Introduction: Bone marrow aspirate and biopsy (BMP) is the most common procedure used by the hematologist to evaluate benign and malignant conditions. During the last decade, several factors associated with this procedure such as higher rates of CT-guidance use, nurse practitioners trained on this skill and the use of drills has increased. It is unclear whether any of these factors have an impact on the quality of the BMP specimens. Hence, we undertook a retrospective study among patients that underwent BMPs in a large academic hospital to evaluate the factors associated with optimal samples.

Aim: To establish clinico-pathological features associated with optimal BMP specimens based on aspirate and trephine quality.

Methods: Adult patients who underwent BMP at Montefiore Medical Center between September 2016 and September 2017 were identified using data from the billing department. Data regarding demographics, indications for the procedure, technique (manual vs. drill), provider (medical staff vs. nurse practitioners), approach (bedside vs. CT guided) and trephine length was obtained by chart review. Stored BMP aspirates and trephines were reviewed by two expert heme-pathologist. Trephine samples were classified as optimal, adequate or inadequate based on fragmentation, aspiration artifact, crush and hemorrhage. Aspirate samples were classified as sufficient or insufficient based on spicular number and quality, cell viability and hemodilution. Overall, BMP samples were considered optimal if there was a combination of an optimal core with sufficient aspirate and inadequate if the core was inadequate and the aspirate insufficient. A combination of a suboptimal core with sufficient aspirate or vice-versa was considered adequate. Clinical characteristics and core length were compared among these three groups using descriptive statistics. A multivariate analysis was performed to identify factors associated with optimal BMP specimens.

Results: A total of 346 BMPs were obtained at Montefiore Medical Center during a 12-month period. Clinical and pathology data was reviewed for 145 cases that were included in the analysis. The median age was 65 years (IQ: 56.5-71.5 years), 53.8% were male and the median BMI was 26.4 kg/m2 (22.5-30.8 kg/m2). The main clinical indication for BMPs was evaluation for cytopenias (43, 29.7%). A manual technique was used in 54.6% while 45.4% were drill-based procedures. BMPs were done at bedside in 55.2% while CT-guidance was used in 44.8% of cases. The procedure was performed by a hematologist attending, fellow or nurse practitioner (NP) in 90 (62.1%), 22 (15.2) and 33 (22.8%) cases, respectively.

Overall, optimal specimens were encountered in 27 cases (18.6%), 106 (73.1%) were adequate and 10 (6.9%) were inadequate. The main outcome was missing for two cases for which only an aspirate was submitted. There were no significant differences in age, gender, BMI, provider or technique among the three groups. (Table 1). When compared to attending physicians, fellows and NPs had a lower rate of optimal specimens (29.6% vs. 25.9% vs 44.4%, p=0.02) while the rate of optimal and adequate samples was similar among fellows and NPs (29.6% vs 25.9% for optimal and 13.1% vs 18.7% for adequate), respectively. The rate of optimal BMPs was significantly higher in the bedside group than the radiology group (88.9% vs 11.1%, p=<0.001). Samples with optimal characteristics had a median length of 13 mm, while samples that were graded as adequate and inadequate had median lengths of 9 and 7 mm respectively. In a multivariate logistic regression analysis, BMP approach was the only variable associated with an optimal specimen (OR: 0.08 95% IC:0.02-0.32, p<0.001).

Conclusion: Optimal BMP specimens are encountered in 18.6% of cases. BMP approach (bedside vs CT-guided) was the only factor associated with an optimal specimen. Patients that underwent BMP by a bedside approach had a 20% higher probability of yielding an optimal sample. Based on this data, BMP by bedside should continue to be the standard of care while CT-guided procedures should be reserved to specific cases only.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution