Background: The venous thromboembolism (VTE) rate in patients with current or recent hospitalization is 330/100,000 person-year vs. 8/100,000 in outpatients. Hence, hospitalists often assess risk and pursue a diagnosis of VTE in their patients. Lower extremity (LE) DVT risk prediction tools such as the Wells score perform poorly on inpatients and hospitalist specific guidelines for the diagnosis of LE DVT on inpatients are lacking. Hence, controversy exists in this workup and it is unknown for which patients bilateral vs. unilateral LE ultrasounds (USs) are appropriate. Vascular medicine and radiology guidelines are ambiguous. Some experts recommend unilateral LE USs only if certain clinical criteria are met (< 60; no cancer; no CHF; no pulmonary disease; ambulatory); others recommend bilateral LE USs exclusively for inpatients. We decided to review our hospitalists’ ordering practice for LE DVT studies to characterize indications and outcomes based on study type (unilateral vs. bilateral).
Methods: Our study was a cross sectional and descriptive in nature. We collected data on consecutive inpatient lower extremity duplex USs orders on Hospitalist patients from 1– 6/2017 and abstracted the ordering indication and study results. This data was merged with administrative discharge abstracts which contained demographics; primary and secondary diagnoses; LOS; and mortality. Our independent variable was duplex type (bilateral vs. unilateral). Our primary outcome variables were indications and study result. We performed bivariate analyses using Chi Square for categorical variables and the Student T test for continuous variables.
Results: During the study period, 312 studies were ordered; 71% bilateral; 13% were positive. Patient demographics are shown in Table 1. Patients with bilateral studies were sicker (mortality of 22% in bilateral vs. 4% in unilateral study patients; p=0.001, see Table 2). The positive rate by type was not different: 12% in bilateral vs. 16% in unilateral studies (p=0.26). However, after exclusion of studies with indication of known VTE and those only finding distal DVTs, the positive rate difference becomes significant, 4.5% bilateral vs. 13% unilateral (p=0.01).
We identified 12 indications used as justification to order the studies. Extremity swelling and leg pain were the most common. The indications with the highest positive study rate were known prior VTE: either a follow-up (43% positive) or a known VTE elsewhere (36% positive). Bilateral studies were more often ordered for patients with known PE (20 studies). Positivity by indication is shown in Table 3. The indications and frequency used broken down by laterality are shown in Table 4.
Conclusions: Almost 1/3 of LE US studies performed by our group were unilateral. This would go against some expert opinion which recommends bilateral studies in all inpatients. Bilateral studies are more often ordered when a patient has a known PE and tends to diagnosis more distal DVTs. Nonspecific / background factors like a hypercoaguable state or respiratory change were poor indications for positivity compared to limb specific swelling, redness, or pain. This suggests something clinical related to the extremity should be a prerequisite. Overall, the positive rate was similar but unilateral studies become significantly more positive when known VTE and distal DVTs are excluded. This is likely related to unilateral exam or history findings. Additional studies are needed to craft evidence based guidelines for the workup of suspected DVT in inpatients.
.png)
.png)