Harmonization of interpretative comments in laboratory hematology reporting: the recommendations of the Working Group on Diagnostic Hematology of the Italian Society of Clinical Chemistry and Clinical Molecular Biology (WGDH-SIBioC)

Introduction

The main goal of harmonizing laboratory testing is to contribute to the quality of patient care and ultimately ameliorate patient outcome. The focus of our profession has long been centered on harmonizing and standardizing processes within the close environment of clinical laboratories, whilst less effort has been catalyzed for improving activities developing outside the physical boundaries of the laboratory [1].

Several lines of evidence suggest that harmonization in laboratory diagnostics should be a much broader enterprise than being merely focused on analytical standardization, thus embracing all the different aspects of the total testing process included in the clinical-laboratory interface [2], [3]. The responsibilities of clinical pathologists should henceforth be broadened to the post-analytic phase of testing, by improving test interpretation and/or recommending additional diagnostic investigations, thus helping hospital physicians and general practitioners to better use test results [4], [5].

The possible misinterpretation of test results carries a tangible risk of inappropriate clinical management, thus ultimately contributing to generating adverse patient outcomes. Thus, it is the responsibility of laboratory professionals to identify the possible drawbacks in laboratory testing and in planning suitable strategies for harmonization [6]. There is a compelling need to improve harmonization in the post-analytical phase of the total testing process, and laboratory specialists can play a pivotal role in this process by educating all stakeholders to help enhancing the quality of this phase of the testing process. One of the most effective strategies encompasses local interventions aimed at providing clinicians with evidence-based and harmonized interpretative comments, so that laboratory reports can be more straightforward and user-friendly [2], [5].

The complete blood cell count (CBC) and leucocyte differential are among the most frequently requested clinical laboratory tests. These analyses can now be performed with fully-automated hematological analyzers. The remarkable technological advances, culminated with the introduction of innovative technologies for cellular analysis, along with the constant evolution of software programs have both contributed to the development and commercialization of a new generation of highly-sophisticated hematological instrumentation, which are now characterized by increased throughput, improved analytic efficiency and the possibility to produce a much larger volume of data [7]. Irrespective of this considerable progress, the morphological analysis of peripheral blood cells (PB) by microscopic revision of appropriately stained blood smears remains crucial even in an era increasingly characterized by molecular diagnostic testing [8], [9], [10].

The inaccurate interpretation of morphological findings and the use of different terminology to describe cellular abnormalities are the leading drawbacks in morphological analysis of peripheral blood smears, finally contributing to insufficient harmonization of this important diagnostic tool [9], [10]. Reliable evidence suggests that the harmonization of this activity is still plagued by heterogeneous reporting practices, and that a different terminology is used for describing blood cell abnormalities, as reflected by the existence of many different recommendations endorsed by different scientific societies [8], [9], [10]. Although the International Council for Standardization in Haematology has published official guidelines for classifying blood film abnormalities and standardizing data reporting as an essential part of the good laboratory and clinical practice [10], local nomenclature adaptations can be still consented, thus allowing a certain degree of flexibility for describing hematological parameters [11], [12], [13].

The SIBioC Italian project for “harmonization of interpretative comments in the laboratory hematology report”

In 2014, the Working Group on Diagnostic Hematology of the Italian Society of Clinical Chemistry and Clinical Molecular Biology (WGDH-SIBioC) has conducted a survey among the Italian laboratories. The WGDH-SIBioC represents a section of the Italian laboratory hematology professionals. It consists of 21 members who are representative of the national context in terms of healthcare facilities. Participation is voluntary, but all members are enrolled according to their degree of personal qualifications, skill and experience in hematology.

The results of this survey showed a remarkable heterogeneity on the use of interpretative comments in laboratory hematology [11]. Therefore, the WGDH-SIBioC was persuaded to initiate a project on “harmonization of interpretative comments in the laboratory hematology report”, aimed to identify the most appropriate comments and propose a standardized reporting system, of which this paper represents an update [12].

