Abstract
Background:
Asymptomatic neutropenia is a common hematology referral, though standardized reference ranges and published clinical outcomes are lacking.
Methods:
In our retrospective analysis, we evaluated demographics, laboratory, and clinical outcomes of adult patients referred to an academic hematology practice for evaluation of neutropenia from 2010–2018. Primary and secondary outcomes included incidence of hematologic disorders and rates of Duffy-null positivity by race, respectively. In a separate analysis, we reviewed absolute neutrophil count (ANC) reference ranges from publicly available AAMC Medical School Member laboratory directories to assess institutional variations.
Results:
In total, 163 patients were included, with disproportionate number of Black patients referred compared to local demographics. Twenty-three percent of patients (n=38) were found to have a clinically relevant hematologic outcome (mean ANC of 0.59 × 109/L), and only 6 were identified with ANC ≥ 1.0 × 109/L. Incidence of hematologic outcomes was lowest among Black patients (p=0.05), and nearly all Blacks who underwent Duffy-null phenotype testing were positive (93%), compared to 50% of Whites (p=0.04). In separate review of laboratory directories, we confirmed wide variation in ANC lower limit of normal (0.91 – 2.40 × 109/L).
Conclusion:
Hematologic disorders were rare in patients with mild neutropenia and among Blacks, highlighting the need to standardize hematological ranges representative of non-White communities.
Keywords: neutropenia, outcome assessment, health care, ethnic and racial minorities, minority health, healthcare disparities, hematologic diseases
Introduction
Incidentally noted neutropenia is a common reason for hematology referral. However, published guidelines directing workup and management are lacking, and clinical outcomes to guide prognostication remain largely unknown (1). Though references ranges are essential for the accurate interpretation of test results and clinical decision making, they generally vary among institutions and tend to be laboratory specific (2–4). Furthermore, referrals for asymptomatic neutropenia highlight non-inclusive hematological ranges that may inadvertently contribute to inequitable care particularly among racial minorities (5).
One such entity that highlights this systemic problem, known as benign ethnic neutropenia (BEN), is a term used to describe an ANC <1.5 × 109/L, in patients with no increased risk of infection and in whom no apparent cause of neutropenia is found by standard workup (6 – 9). Though BEN has historically been linked to ethnicity given its higher prevalence among persons of African ancestry, we now know that the Duffy-null [Fy(a-b−)] phenotype correlates with a higher sensitivity and specificity for this diagnosis (10). The absence of Duffy antigens, as in the Fy(a-b−) phenotype, prevents malaria merozoites from invading red blood cells (RBC), which explains why people living in geographic areas with high rates of malaria more commonly harbor this phenotype (11–13). As such, studies have proposed replacing BEN with more accurate terminology such as Duffy-null associated neutrophil count (DANC) (14).
Established hematological reference ranges that are historically built on “normal” values among those of European descent disproportionately impact individuals who harbor the Duffy-null phenotype, most of whom are of Black race, leading to inaccurate clinical decision making and excessive diagnostic testing (15–17). Gaining a comprehensive understanding of the demographics and clinical outcomes of patients with asymptomatic neutropenia, as well as variations in institutional references ranges, has the potential to change diagnostic and treatment paradigms. In this study, our aims were two-fold: (1) to describe the demographic and clinical outcomes of patients referred to a hematologist for evaluation of asymptomatic neutropenia at our single institution, and (2) to evaluate the variations in ANC reference ranges among publicly available laboratory directories in North America.
Methods
Study population
In this single-center retrospective analysis, we identified all adult patients (≥18 years) referred to hematology at Oregon Health & Science University (OHSU) for asymptomatic neutropenia over an 8-year period (2010 – 2018). This study was approved by the OHSU Institutional Review Board. Patients identified for screening through database query then underwent manual chart review. Asymptomatic neutropenia was defined as an ANC below the internal reference range of 1.8 × 109/L (18), without history of recurrent or abnormal infections. Patients who had other cytopenias were included in the study as long as the primary referral reason was asymptomatic neutropenia. Patients with signs or symptoms of infection or malignancy, including fever, night sweats and/or unintentional weight loss, at the time of referral were excluded from the study. Unknown race or race groups with fewer than 5 events were excluded from analysis. Baseline characteristics, including pertinent demographics, ANC at the time of referral, and number of tests performed for workup, including bone marrow biopsy and RBC antigen testing, were ascertained through individual chart review.
Reference range acquisition
Using a list of Association of American Medical Colleges (AAMC) Medical School Members (19), we then reviewed and recorded adult ANC reference ranges from online laboratory directories to evaluate North American variations. Institutions were excluded from data collection if laboratory directories were not publicly available through online search.
Outcome definitions
Our primary outcome of interest was the incidence of a clinically significant hematologic disorder among all patients referred for asymptomatic neutropenia, defined as a newly diagnosed non-malignant, premalignant, or malignant hematologic condition. Hematologic outcomes were compared to those found to have an underlying attributable non-hematologic condition (including viral, nutritional, and drug-induced) and those with no readily identifiable cause. Our secondary outcome included rates of Duffy-null positivity among patients who underwent RBC antigen testing. Separate subgroup analyses stratifying outcomes by ANC severity at the time of referral and by race were also conducted.
Statistical methods
Descriptive statistics were used to summarize baseline and outcome characteristics, with categorical variables presented as number and percentage and numerical variables presented as mean and standard deviation (SD). Proportions were compared with a Fisher’s exact test and continuous variables were compared with a Mann Whitney or a Kruskal Wallis after checking for normality with Shapiro Wilk test. The statistical significance level was set at ≤ 0.05.
Results
Cohort characteristics
The final study population comprised of 163 adult patients with a median age of 55 years (range: 19 – 96). Ten percent of patients identified as Asian (n=17), 16% as Black (n=26), and 74% as White (n=120) (Table 1), with a disproportionate number of Black patients referred compared to local population demographics in Portland, Oregon (9% Asian, 6% Black, and 75% White) (20). Black patients were younger compared to White and Asian counterparts, though not statistically significant (p = 0.09).
Table 1.
Patient demographics, laboratory values, and outcomes stratified by race
| Asian (N=17) |
Black (N=26) |
White (N=120) |
Overall (N=163) |
|
|---|---|---|---|---|
|
| ||||
| Age | ||||
| Median [IQR] | 58 [28–78] | 47 [19–86] | 57 [24–96] | 55 [19–96] |
| Sex | ||||
| Female | 14 (82.4%) | 10 (38.5%) | 77 (64.2%) | 101 (62.0%) |
| Male | 3 (17.6%) | 16 (61.5%) | 43 (35.8%) | 62 (38.0%) |
| RBC Antigen Test | ||||
| Not tested | 16 (94.1%) | 12 (46.2%) | 112 (93.3%) | 140 (85.9%) |
| Negative | 0 (0%) | 1 (3.8%) | 4 (3.3%) | 5 (3.1%) |
| Duffy-null Fy(a-b−) | 1 (5.9%) | 13 (50.0%) | 4 (3.3%) | 18 (11.0%) |
| ANC at Referral (×109/L) | ||||
| Mean (SD) | 1.05 (0.44) | 0.99 (0.25) | 0.88 (0.43) | 0.91 (0.41) |
| Outcomes | ||||
| AML | 0 (0%) | 0 (0%) | 2 (1.7%) | 2 (1.2%) |
| Autoimmune neutropenia | 1 (5.9%) | 0 (0%) | 4 (3.3%) | 5 (3.1%) |
| Low-grade B cell lymphoma | 0 (0%) | 0 (0%) | 1 (0.8%) | 1 (0.6%) |
| CLL | 0 (0%) | 0 (0%) | 2 (1.7%) | 2 (1.2%) |
| Congenital neutropenia | 0 (0%) | 0 (0%) | 2 (1.7%) | 2 (1.2%) |
| Hairy cell leukemia | 0 (0%) | 0 (0%) | 1 (0.8%) | 1 (0.6%) |
| Hodgkin’s lymphoma | 0 (0%) | 0 (0%) | 1 (0.8%) | 1 (0.6%) |
| ITP | 0 (0%) | 0 (0%) | 3 (2.5%) | 3 (1.8%) |
| LGL leukemia | 0 (0%) | 1 (3.8%) | 5 (4.2%) | 6 (3.7%) |
| MDS | 1 (5.9%) | 0 (0%) | 11 (9.2%) | 12 (7.4%) |
| MGUS | 0 (0%) | 1 (3.8%) | 1 (0.8%) | 2 (1.2%) |
| PNH | 0 (0%) | 0 (0%) | 1 (0.8%) | 1 (0.6%) |
| Nutritional deficiency | 0 (0%) | 0 (0%) | 2 (1.7%) | 2 (1.7%) |
| Drug-induced | 0 (0%) | 0 (0%) | 2 (1.7%) | 2 (1.2%) |
| Viral | 1 (5.9%) | 0 (0%) | 3 (2.5%) | 4 (2.5%) |
| No clinical outcome | 14 (82.4%) | 24 (92.3%) | 79 (65.8%) | 117 (71.8%) |
Abbreviations: IQR, interquartile range; RBC, red blood cell; ANC, absolute neutrophil count; AML, acute myeloid leukemia; CLL, chronic lymphocytic leukemia; ITP, immune thrombocytopenic purpura; LGL, large granular lymphocytic; MDS, myelodysplastic syndrome; MGUS, monoclonal gammopathy of undetermined significance; PNH, paroxysmal nocturnal hemoglobinuria
The average ANC at referral was 0.91 × 109/L (SD = 0.41), which did not differ significantly by race (p=0.11). When stratified by degree of severity, 23 patients (14%) had ANC < 0.50 × 109/L (severe), 59 patients (36%) had ANC ≥ 0.50 to < 1.00 × 109/L (moderate), and 81 patients (50%) had ANC ≥ 1.00 × 109/L (mild), of which 3 had ANC > 1.5 × 109/L. The average number of tests performed at referral, including repeat complete blood cell count with differential, was 7. Bone marrow biopsies were obtained in 27 patients with an average ANC of 0.47 × 109/L and RBC antigen testing was obtained in 23 patients.
Incidence of clinically significant hematologic outcomes
In total, 38 patients (23%) were found to have clinically significant hematologic outcomes with mean ANC of 0.59 × 109/L: myelodysplastic syndrome (MDS) (n=12), large granular lymphocytic (LGL) leukemia (n=6), autoimmune neutropenia (n=5), immune thrombocytopenic purpura (ITP) (n=3), acute myeloid leukemia (AML) (n=2), chronic lymphocytic leukemia (CLL) (n=2), monoclonal gammopathy of undetermined significance (MGUS) (n=2), congenital neutropenia (n=2), hairy cell leukemia (n=1), low grade B cell lymphoma (n=1), Hodgkin’s lymphoma (n=1), and paroxysmal nocturnal hemoglobinuria (PNH) (n=1) (Table 1). Eight patients were found to have an attributable non-hematologic condition, including viral (n=4), nutritional deficiency (n=2), and drug-induced (n=2). Resolution of neutropenia occurred in 30% of patients.
Hematologic outcomes by race and ANC
Rates of hematologic outcomes differed by race, with only 8% of Black patients (n=2) found to have a clinically significant hematologic diagnosis (MGUS and LGL leukemia), compared to 12% of Asians (n=2), and 28% of Whites (n=34) (p=0.05) (Figure 1). Of the two Black patients found to have a clinically significant hematologic diagnosis, one patient was found to have LGL leukemia with an ANC of 1.0 × 109/L and the other patient was found to have MGUS with a referral ANC of 0.8 × 109/L. In the patient with MGUS, neutropenia was felt to be an incidental finding and not attributed to MGUS. When stratified by ANC severity, the majority of patients were referred for mild neutropenia. ANC severity varied significantly by race, with no Black patients referred with an ANC < 0.50 × 109/L, whereas 12% of Asians and 18% of Whites had severe neutropenia (p=0.03). Finally, among patients with an ANC ≥ 1.0 × 109/L, only 6 patients were found to have an underlying hematologic disorder, of which only 2 were malignant conditions: LGL leukemia (n=1), MDS (n=1), ITP (n=3), and autoimmune neutropenia (n=1). Both patients with malignant conditions (LGL and MDS) had an ANC of 1.0 × 109/L at time of referral. The patient with LGL leukemia was asymptomatic at the time of referral with only mild anemia of 12.7g/dL. In the patient with MDS, no other cytopenias were detected aside from a mildly low total white cell count of 2.65 × 109/L.
