examinations on preoperative and postoperative day 5 to 7 was 4.7%, and all had benign tumors.18 More recently, DeSilva et al19 detected an incidence of 30% of DVT in acute strokes in Asians in Singapore with lower-limb paresis at days 7 to 10, rising to 45% at 25 to 30 days poststroke, indicating that the asymptomatic DVT incidence locally is not a rare problem. Local variations exist, because there is no national standard screening protocol in these populations. Hence, the objectives of this study were (1) to determine the incidence of asymptomatic lower-limb DVT among Asian neurorehabilitation admissions in a single rehabilitation center in Singapore and to compare this with similar patient populations locally and worldwide; (2) to determine whether there was a correlation between D-dimer assay and the ultrasound diagnosis of DVT; and (3) to determine whether there were demographic, clinical, or rehabilitation predictors or risk factors for DVT in this population, particularly with regard to functional rehabilitation data such as lower-limb motricity index score and rehabilitation outcome measures using the FIM. Methods  Study Design This study consisted of a prospective observational study of consecutive admissions to a single rehabilitation center for acute strokes and brain injuries greater than 15 years of age over a 21-month period from April 2005 through January 2007. The rehabilitation center has direct links to an acute stroke unit and level 2 trauma center serving most stroke and BI acute admissions for the country. All acute stroke or BI diagnoses were made by admitting neurologists, neurosurgeons, or emergency department physicians and confirmed by neuroimaging studies (computed tomography or magnetic resonance imaging brain scans) within 3 to 6 hours of admission. Prior to data collection, the study also obtained approval from the local ethics committee of the hospital's institutional review board. The study did not receive external funding. Subjects Patients were included in the study if they had a diagnosis of stroke (ischemic or hemorrhagic stroke including spontaneous subarachnoid hemorrhage), TBI, and other acquired BI (including hypoxic encephalopathy, cerebral tumors, metabolic encephalopathy, viral meningoencephalitis, poisonings, and other causes). Recurrent strokes and brain injuries and those receiving antiplatelet or warfarin therapy for secondary stroke prophylaxis were included. Patients were excluded if they had spinal cord injury or were readmissions from the community for reasons of secondary functional decline not related to acute stroke or BI. Data from subjects who refused venepuncture required for D-dimer assay; patients who refused or could not undergo venous duplex ultrasonography because of agitation, restlessness, or limb fractures; and those with DVT or PE diagnosed prior to transfer to rehabilitation were also excluded. The DVT Screening Protocol The DVT screening protocol consisted of a single quantitative D-dimer assay using latex agglutination methods performed within 24 to 48 hours of rehabilitation admission. The LIATEST D-dimer assay, which is an immuno-turbidimetric quantitative assay method based on a latex microparticle agglutination test, was used. All venous duplex ultrasonography studies were performed in a single radiology department within the rehabilitation center by the same radiographer with specific experience and reported by trained radiologists using a Toshiba Xario venous Doppler ultrasound machine, version 2002,a during the entire study period. Targeted single-episode imaging of the hemiplegic (weaker) lower extremity was performed if the D-dimer level was elevated (≥0.34μg/mL). This predetermined cutoff value was the 97.5 maximal percentile value obtained from an analysis of 200 healthy local hospital staff carried out by the hematology department (Kuperan P, 2005, unpublished data). No representative sampling of hospital patients was done in this cutoff determination. Routine imaging of both lower limbs was not done except in those with bilateral lower-limb weakness for logistic and cost containment reasons. Routine follow-up imaging was also not performed. The presence of DVT was determined by a positive venous duplex ultrasonography result of either proximal or distal DVT. DVT was classified into proximal (iliac, common superficial, deep femoral, popliteal vein involvement) or distal DVT (posterior tibial, peroneal, gastrocnemius, soleal, perforator vein involvement). Venous duplex ultrasonography findings were classified as absent DVT (normal study done with elevated D-dimer assay), present DVT (proximal or distal DVT), or not performed (normal D-dimer assay levels). Routine pulmonary angiography or spiral computed tomography lung scans were not performed to exclude PE, and the diagnosis of PE was made on clinical grounds and investigated