After reaching widespread consensus, the WGDH-SIBioC has revised and integrated the comments of 13 Italian volunteer laboratories. In 2016, the Working Group on Diagnostic Hematology of the Italian Society of Clinical Chemistry and Clinical Molecular Biology (WGDH SIBioC) published its first consensus based recommendation for interpretative comments in laboratory hematology reporting whit the purpose of evaluating comments and the aim to (a) reducing their overall number, (b) standardizing the language, (c) providing information that could be easily comprehended by clinicians and patients, (d) increasing the quality of the clinical information, and (e) suggesting additional diagnostic tests when necessary. This paper represents a review of the recommendations of the former document [12]. These laboratories where selected by the coordinator of WGDH-SIBioC from among the members in this group. These laboratories were sufficiently representative of the national situation in terms of healthcare facilities (i.e. both general and university hospitals were included), case mix, workload and geographic distribution. The workload of these laboratories ranges between 500 and 2500 samples per day. The rate of reviewed blood smears ranges between 0.5% and 20% (Median=5, 95% Confidence Interval=5 to 5) on the basis of the technology used, of the blood smear review criteria adopted in each laboratory, and its particular “case mix”.

Overall, 423 comments were collected, 280 of which pertained to white blood cells (WBC) abnormalities (Abn-WBC), 78 to red blood cells (RBC) abnormalities (Abn-RBC) and 13 to platelet (PLT) abnormalities (Abn-PLT). Fifty-two additional comments were referred to non-cell specific abnormalities or to the pre-analytical phase. Similar comments (i.e. with overlapping significance but different terminology) were initially merged into a single comment. The comments were then revised, using a “consensus opinion” [14], and considering the following aspects:

1. Irrespective of advances in laboratory technology and automation, the morphological evaluation of PB still provides high-value clinical information. Nevertheless, the abnormalities found in blood smears are often non-diagnostic and necessitate more specific diagnostic investigations [15], [16].

2. The clinical information available to the laboratory is often scarce, so making the clinical interpretation of blood smear abnormalities rather challenging [5].

3. Laboratory test results can often be misinterpreted. Published data highlights that the post-post analytical phase (i.e. test interpretation and clinical actions undertaken), remains extremely critical [5], [17], [18].

4. Patients have increasing access to medical information and laboratory tests results. Therefore, interpretive comments should not contain clinically unproven or unreasonably alarming statements [5], [13], [17], [18].

5. State of art in this field [9], [10], [11], [13], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49].

6. Laboratory professionals must be engaged in improving a specific culture in support of diagnostic reasoning [5], [43], [50].

7. The diagnostic process involves a large number of health care professionals, including general practitioners and laboratory professionals. Improving diagnostic efficiency requires a team-based approach for improving diagnostic accuracy and efficiency [44].

The interpretive comments generated according to the previously described strategy were then discussed and validated in a final “consensus conference” of the WGDH-SIBioC [12], [14]. In order to comply with all these different aspects, the WGDH-SIBioC has hence developed its recommendations focusing on two different aspects, the former pertaining the text of the comments (so including standardized criteria for selecting and including comments in the laboratory report), and the latter related to the clinical significance of these comments, which should be clarified to, and agreed with, physicians through organization of official meetings, audits or broadcasts.

After publication of the 2016 revision of the World Health Organization (WHO) classification of hematologic malignancies, the WGH-SIBioC then decided to revise and update the comments. The main goals of this revision were: (a) reduce their number, (b) standardize the language, (c) provide information that could be easily comprehended by clinicians and patients, (d) ensure a correct and clinically valuable interpretation of laboratory results and (e) suggest additional diagnostic investigations, when necessary [12], [13], [15], [16], [22], [23], [24].

The final outcome of WGDH-SIBioC was the publication of a consensus agreement, taking into account literature data, individual practices and agreement among the WG members.