Figure 1.
(A). Differences in outcome* by race, (B). Differences in Duffy-null positivity by race, and (C). Differences in ANC severity** by race
Abbreviations: ANC, absolute neutrophil count; RBC, red blood cell
*Outcomes were stratified by none identified, non-hematologic (including viral, nutritional deficiency, or drug-induced), or clinically significant hematologic disorders (including non-malignant, pre-malignant, and malignant hematologic disorders)
**Severity categories were defined as: mild = ANC ≥ 1.00 × 109/L, moderate = ANC ≥ 0.50 to < 1.00 × 109/L, and severe = < 0.50 × 109/L
Duffy-null testing
Among Black patients who underwent testing for Duffy antigen expression (n=14), nearly all patients were positive for the Duffy-null phenotype [Fy(a-b−)], with the exception of one patient who was Fy(a-b+). In contrast, in Non-Black patients tested (n=8), only 50% (n=4) were positive for this phenotype (p=0.04) (Figure 1).
ANC Reference Ranges
Upon review of the 171 AAMC Medical School Members, we identified 33 publicly available adult ANC reference ranges from online laboratory directories (Supplemental Table 1). Among these institutions, we found variations in the ANC lower limit of normal (LLN), ranging from 0.91 – 2.40 × 109/L with an average LLN of 1.75 × 109/L (Figure 2). When organized by geographic region, we did not find significant variation (p=0.31). However, 4 institutions included in our analysis reported discordant references ranges by sex.
Figure 2.
(A). Absolute neutrophil count (ANC) by lower limit of normal among institutions with publicly available reference ranges (n=33) and (B). stratified by region in North America. ANC values for four institutions that reported discordant ANC values by sex are represented separately.
Discussion
In summary, we identify several important conclusions from this study. First, we found that nearly all patients with an ANC ≥ 1.0 × 109/L (97%) had no identified underlying malignant hematologic disorder detected by standard workup. Second, in comparison to the local demographics, Black individuals were disproportionately referred to a hematologist for asymptomatic neutropenia, were significantly more likely to harbor the Duffy-null phenotype, and less likely to have a clinically significant hematologic outcome. Finally, we confirmed that laboratory ANC references ranges vary among institutions throughout North America highlighting significant clinical implications in the overall approach to the initial workup and management of asymptomatic neutropenia. Furthermore, our findings shed light on racial inequities within hematology and a critical need to standardize hematological references ranges that include demographic variants that are more representative of the community they serve.
We additionally propose that routine RBC antigen typing could potentially reduce the financial and emotional burden of an extensive workup for a clinically insignificant variant. This aligns with recently released recommendations from the American Society of Clinical Pathology (ASCP) in collaboration with the American Society of Clinical Laboratory Science (ASCLS) to not perform an extensive work up in otherwise healthy individuals with neutropenia who are of African or Middle Eastern ancestry prior to Duffy-null phenotype testing (21). This evidence-based suggestion, published through the Choosing Wisely campaign, corroborates the findings of our study, other reports (22), and our own experience.
Our study has several strengths, including careful identification of the study cohort and thorough evaluation of clinical outcomes through manual chart review by hematologists. Our study period concluded in 2018, ensuring a minimum of 3 years of follow up by the time of manual chart review. Limitations include limited sample size and the retrospective nature of data collection. Additionally, we acknowledge that not every patient underwent invasive testing with bone marrow biopsy, a gold standard for definitive diagnosis, though reassuringly no identified hematologic transformation was identified through routine follow up. Lastly, only a minority of AAMC Medical School Members had publicly available ANC reference ranges, and as such these findings may not fully represent the North American ANC distribution. Furthermore, we did not expand our review of laboratory reference ranges to include other continents, though we suspect that this systemic problem exists worldwide and hope that our findings trigger additional investigation to inform change.