accordingly. Data and Outcome Variables Data of interest included acute demographic variables such as age, sex, race, acute hospital LOS, admitting diagnoses (stroke, SAH, TBI, acquired non-TBI), stroke type (ischemic or hemorrhagic), BI type (traumatic, nontraumatic) and comorbid conditions (cardiac failure, atrial fibrillation, ischemic heart disease, cancer, diabetes hypertension, previous stroke, or cancer diagnosis). Data regarding acute management included heparin prophylaxis, stroke treatment (none, antiplatelet therapy, anticoagulation with warfarin), and presence or absence of neurosurgical procedures. Rehabilitation and functional data included acute and rehabilitation LOS (days), lower-limb motor power as measured by the lower-limb motricity index score (20), and functional independence as measured by the FIM score. The lower-limb motricity score ranges from 0 to 15 and measures motor strength in the hip, knee, and ankle.20 In addition, the FIM-motor subscale and FIM-walk subscale were also evaluated.21 The FIM score, which is a global indicator of functional impairment and burden of care, has 7 levels for its 18 individual items and has a highest total score of 126 and a lowest score of 18. It was scored within 72 hours of rehabilitation admission by trained rehabilitation therapists, and subscores include FIM-motor (subtotal score=91) and FIM-walk (subtotal score=7).21 A FIM-walk score of 1 to 2 indicates that the patient is nonambulatory, while a FIM-walk score of 6 to 7 indicates independence. Severity of TBI was measured using the acute admission GCS scores and the Westmead PTA scale scores on admission to the rehabilitation center.22, 23 Neurorehabilitation Program The center's rehabilitation program consisted of intensive therapies 2 to 3 hours daily, 5.5 days a week, with rest on weekends. The emphasis was on mobility and functional activities training and discharge planning. All patients received thigh-length graduated pressure compressive stockings for both lower limbs on admission to rehabilitation until discharge as routine nonpharmacologic DVT prevention and early mobilization out of bed within 24 hours of rehabilitation admission if medically stable. Subcutaneous low-molecular-weight heparin prophylaxis was not routinely practiced for all patients with severe limb paresis after stroke. However, this was continued if heparin therapy was commenced at the acute stroke unit and discontinued when patients were ambulatory for 20m with or without aid. Statistical Analysis Statistical analysis was carried out using SPSSb and STATA 10.0.c All statistical tests were carried out at a 5% level of significance. To ascertain how the identified factors were associated with DVT (0, absent; 1, present) and D dimer (0, normal; 1, elevated), univariate and multivariate logistic regression were employed. In situations where there was cell sparsity, the exact logistic regression was used. The final model was chosen based on backward elimination, with the removal probability fixed at 0.2. Results Altogether, 419 consecutive neurorehabilitation admissions were studied from November 2005 to April 2007. A total of 13 patients were excluded from the analyses because of a prerehabilitation diagnosis of DVT in 2 and absent D-dimer or venous duplex ultrasonography results in 11 as a result of procedure refusal, agitation, or limb fractures. The 419 patients were admitted on an average ± SD of 26.0±42.8 days postneurologic event. These patients formed 64.6% of a total of 650 inpatient admissions to the center during the 21-month study period. The mean age was 59.0±15.0 years (range, 15–94y), and 59.9% (251) were men. Ethnic groups included Chinese (76.8%), Malay (14.3%), Indian (7.2%), and other races (1.7%). This distribution corresponded to the local population demographics.22 Ischemic stroke comprised half of the diagnostic groups (212), followed by hemorrhagic strokes including spontaneous subarachnoid hemorrhage (129), TBI (59), and other non-TBI (19). The non-TBI group was composed of cerebral tumors (11), hypoxic encephalopathy (5), and other injuries (3). For univariate analysis, the ethnic groups were regrouped into Chinese (322) and non-Chinese (97) ethnic groups. The diagnostic etiologies were regrouped into ischemic strokes (212), hemorrhagic strokes (129), and other brain injuries (78). Similarly, the sites of ischemic strokes were regrouped into anterior circulation (88), posterior circulation (46), and lacunar stroke (78) categories. For the hemorrhagic strokes, the sites of stroke were regrouped into subcortical (basal ganglia, thalamic, cerebellum, and pons; 88) and cortical/other regions (18). The median time to rehabilitation was 26.0 days (range, 1–378d), and median rehabilitation LOS was 28.0 days (range, 3–379d). The FIM total, subset motor, and subset walking scores were 58.6±23.5 (SD), 36.2±16.6, and 2.6±1.4, respectively. The mean lower-limb motricity index score was 7.4±5.1, and the median score was 9.0. The comorbid medical conditions and acute management are summarized in table 1. Hypertension followed by diabetes mellitus, ischemic heart disease, and a previous history of stroke were the most common comorbidities, and these were not different from the local literature for stroke risk factors. | ⁎ Using the Oxfordshire community classification of stroke. |
