A Prospective Comparison of Bone Marrow Biopsy Adequacy with and without Use of Powered Assistive Drill in the Clinical Setting Versus Interventional Radiology

Background

The adequacy of bone marrow biopsies is most often defined by trephine length. By this definition, adequacy has been associated with the experience of the procedure operator. In small prospective studies, powered assistive drills have also affected adequacy as their use has been shown to yield longer trephine lengths as compared to manual devices. Within our Central Texas academic institution tertiary care center, we have noted the two following recent changes in our bone marrow biopsy practice: routine implementation of the powered assistive drill and the use of interventional radiology for CT-guided bone marrow biopsies of more difficult patients (i.e. obese body habitus, altered anatomy, previous failed attempts, etc.), and those who desire sedation. This study aims to review the impact of powered assistive drill use and referral to interventional radiology on the adequacy of bone marrow biopsies at our institution.

Methods

Consecutive bone marrow biopsies performed at the Baylor Scott & White Health clinical site 1 (CS1), one of the satellite clinics (CS2), and interventional radiology (IR) from January 1, 2011, through December 31, 2013, were eligible for inclusion. Patients less than 18 years of age were excluded. A prospective registry recorded the date of procedure, patient medical record number, patient age, as well as the presence of spicules within the aspirate (yes or no), and trephine length (in mm) determined by pathology. Procedure location and use of the powered assistive drill for each procedure were determined by billing codes. Trephine length was compared between locations and by use of the assistive drill with Wilcoxon two-sample tests.

Results

In total, 888 bone marrow biopsies were performed. 753 were performed in the clinical setting (554 at CS1 and 199 at CS2), and 135 in IR. Overall, 326 biopsies utilized the powered drill and 562 were performed manually. Mean trephine length, independent of assistive drill use, was 10.1 mm, 10.3 mm, and 7.4 mm at CS1, CS2, and IR locations respectively; with a significantly shorter trephine length observed in IR as compared to CS1 (p < 0.0001), and CS2 (p <0.0001). The mean trephine length of biopsies obtained with use of the assistive drill was significantly longer than those performed manually at CS1 (10.8 mm vs 9.3 mm [p = 0.0004]), and in IR (10.6mm vs 5.9 mm [p <0.0001]), respectively. There was no difference in mean trephine length between CS1 and IR with use of the drill (10.8 mm and 10.6 mm [p = 0.8123]). The presence of spicules did not differ with the use of the assistive drill at either location (p = 0.9463). CS2 has not yet implemented the routine use of the powered assistive drill.

Discussion

A statistically significant increase in mean trephine length was observed with use of the assistive drill between CS1 and IR, though the only difference of clinical significance was that observed in IR. With use of the drill, however, the mean trephine lengths between our clinical site and interventional radiology was no longer statistically significant. The use of the assistive drill in IR simply increased trephine length to equal those obtained routinely in the clinical setting. This suggests the use of powered assistive devices, in a patient population for whom CT-guidance is necessary, may increase the likelihood of obtaining a trephine of similar adequacy to those performed on the “ideal” patient at the bedside by experienced operators.