Conclusion
In summary, mild asymptomatic neutropenia with an ANC ≥ 1.0 × 109/L was a benign finding in nearly all patients in our study. The vast majority were not found to have an underlying hematologic disorder, suggesting that an ANC threshold of < 1.0 ×109/L may be more appropriate to trigger further hematologic workup. Upfront RBC antigen testing may assist clinicians in determining if further workup for neutropenia is necessitated, particularly among Black patients. Finally, we identify a critical need to develop and standardize ANC references ranges to limit care inequities that predominantly affect non-White communities.
Supplementary Material
Supplemental Table 1. Absolute neutrophil count (ANC) range by institution through review of publicly available Association of American Medical Colleges (AAMC) Medical School Member laboratory directories
What is the NEW aspect of your work?
Through review of AAMC Medical School Member laboratory directories, we found that the absolute neutrophil count (ANC) lower limit of normal varies broadly across institutions (0.91 – 2.40 × 109/L), highlighting non-inclusive ranges that may contribute to inequitable care among racial minorities.
What is the CENTRAL finding of your work?
In comparison to the local demographics, Black individuals were disproportionately referred to a hematologist for asymptomatic neutropenia, were significantly more likely to harbor the Duffy-null phenotype, and less likely to have a clinically significant hematologic outcome.
What is (or could be) the SPECIFIC clinical relevance of your work?
Standardized hematological ranges representative of non-White communities and early incorporation of routine Duffy-null phenotype testing among patients of African or Middle Eastern ancestry may reduce the financial and emotional burden of an extensive workup for a clinically insignificant variant.
Funding Sources:
JJS is supported by the National Heart Lung and Blood Institute; National Institute of Health (HL151367); RM is supported by the National Heart Lung and Blood Institute; National Institute of Health (F30HL163918). KLM is supported by the American Society of Hematology Research Training Award for Fellows.
Footnotes
Conflict of Interest: JJS reports receiving consulting fees from Aronora INC. The remaining authors report no conflicts of interest.
Ethical Approval: This retrospective chart review study involving human participants was in accordance with the ethical standards of the OSHU Institutional Review Board committee who approved this study.
Data Sharing Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1.Valent P Low blood counts: immune mediated, idiopathic, or myelodysplasia. Hematology 2010, the American Society of Hematology Education Program Book. 2012;485–491. [DOI] [PubMed] [Google Scholar]
- 2.Nah EH, Kim S, Cho S, Cho HI. Complete Blood Count Reference Intervals and Patterns of Changes Across Pediatric, Adult, and Geriatric Ages in Korea. Ann Lab Med. 2018. Nov;38(6):503–511. doi: 10.3343/alm.2018.38.6.503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Katayev A, Balciza C, Seccombe DW. Establishing reference intervals for clinical laboratory test results: is there a better way? Am J Clin Pathol. 2010. Feb;133(2):180–6. doi: 10.1309/AJCPN5BMTSF1CDYP. [DOI] [PubMed] [Google Scholar]
- 4.Ozarda Y, Higgins V, Adeli K. Verification of reference intervals in routine clinical laboratories: practical challenges and recommendations. Clin Chem Lab Med. 2018. Dec 19;57(1):30–37. doi: 10.1515/cclm-2018-0059. [DOI] [PubMed] [Google Scholar]
- 5.Merz LE, Achebe M. When non-Whiteness becomes a condition. Blood. 2021. Jan 7;137(1):13–15. doi: 10.1182/blood.2020008600. [DOI] [PubMed] [Google Scholar]