Almost half of the subjects (195) received antiplatelet therapy for secondary stroke prophylaxis, and only 4.5% of patients (19) were anticoagulated with warfarin for stroke management. The remainder (205) received no blood thinners because of underlying cerebral hemorrhage, SAH, or TBI, which contraindicated their use. In this cohort, only 7 (1.7%) patients received routine prophylaxis with subcutaneous heparin on admission. D-dimer levels were elevated in 247 (58.9%) subjects, and all of these 247 patients underwent venous duplex ultrasonography. Twenty-one new DVTs were detected in 420 admissions with this protocol—an overall DVT incidence of 5.01% for this cohort. There were no symptomatic PE cases. Table 2 shows the clinical factors that were correlated with elevated D-dimer levels. Older, more dependent patients with lower total and motor FIM admission scores and those of non-Chinese ethnicity were significantly associated with elevated D-dimer levels (>0.34μg/L). However, the effect of motor FIM score on D-dimer levels was masked by the more dominating effect of total FIM score in the multivariate model. As such, only age, non-Chinese ethnicity, and higher dependency as measured by lower total admission FIM score were identified as significant factors affecting D-dimer levels in the multivariate model (see table 2). FIM motor subscores and lower-limb motricity score were not significantly associated with elevated D-dimer levels. | ⁎ Statistically significant at 5%. |
Logistic regression analyses were performed for a total of 247 cases with venous duplex ultrasonography results to determine whether there were demographic or clinical predictors, which correlated with DVT. D-dimer levels were found to be strongly predictive of DVT for this sample, and the effect was positive (OR=1.17; 95% CI, 1.02–1.34; P=.02) (table 3). | ⁎ Statistically significant at 5%. †Sample=121. ‡Analyzed with exact logistic regression in view of the 0 cells. |
Heparin prophylaxis, which was carried out in only 7 (1.7%) patients, was not found to be protective against DVT, because only 1 heparinized patient had an elevated D-dimer level and underwent venous duplex ultrasonography (P=.17). Other factors such as age, sex, Chinese ethnicity, comorbid conditions, neurosurgical operations, presence of atrial fibrillation, antiplatelet therapy, FIM total and subset FIM-motor and FIM-walk scores, and lower-limb motricity score were entered into the multivariate logistic regression model. With the exception of D-dimer levels, all were not significantly associated with the presence of DVT. There was no evidence of interaction effects among the variables. However, it was noteworthy to find that patients with less severe limb paresis, represented by a higher lower-limb motricity score, were more likely to have a slightly lower risk of DVT (OR=0.84; 95% CI, 0.88–1.04; P=.30), although the effect was not statistically significant (see table 3). Further subanalyses of 59 patients with TBI in this predominantly older (mean age, 50.4±20.5y) and more severely injured (mean admission GCS, 9.4±3.9) cohort were carried out. The demographic profile of these patients was similar to the local older TBI population (men:women=1.8:1; 64.4% Chinese; 40.4% with admission; GCS<8; (Yap SGM, 2008, unpublished data). An elevated D-dimer level was present in 36 (61%) patients, and all underwent venous duplex ultrasonography examination. The prevalence of DVT was 5.1% in this cohort (3 of 59 patients, 2 proximal and 1 distal DVT), and this mirrored the prevalence in the general neurorehabilitation population. Discussion Studies of patients in neurorehabilitation settings after acute stroke report incidences ranging from 8.5% to 18% when asymptomatic screening is performed compared with a reported incidence of 1.6% for patients with symptomatic DVT. Screening protocols vary according to unilateral or bilateral lower-limb imaging, whether distal imaging is performed as well, and whether the D-dimer assay is used to correlate DVT findings. During BI rehabilitation in the white population, incidences of asymptomatic DVT range from 8% to 32% using a variety of methods such as Doppler ultrasonography and impedance plethysmography.10, 11, 12, 13, 14 To our knowledge, this is the first local study to establish prospectively the incidence of asymptomatic lower-limb DVT in a neurorehabilitation population using a DVT screening protocol employing cost-effective single targeted venous duplex ultrasonography in those with elevated D-dimer assay on admission, and to attempt correlation with demographic and rehabilitation data. It was not the purpose of this study to determine the sensitivity or negative predictive value of the D-dimer assay or venous duplex ultrasonography for diagnosis of DVT. Our findings confirm suspicions that the incidence of DVT is uncommon compared with incidences reported in comparable patient populations in the West and in Asia and Singapore. For the 247 out of 420 patients with an elevated D-dimer assay who underwent venous duplex ultrasonography, 21 new cases of DVT were detected, giving an incidence of 5.01%. D-dimer assay levels were overall highly correlated with the presence of DVT, lending further support to our screening protocol (see table 3). In a mixed rehabilitation population (orthopedic, stroke, spine, and cancer group), Sachdev et al2 found a DVT incidence of 34% (77% had proximal DVT) in 579 inpatients who underwent bilateral limb venous duplex ultrasonography at 6 weeks postevent. In the BI rehabilitation patients, the incidence of DVT from various authors is lower. Yablon et al12 studied BI rehabilitation admissions within less than 120 days of event using routine proximal venous duplex ultrasonography, and the incidence of occult DVT was 11.1% with significant correlations between D-dimer assay and the presence of DVT, particularly in those with brain tumors. Only 10% of this population received anticoagulation. Cifu et al13 studied a smaller similar BI rehabilitation admissions group within 2 months of injury, in whom all received either subcutaneous heparin or equivalent intermittent compression devices, and his incidence of occult DVT was similar (11%) in 153 patients without significant predictive factors. Patients with TBI had a higher incidence of occult DVT (18%) than patients without TBI (4%). His study emphasized the importance of baseline screening for occult DVT in the brain-injured population who had routine prophylaxis as standard of care.13 Similarly, in 164 brain-injured patients undergoing rehabilitation within 4 months of event, Meythaler et al14 found a proximal DVT incidence by routine ultrasound Doppler screening of 8.5% without significant predictive factors. To date, there are no local studies on the incidence of DVT in the BI population undergoing rehabilitation. The incidence of occult DVT after acute stroke ranges from 11% to 66% because of differences in population ethnicity; age; severity of stroke and motor weakness; DVT prophylaxis and antiplatelet therapy, which reduce DVT incidence; screening methods; and time from event.19 In Asians, this is believed to be much lower than in whites, and in the Chinese stroke population, the occult DVT incidence ranges from 17% to 21% using fibrinogen scanning and ultrasonography, respectively.16, 23, 24 Locally, a recent study by DeSilva et al19 at the Singapore General Hospital after acute ischemic stroke within 7 to 10 days using bilateral Doppler ultrasound suggested that occult DVT is present in 30% of patients, and this rises to 45% at 4 weeks, with a higher proportion of distal DVT (66%–77%) in general. Notably, our study had a higher proportion (52.3%) of proximal DVT than the DeSilva19 study, and this may be a result of differences in screening protocol in which untargeted venous ultrasound screening results in a higher incidence of distal DVT. Our relatively low incidence of DVT despite moderate to severe functional dependency could be attributed to several factors. First, local racial (predominantly Chinese) and ethnic/genetic factors contributing to a lowered thrombotic potential, early mobilization, and transfer to intensive neurorehabilitation (median time to rehabilitation ≤2wk) and early commencement of acute stroke unit rehabilitation therapies such as assisted walking and mobilization out of bed within 24 hours of admission. In addition, etiologies such as TBI, which in several studies in white subjects indicate a higher risk of DVT in a neurorehabilitation population because of concomitant polytrauma or lower-limb fractures or vascular or soft tissue injuries, may delay mobilization, transfer to rehabilitation, and onset of walking. However, TBI comprised a minority (14.1%) of our study population, and they did not have a significantly higher DVT incidence despite the severity of injuries as shown by a mean admission GCS of 9 and mean PTA scores of 52 days. The current evidence also questions whether lower-limb fractures are truly high-risk indicators for DVT in patients with brain trauma, as previously believed.12, 25 The presence of traditional risk factors such as cardiac failure, atrial fibrillation, and cancer also failed to correlate significantly with the presence of DVT, likely because of their