Proposals of comments

The CBC and the microscopic revision of blood smear are the mainstays of laboratory hematology. Accurate preparation and staining of blood films, along with experience and skill of laboratory professionals, are additional aspects needed for generating reliable interpretative comments [5], [8], [9], [10], [41], [42], [43]. Therefore, the comment should not be automatically released by the laboratory information system and it should really convey a substantial clinical value [13]. The comments of the WGDH-SIBioC, revised and updated according to the new WHO 2016 classification of hematologic malignancies, are shown in Tables 1 and 2 [22], [24]. A brief description of the various abnormalities of each cell lineage is also provided in the following sections of this article.

Qualitative abnormalities of lymphoid cells

Lymphocytes may display remarkable morphological variations as a consequence of different immunological triggers which characterize both inflammatory, infectious diseases (mainly of viral etiology), or neoplastic diseases (especially leukemias, lymphomas and other lymphoproliferative disorders), which can then be reflected by the presence of lymphocytes with peculiar morphological abnormalities [10], [13], [15], [20], [21].

The WGDH-SIBioC recommends that the definition of “reactive lymphocytes” should be used to describe lymphocytes only when benign etiology is almost certain, whilst the definition of “atypical lymphocytes”, with an accompanying description of cells, should be used to describe lymphocytes with likely malignant or clonal etiology. When doubts remain about the reactive or neoplastic origin of these large cells, especially in the absence of lymphocyte-associated polymorphism, or in the presence of a persistent lymphocytosis without clinically significant morphological abnormalities, caution must be used and the specific comment “immunophenotyping is recommended” may be added [12], [13].

Atypical lymphocytes displaying morphological features of malignancy such as prolymphocytes, hairy and lymphoma cells (i.e. in patients with Sezary syndrome or marginal lymphoma), should be counted within a separate population. These morphological characteristics need to be confirmed with immunophenotyping or molecular biology before definitive classification [20], [21], [22], [23], [27], [28], [29].

The use of the term “atypical lymphocytes” emphasizes the limited diagnostic value of morphological assessment for diagnosing lymphoproliferative disorders, as the definitive diagnosis of these conditions can only be made in combination with phenotypic, genetic, molecular and clinical features [10], [15], [20], [21], [22], [23].

Quantitative or qualitative abnormalities in myeloid cells

Neutrophil abnormalities, potentially needing an accompanying comment, are summarized in Table 1. These comments may be related to quantitative (e.g. severe neutropenia alone) and/or qualitative abnormalities. As regards circulating immature granulocytes, the WGDH-SIBioC recommends a detailed classification into promyelocytes, myelocytes and metamyelocytes [10], [13], [21], [23], [24].

The term “dysplastic changes of granulocytes” can be used to characterize morphologically abnormal granulocytes, usually observed in concomitance with impaired cell development and maturation. Some examples of dysplasia include unusually large or small cells, abnormalities of nuclear lobulation or chromatin density (nuclear hyposegmentation or hypersegmentation), cytoplasmic abnormalities including hypogranulation, hypergranulation and abnormal granulation (i.e. large fused granules, Auer bodies) [10], [15], [23], [24], [30].

Blasts

Regardless of the overall number, the presence of blasts must always be reported in absolute value and/or percentage [8], [9], [10], [23], [24], [31]. The WGDH-SIBioC also recommends adding the comment “presence of blasts >0 and <1%” to the laboratory report because the imprecision of a single count is dramatically high when the percentage of blasts is <1%, whilst blasts counting remains critical for accurate classification of myelodysplastic syndromes [24]. Notably, a considerable morphologic overlap exists between lymphoblasts and some types of undifferentiated or minimally differentiated myeloblasts, so that morphological analysis alone may not be sufficiently accurate for making a definitive diagnosis. Immunophenotyping is always necessary for reaching a more accurate diagnostic classification. We advise against the use of the term lymphoblast based exclusively on morphology, due to the possible occurrence of acute myeloblastic leukemia with minimal differentiation or leukemia of ambiguous lineage. Nevertheless, the cytoplasmic characteristics of these blasts must still be described in the laboratory report (i.e. granules, Auer bodies, basophilia). This information provides a useful clinical information for classifying the type of blasts [8], [9], [10], [20], [21], [23], [24], [30], [32].

When blasts are observed during the first revision of blood smears, even when these cells are present in small numbers, the prescribing physician should be contacted for preventing misinterpretation. Ideally, the contact should be direct (e.g. by telephone) and should always be recorded [13], [16], [19]. The prescribing physician should also be contacted during routine follow-up, when significant deviations from previous data are noticed (i.e. cytological remission or significant increase of blasts number) [9], [10], [11], [15], [20], [21], [22], [23], [24].

Nucleated red blood cells (NRBC)

Outside of the neonatal period, NRBC only populate peripheral blood in association with some pathological conditions. The NRBC should hence be correctly identified, even when present in small numbers, because their identification may reflect a clinically significant condition and predict a poor outcome. The recommendation of the WGDH-SIBioC is to report NRBC in absolute number within the CBC profile, and the WBC count should consequently be corrected for the NRBC number [10], [13]. No evidence has been found that recognizing the various stages of erythroid cells differentiation and/or reporting qualitative anomalies may be clinically useful [15], [20], [21], [22], [23], [25].

Red blood cells (RBC)

The analysis of RBC morphology is usually considered the best diagnostic tool for triggering clinical or laboratory follow-up in anemic patients [33], [42]. Nevertheless, morphological erythrocyte abnormalities should only be reported when these can be considered clinically significant [13], [33], [42] and should be interpreted with caution, especially in infancy and childhood [42]. Notably, the assessment of some RBC parameters generated by the new generation of fully-automated haematological analyzers is more accurate and precise than using microscopic analysis [7], [10].

Reticulocytes

The morphologic abnormalities of reticulocytes usually lack clinical value, but the presence of reticulocytopenia or true reticulocytosis [47], [48] in association with results of other tests (e.g. RBC count, hemoglobin and morphological erythrocyte abnormalities) may facilitate the differential diagnosis of anemia, especially when accompanied by an appropriate comment. In particular, the observation of reticlocytosis in association with spherocytes, schistocytes or marked RBC heterogeneity is highly suggestive for hemolytic anemia [47]. Therefore, reticulocytes only provide limited information and do not retain a definitive diagnostic value.

The immature reticulocyte fraction (IRF), defined as the ratio between young or immature reticulocytes and the total number of reticulocytes, is an early and sensitive index of erythropoiesis. IRF increases before the total reticulocyte count and can be used to monitor bone marrow or stem cell regeneration post-transplant or chemotherapy.

In anemias characterized by reduced erythropoiesis, such as iron deficiency or anemia of chronic disease, the total reticulocyte count is reduced whilst the IRF is normal. After treatment for a nutritional anemia (vitamin B12, folate or iron deficiencies) IRF precedes the increase of the total reticulocyte count by several days, and is hence useful for monitoring therapeutic response.

The reticulocyte hemoglobin content provides an indirect measure of the functional iron available for new RBC production over the previous 3–4 days, also providing early measure of response to iron therapy, as it only increases within 48 h from the initiation of intravenous iron therapy. The parameter is especially important because its reduction mirrors iron-deficient erythropoiesis, even in conditions in which traditional biochemical markers such as ferritin and transferrin have an inadequate diagnostic efficiency (e.g. in patients with inflammation or anemia of chronic disease). Notably, no external quality assessment (EQA) scheme is available for this parameter [51], [52].

Platelets (PLT)

Besides description of analytical interferences (i.e. clumps, satellitism), the clinical utility of comments regarding PLT morphology is usually limited [10], [13], [36], [37], [38], [39], [41], [46], [48], [49]. The size of the PLTs may be sometimes diagnostic, especially when directly associated with an abnormal platelet count [10].

Platelets may be smaller than normal or, more often, larger than normal. The macroplatelet is defined by as an increase in size larger than half of a RBC, whilst giant platelet is defined as a size increase comparable to a normal erythrocyte. Microplatelets are conventionally defined as elements with a size in the range of 1 μm. Platelet size has diagnostic significance, especially when related to platelet count. Small or normal-sized platelets in association with thrombocytopenia suggest that the potential cause is insufficient bone marrow production, whilst thrombocytopenia with large platelets is more likely due to peripheral destruction or consumption with compensatively enhanced bone marrow production. Platelet size is also useful for identifying the possible cause of thrombocytosis. In reactive thrombocytosis (e.g. due to severe infections or inflammation), platelets are usually of normal size. Conversely, when thrombocytosis is observed in association with a myeloproliferative disorder (i.e. chronic myeloid leukemia, essential thrombocythaemia or polycythaemia vera), platelet size is generally increased and some giant platelets may also be identified.

The results of derived platelet parameters are highly dependent on the specific technology used for their assessment and are also influenced by anticoagulant used for collecting blood and the time passed between sample collection and analysis (e.g. the so-called EDTA-induced swelling phenomenon).

Platelets may display abnormal or absent granulation. When an artifact can be excluded, the identification of abnormal granulated platelets has diagnostic significance. It usually occurs as a rare congenital anomaly abnormality (e.g. the gray platelet syndrome), but can often be due to bone marrow diseases such as myeloproliferative or myelodysplastic disorders.

As for RBCs, morphological abnormalities of PLTs should only be reported when they have clinical significance [36], [37], [38], [39].

EDTA-dependent thrombocytopenia is one of the most frequent abnormalities associated with spurious platelet counts using hematology instrumentation. The performance of the “platelet clumps” flag is seldom inefficient using some fully-automated hematology analyzers. Therefore, microscopic analysis of blood smear is almost unavoidable to rule out platelet clumps in patients with low PLT count and unavailability of previous data [49], [53], [54], [55].

Discussion

Expert counseling on laboratory test results is an important post-analytic activity and an added value for establishing a positive relationship between the laboratory and the clinics [3], [5], [11], [12], [50]. Changing the role of clinical laboratories, from “passive analysis” of biological specimens to providing “expert counseling” on test results should hence be seen as a valuable paradigm shift in health care.

Reconsidering the laboratory-clinician interface not only may generate significant improvement of quality care as a whole, but can also lower the risk of medical errors and decrease health care expenditure [1], [2], [3], [4], [5], [6], [17], [18].

Nevertheless, harmonization initiatives aimed to enhance the clinical-laboratory interface are necessary principles for achieving the goal of improved relationship and communication inside and outside the laboratory. Harmonization is becoming more important as hospitals and laboratories are merged together, with development of large hospital and laboratory networks characterized by patients accessibility from multiple sites [10].

By publication of this consensus paper, the main aim of the WGDH-SIBioC is proposing a list of simple recommendations, based on a language that can be understood by the vast majority of health care professionals, aimed to enhance the overall quality of the cytomorphological diagnostic process. The list of interpretative comments for the various abnormalities discussed in this document is not projected as being exhaustive, nor do quantitative anomalies always need to be highlighted in the laboratory report [10], [15], [20], [21], [22], [23], [24], [37], [38].

The practice of including interpretative comments in laboratory reports does not rule out the need to directly interact with clinicians under some circumstance, especially, improving the timely and effective communication of critical test results with significant impact on medical decisions and patient outcome [3], [5]. This contact should be direct (e.g. by phone) and must always be recorded [19].

By promoting harmonization of interpretative comments in laboratory hematology, the WGDH-SIBioC believes that laboratory professionals may provide strategic leadership for improving diagnostic efficiency and promoting better health care delivery. Scientific associations should play a pivotal role for promoting the importance of harmonization.