- 6.Ortiz MV, Meier ER, Hsieh MM. Identification and clinical characterization of children with benign ethnic neutropenia. J Pediatr Hematol Oncol. 2016; 38 ( 3 ): e140 – e143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Shoenfeld Y, Weinberger A, Avishar R, et al. Familial leukopenia among Yemenite Jews. Isr J Med Sci. 1978; 14 ( 12 ): 1271 – 1274. [PubMed] [Google Scholar]
- 8.El Hadidi KT, Medhat BM, Abdel Baki NM, et al. Characteristics of systemic lupus erythematosus in a sample of the Egyptian population: a retrospective cohort of 1109 patients from a single center. Lupus. 2018; 27 ( 6 ): 1030 – 1038. [DOI] [PubMed] [Google Scholar]
- 9.Haddy TB, Rana SR, Castro O. Benign ethnic neutropenia: what is a normal absolute neutrophil count? J Lab Clin Med. 1999; 133 ( 1 ): 15 – 22. [DOI] [PubMed] [Google Scholar]
- 10.Howes RE, Patil AP, Piel FB, et al. The global distribution of the Duffy blood group. Nat Commun. 2011; 2 ( 1 ): 266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gunalan K, Lo E, Hostetler JB, Yewhalaw D, Mu J, Neafsey DE, Yan G, Miller LH. Role of Plasmodium vivax Duffy-binding protein 1 in invasion of Duffy-null Africans. Proc Natl Acad Sci U S A. 2016. May 31;113(22):6271–6. doi: 10.1073/pnas.1606113113. Epub 2016 May 17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Palmblad J, Höglund P. Ethnic benign neutropenia: a phenomenon finds an explanation. Pediatr Blood Cancer. 2018; 65 ( 12 ): e27361. [DOI] [PubMed] [Google Scholar]
- 13.Martins ML, da Silva AR, Santos HC, et al. Duffy blood group system: new genotyping method and distribution in a Brazilian extra-Amazonian population. Mol Cell Probes. 2017; 35 : 20 – 26. [DOI] [PubMed] [Google Scholar]
- 14.Merz LE, Story CM, Osei MA, Ren S, Park HS, Jolley K, Yefidoff Freedman R, Neuberg D, Smeland-Wagman R, Kaufman R, Okam Achebe M. Absolute Neutrophil Count by Duffy Status Among Healthy Black and African American Adults. Blood Adv. 2022. Aug 22:bloodadvances.2022007679. doi: 10.1182/bloodadvances.2022007679. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hershman D, Weinberg M, Rosner Z, Alexis K, Tiersten A, Grann VR, Troxel A, Neugut AI. Ethnic neutropenia and treatment delay in African American women undergoing chemotherapy for early-stage breast cancer. J Natl Cancer Inst. 2003. Oct 15;95(20):1545–8. doi: 10.1093/jnci/djg073. [DOI] [PubMed] [Google Scholar]
- 16.Lerma P, Besa E, Klease C, Joseph R, Ozer H. Benign neutropenia patients are subjected to frequent treatment delays when current treatment guidelines are used [abstract 1627]. Proc ASCO 1999; 18 : 422a. [Google Scholar]
- 17.Manu P, Sarvaiya N, Rogozea LM, Kane JM, Correll CU. Benign Ethnic Neutropenia and Clozapine Use: A Systematic Review of the Evidence and Treatment Recommendations. J Clin Psychiatry. 2016. Jul;77(7):e909–16. doi: 10.4088/JCP.15r10085. [DOI] [PubMed] [Google Scholar]
- 18.CBC with Differential. OHSU Lab Services. Retrieved November 22, 2022, from https://www.ohsu.edu/lab-services/cbc-differential [Google Scholar]
- 19. https://members.aamc.org/eweb/DynamicPage.aspx?site=AAMC&webcode=AAMCOrgSearchResult&orgtype=Medical%20School .
- 20.U.S. Census Bureau (2021). QuickFacts, Portland City, Oregon. Retrieved from [https://www.census.gov/quickfacts/portlandcityoregon]
- 21.Choosing wisely. American society for clinical pathology and American society for clinical laboratory science. Five things Physicians and patients should question Choosingwisely.org. Nov 2022 [Google Scholar]
- 22.Chaitowitz Mark, Van Metre Laura, and Goel Swati. “Hematology referrals for leukopenia in the bronx rarely result in clinically significant diagnoses: first steps towards and evidence-based referral guideline.” (2016): 4743–4743 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Table 1. Absolute neutrophil count (ANC) range by institution through review of publicly available Association of American Medical Colleges (AAMC) Medical School Member laboratory directories
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.