small representation (see Table 1, Table 3). Weaker patients with a lower-limb motricity score had a tendency toward a higher risk of DVT, but walking ability as measured by the FIM-walk and FIM-motor subscores did not influence DVT risk. These factors may be reflective of more serious stroke or BI and a higher likelihood of medical complications, which could increase the risk of DVT through prolonged immobility.21 In addition, antiplatelet therapy in 46.5% of patients and subcutaneous low-molecular-weight heparin prophylaxis in 1.7% were not protective in our cohort26 (see table 3). Patients with DVT also did not have a significantly longer rehabilitation LOS, likely because of prompt diagnosis and appropriate treatment through this admission screening protocol, thus allowing patients to continue intensive rehabilitation with minimal disruption and delays. Elevated D-dimer assay levels were significantly correlated with an older, non-Chinese population with higher functional dependency but not more severe limb weakness and poorer walking ability as measured by lower limb motricity score and FIM-walk scores, respectively. This has a possible impact on the use of D-dimer on older Asian neurorehabilitation populations who may have higher false-positive rates of D-dimer assay and prompt unnecessary venous duplex ultrasonography imaging, which may result in higher rehabilitation costs. Certain limitations in this study also need to be highlighted. Routine venous duplex ultrasonography imaging of both lower limbs in all patients regardless of the side of hemiplegia could result in a higher pick-up rate of DVT and PE and approach that in white populations. The cut-off range for D-dimer assay used in our institution at the time of the study was not substantiated by studies of a comparative hospitalized population; however, at this level, it was sufficient to detect a low incidence of DVT. Follow-up venous duplex ultrasonography was also not done routinely. Based on previous local studies, the maximal thrombotic potential appears to be longer than that reported in studies of white subjects (6wk vs 1wk), and it is possible that routine sequential venous duplex ultrasonography 2 weeks after the initial screening test in our population could possibly yield a higher rate of DVT.1, 19 Last, routine pulmonary angiography to detect PE would likely yield a very low incidence of PE with an unnecessary increase in health care costs. With a selected and targeted low-cost protocol on admission to rehabilitation, the DVT incidence is low, and the clinical PE rate is negligible. Early initiation of walking and reduced functional dependency together with nonpharmacologic DVT prevention measures may have appreciable beneficial effects for DVT prevention in our cohort. While no significant predictive factors were found for DVT, increased functional dependency in older non-Chinese patients correlated with elevated D-dimer levels and may contribute to falsely elevated D-dimer levels and prompt increased venous duplex ultrasonography workup in this subgroup. Conclusion Our results confirm that asymptomatic DVT is indeed uncommon as believed among Asians undergoing neurorehabilitation without any particularly susceptible diagnostic group. Traditional clinical risk factors for DVT were not evident in this study, and there were no functional or rehabilitation predictors for DVT. The routine use of venous duplex ultrasonography as a screening tool for DVT for all admissions is not supported by our study, and D-dimer assay coupled with targeted selective venous duplex ultrasonography were adequate to screen for DVT. Further research is required into the cost-effectiveness of such a screening protocol locally and to determine whether there is indeed a late peak in DVT incidence during the rehabilitation phase. Suppliers References 1. 1Kelly J, Rudd A, Lewis R, Hunt BJ. Venous thromboembolism after acute stroke. Stroke. 2001;32:262–267. 2. 2Sachdev U, Teodorescu VJ, Chao M, et al. Incidence and distribution of lower extremity deep vein thrombosis in rehabilitation patients: implications for screening. Vasc Endovascular Surg. 2006;40:205–211. MEDLINE |
CrossRef
3. 3Warlow C, Ogston D, Douglas AS. Venous thrombosis following strokes. Lancet. 1972;1:1305–1306. MEDLINE 4. 4Prandoni P, Lensing AW, Cogo A, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996;125:1–7. MEDLINE 5. 5Fraser D, Moody A, Davidson I, Martel A, Morgan P. Diagnoses of lower limb deep venous thrombosis (DVT): prospective blinded analysis of magnetic resonance direct thrombus imaging. Vessel WMI Mechanics and Morphology; Society of Magnetic Resonance in Medicine. 2000;. 6. 6Kelly J, Rudd A, Lewis RR, Coshall C, Moody A, Hunt BJ. Venous thromboembolism after acute ischaemic stroke: a prospective study using magnetic resonance direct thrombus imaging. Stroke. 2004;35:2320–2326.
CrossRef
7. 7Kelly J, Rudd A, Lewis RR, et al. Screening for proximal deep vein thrombosis after acute ischaemic stroke: a prospective study using clinical factors and plasmas D-dimers. J Thromb Haemost. 2004;1:1321–1326. MEDLINE |
CrossRef
8. 8Harvey RL, Roth EJ, Yarnold PR, Durham JR, Green D. Deep venous thrombosis after stroke. Stroke. 2006;27:1516–1520. MEDLINE 9. 9Akman MN, Cetin N, Bayramoglu M, Isiklar I, Kilinc S. Value of the D-dimer test in diagnosing deep vein thrombosis in rehabilitation inpatients. Arch Phys Med Rehabil. 2004;85:1091–1094. Abstract | Full Text |
Full-Text PDF (49 KB)
|
CrossRef
10. 10Oczkowski WJ, Ginsberg JS, Shin A, Panju A. Venous thromboembolism in patients undergoing rehabilitation for stroke. Arch Phys Med Rehabil. 1992;73:712–716. MEDLINE 11. 11Sioson ER, Crowe WE, Dawson NV. Occult proximal deep venous thrombosis: its prevalence among patients admitted to a rehabilitation hospital. Arch Phys Med Rehabil. 1988;69:183–185. MEDLINE 12. 12Yablon SA, Rock WA, Nick TG, Shere M, McGrath SM, Goodson KH. Prevalence and risk factors in rehabilitation admissions with brain injury. Neurology. 2004;63:485–491. 13. 13Cifu D, Kaelin DL, Wall BE. Deep venous thrombosis: incidence on admission to a brain injury rehabilitation program. Arch Phys Med Rehabil. 1996;77:1182–1185. Abstract |
Full-Text PDF (560 KB)
|
CrossRef
14. 14Meythaler JM, DeVivo MJ, Hayne JB. Cost-effectiveness of routine screening for proximal deep venous thrombosis in acquired brain injury patients admitted to rehabilitation. Arch Phys Med Rehabil. 1996;77:1–5. Abstract |
Full-Text PDF (652 KB)
|
CrossRef
15. 15Gallus AS, Lee LH, Coghlan DW. New aspects of the blood coagulation cascade, anticoagulants and vein thrombosis in Asia. Ann Acad Med Singapore. 2002;31:685–696. MEDLINE 16. 16Sun KK, Wang C, Pang BS, et al. Study on the risk factors of deep venous thrombosis in acute hospitalised stroke patients. Zhonghua Liu Xing Bing Xue Za Zhi. 2004;25:1010–1023. MEDLINE 17. 17Lee LH, Gu KQ, Heng D. Deep vein thrombosis is not rare in Asia: the Singapore General Hospital experience. Ann Acad Med Singapore. 2002;31:761–764. MEDLINE 18. 18Kumar K, Tang KK, Thomas J, Chumpon C. Is post-operative deep vein thrombosis a problem in neurosurgical patients with brain tumours in Singapore?. Singapore Med J. 2002;43:345–349. MEDLINE 19. 19DeSilva DA, Pey HB, Wong MC, Chang HM, Chen CPLH. Deep vein thrombosis following ischaemic stroke among Asians. Cerebrovasc Dis. 2006;22:245–250. MEDLINE |
CrossRef
20. 20Collin C, Wade D. Assessing motor impairment after stroke: a pilot reliability study. J Neurol Neurosurg Psychiatry. 1990;53:576–579. MEDLINE |
CrossRef
21. 21Wright J. The FIM (TM). 2000;http://www.tbims.org/combi/FIM. 22. 22Mosenthal AC, Lavery RF, Addis M, et al. Isolated traumatic brain injury: age is an independent predictor of mortality and early outcome. J Trauma. 2002;52:907–911. MEDLINE 23. 23Lewin W, Marshall TF, Roberts AH. Long term outcomes after severe head injury. BMJ. 1979;2:1533–1538. 24. 24Population and health facts, Singapore 2007. http://www.moh.gov.sg/mohcorp/statistics/aspx. 25. 25White RH, Zhiu Z, Romano PS. Incidence of idiopathic deep vein thrombosis and secondary thromboembolism among ethnic groups in California. Ann Intern Med. 1998;128:737–740. MEDLINE 26. 26Tso SC. Deep venous thrombosis after strokes in Chinese. Aust N Z J Med. 1980;10:513–514. MEDLINE a Department of Rehabilitation Medicine, Tan Tock Seng Hospital, Republic of Singapore b Clinical Research Unit, Tan Tock Seng Hospital, Republic of Singapore  Correspondence to Karen Chua, MBBS, MRCP, FAMS, Dept of Rehabilitation Medicine, Tan Tock Seng Hospital, Singapore, 17 Ang Mo Kio Ave 9, Singapore 569766, Republic of Singapore
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprints are not available from the author. PII: S0003-9993(08)01473-1 doi:10.1016/j.apmr.2008.05.025 © 2008 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT