Background

Anemia is a common complication of chronic kidney disease (CKD). The management of anemia in CKD patients must strike an appropriate balance between stimulating generation of erythroblasts (erythropoiesis) and maintaining sufficient iron levels for optimum hemoglobin (Hb) production.¹ As such, it is important to assess iron stores and the availability of iron for erythropoiesis, as adequate iron status is integral to both iron and anemia managements in CKD patients.

Classical iron status tests, of which ferritin and transferrin saturation (TSAT) are the most widely used, exhibit large biological variability in the context of underlying inflammation of CKD.^2–4 The accurate assessment of iron status is dependent on the validity and reliability of laboratory test results, and differences in test performance pose a dilemma regarding the most appropriate test to guide treatment decisions. Several novel biomarkers of iron status have been proposed as alternatives to the classical iron status tests. These include hemoglobin content of reticulocytes (CHr), reticulocyte hemoglobin equivalent (RetHe), percentage of hypochromic erythrocytes (%HYPO), erythrocyte zinc protoporphyrin (ZPP), soluble transferrin receptor (sTfR), and hepcidin. In addition, Superconducting Quantum Interference Devices (SQUIDs) are an alternative non-invasive means for detecting and quantifying liver iron content, via the paramagnetic properties of iron (magnetic resonance diminishes in the liver as iron concentration increases).

The Tufts Evidence-based Practice Center (EPC) conducted a Comparative Effectiveness Review (CER) to systematically evaluate studies that examined the impact on patient-centered outcomes of using the newer laboratory biomarkers as a replacement for or as an add-on to classical laboratory biomarkers of iron status for assessing iron status and the management of iron deficiency in adult and pediatric CKD patients (nondialysis and dialysis).⁵ The Key Questions for the CER are presented below:

Key Question 4

What factors affect the test performance and clinical utility of newer markers of iron status, either alone or in addition to older laboratory biomarkers, in stages 3–5 (nondialysis and dialysis) CKD patients with iron deficiency?

Combining the evidence addressing Key Questions 2, 3, and 4, the CER concluded that there is currently insufficient data to determine if most newer laboratory biomarkers of iron status are better than classical markers for predicting iron deficiency as defined by a response to an iron challenge test. However, it may be that CHr and %HYPO have better predictive ability for a response to intravenous (IV) iron treatment than classical markers (TSAT <20% or ferritin <100 ng/mL) in HD CKD patients. In addition, results from two randomized controlled trials (RCTs) showed a reduction in the number of iron status tests and resulting IV iron treatments administered to patients whose iron management was guided by CHr, compared with those guided by TSAT or ferritin. These results suggest that CHr may reduce potential harms from IV iron treatment by lowering the frequency of iron testing; however, the evidence for the potential harms associated with testing or test-associated treatment is insufficient.

Nevertheless, the strength of evidence supporting these conclusions is low and there remains considerable clinical uncertainty regarding the use of newer markers in the assessment of iron status and the management of iron deficiency in stages 3–5 CKD patients (both nondialysis and dialysis). In addition, factors that may affect the test performance and clinical utility of newer laboratory markers of iron status remain largely unexamined.

Table A summarizes the evidence gaps identified in the CER. One major evidence gap concerns the dearth of pediatric studies. Addressing this gap would require a specially-composed stakeholder group for determining specific Future Research Needs (FRN). For this reason, the current FRN project is focused on adult CKD patients (nondialysis and dialysis).

Table A

Research gaps and suggestions for future research.

Methods

Results

Based on the CER’s Future Research section and our discussion with stakeholders, 17 Future Research Needs topics were nominated. We considered the five topics with endorsement by at least fifty percent of the nine voting stakeholders as the highest priority FRN topics. The topics chosen as the highest priority Future Research Needs are listed in Table B.

Table B

Top five Future Research Needs as indicated by participating stakeholders.

Nomination of FRN Topic 1 highlights the current lack of a well-accepted reference standard for diagnosing iron deficiency in CKD patients, which is consistent with the findings of our CER. To compare the test performance among medical tests, a common reference standard is needed. When studies use different definitions of a reference standard, the results cannot be compared or “summed up” across studies. Without using the same definition of reference standard, conducting more studies on the test performance of existing or new medical test is unlikely to build up the body of evidence and therefore impact current practice.

Based on stakeholder discussion, it appears that iron staining of a bone marrow biopsy specimen is widely regarded as the “gold standard” for the diagnosis of (absolute) iron-deficient anemia, although this viewpoint is not universally accepted in the setting of CKD. Bone marrow iron may have limited clinical use due to the risks of infection or bleeding at the biopsy site. Despite these limitations, a bone marrow biopsy remains the most accurate measure that reflects stored iron, and thus should be used to define absolute iron deficiency. On the other hand, there is currently lack of a well-accepted reference standard for functional deficiency. Thus, we suggest that an expert panel be convened to standardize the definition for functional iron deficiency and determine which definition should be considered the preferred reference standard for diagnosing functional deficiency. The panel should also assess which intermediate outcomes (e.g., erythropoiesis-stimulating agent [ESA] or iron treatment dosages) or test characteristics (e.g., test availability/accessibility, cost) are appropriate to consider in determining the ideal definition for functional iron deficiency. In addition to representatives from all stakeholder categories, this panel should specifically include authoritative bodies and major professional organizations with relevant interests in iron deficiency, using a process similar to that used by the Centers for Disease Control and Prevention (CDC) Lipid Standardization Program (www.cdc.gov/labstandards/lsp.html).

FRN Topic 2 was Key Question 2 in the original CER. Although we did find studies comparing classical and newer tests for diagnosing absolute iron deficiency, they used classical laboratory biomarkers (alone or in combination with each other) as the reference standard for iron deficiency, essentially measuring the concordance between classical and newer biomarkers of iron status. Thus, we were unable to answer the question. We suggest that future research on this topic follow a prospective cohort design, as such studies would allow for multiple tests to be compared all together and potential biases could be minimized. In addition, depletion of bone marrow iron should be used as the reference standard for absolute iron deficiency, as iron staining of a bone marrow biopsy specimen is considered the most accurate measure that reflects stored iron. However, we expect that it will be difficult to recruit patients for such studies, because bone marrow aspiration is painful and poses risks to some patients.

FRN Topic 3 was also part of Key Question 2 in the original CER. Based on our post-hoc observation of this body of literature, we found that current studies often used a response to intravenous (IV) iron treatment as the reference standard for functional iron deficiency. However, there was no uniform regimen of IV iron in terms of dosage and iron formulation. There was also a wide range of durations of IV iron treatment across studies. These variations in the reference standards in the published studies resulted in incomparable study results, and limited the strength of body of evidence. Therefore we suggest that it is vital to establish a preferred reference standard for functional iron deficiency (also see FRN Topic 1), before future research on this topic go forward. Future research on this topic should use the same reference standard, in order to grow the body of evidence on this important research question in a manner amenable to systematic review and meta-analysis, to compare results across studies.

Similar to FRN Topic 2, we also suggest that future research on this topic also utilize a prospective cohort design, and the agreed upon (by the consensus expert panel) definition should be used as the reference standard for diagnosing functional iron deficiency. If the reference standard for functional iron deficiency is defined by a response to ESA or iron treatment, a sufficient washout period (at least 4 weeks) is needed to stop ESA and iron treatment before ascertainment of the baseline test measurements. Stopping treatment may not be feasible for some clinical settings. In this case, studies should recruit only CKD patients who did not receive ESA and iron treatment within 1 month before baseline test measurements.

Our CER did not directly address FRN Topics 4 and 5, so we suggest that as a first step, prior to conducting new research studies on these topics, an expert panel be convened to determine and prioritize the elements needed for design future research on these topics (such as which biomarkers and outcomes are appropriate and relevant to consider in determining an ideal marker to monitor response to therapy and repletion status, or monitoring iron overload). We noted that markers that can separate cell hemoglobinization from cell production would provide researchers a better chance of understanding the underlying mechanism, potentially enabling better targeted treatments and therefore better managed anemia. Markers that have specificity to responses related to ESA or iron delivery would be optimal.

In terms of the expert panel composition, to balance viewpoints the panel should specifically include clinical chemists and hematologists. Following identification of the most appropriate biomarkers by the expert panel and determination of a reference standard, prospective cohort studies with consecutive CKD patients are again, as in our previous suggestions, the ideal study design that should be conducted for future research on these topics. Moreover, future research on these topics should assess clinical or patient-centered outcomes.

Discussion

The prioritization of topics for future research is a stakeholder-driven process. Our stakeholder panel represented a broad range of perspectives, from a well-informed patient advocate to clinical experts and policy makers. However, our stakeholder panel is unlikely to represent all perspectives because we did not use formal sampling methods to select our stakeholder members. The process of engagement though multiple teleconferences enabled us to get a well-rounded perspective; and we believe that the process of sharing the minutes of the discussions with all stakeholders enabled them to appreciate one another’s viewpoints. We were able to obtain input from all members of the stakeholder panel, and the final list showed a clear preference for the top five priorities.

There are several challenges for designing future research on the priority topics. The ideal study design and sampling populations would vary depending on the purpose of using a medical test (e.g., screening, diagnosis, prognosis, monitoring, etc.), in order to maximize the internal and external validity of a study. One of the challenges in the current FRN project is that an iron status test can be used for the purposes of screening, diagnosis, and/or monitoring for iron deficiency anemia in CKD patients. However, the same iron status tests can be used for the purposes of diagnosis and monitoring without considering the biases in interpretations of test results that are highly likely in these instances. Another challenge is that patients’ iron status can change spontaneously (due to changes in diet, or due to unrelated metabolic and inflammatory conditions) or due to the treatment received. Therefore, it is important to control for these confounding factors in future research studies, which will require large sample sizes to reach sufficient statistical power.

Conclusions

Our CER and FRN stakeholder discussions pointed out the top future research gaps, and highlighted the great deal of confusion/uncertainty in using newer and/or classical laboratory biomarkers of iron status for the purposes of the diagnosis or monitoring of iron deficiency anemia in CKD patients. The chief factor causing this confusion and uncertainty is the lack of a well-accepted reference standard for iron deficiency anemia. The most effective first step would be to establish a common reference standard for iron deficiency anemia, considering two separate and distinct definitions: absolute versus functional iron deficiency. The ideal reference standard should be independent of the index tests and test-directed treatment to maximize the internal validity of study results.

References

1.

Anaemia Management in Chronic Kidney Disease - Rapid Update 2011. National Clinical Guideline Centre; 2011. www​.ncbi.nlm.nih.gov​/books/NBK65530/pdf/TOC.pdf.

2.

Winkel P, Statland BE, Bokelund H. Factors contributing to intra-individual variation of serum constituents: 5. Short-term day-to-day and within-hour variation of serum constituents in healthy subjects. Clin Chem. 1974 Dec;20(12):1520–27. [PubMed: 4430129]

3.

Winkel P, Statland BE, Bokelund H. The effects of time of venipuncture on variation of serum constituents. Consideration of within-day and day-to-day changes in a group of healthy young men. Am J Clin Pathol. 1975 Oct;64(4):433–47. [PubMed: 1239188]

4.

Wish JB. Assessing iron status: beyond serum ferritin and transferrin saturation. Clin J Am Soc Nephrol. 2006 Sep;1 Suppl 1:S4–S8. [PubMed: 17699374]

5.

Chung M, Moorthy D, Hadar N, et al. Biomarkers for Assessing and Managing Iron Deficiency Anemia in Late-Stage Chronic Kidney Disease. Comparative Effectiveness Review No. 83. Rockville, MD: Agency for Healthcare Research and Quality; Oct, 2012. (Prepared by the Tufts Evidence-based Practice Center under Contract No. 290-2007-10055-I.) AHRQ Publication No. 12(13)-EHC140-EF. www​.effectivehealthcare​.ahrq.gov/reports/final.cfm. [PubMed: 23762920]

6.

Concannon T, Meissner P, Grunbaum J, et al. A New Taxonomy for Stakeholder Engagement in Patient-Centered Outcomes Research. Journal of General Internal Medicine. 2012:1–7. [PMC free article: PMC3403141] [PubMed: 22528615]

Footnotes

1

Newer laboratory biomarkers: content of Hb in reticulocytes, percentage of hypochromic red blood cells, erythrocyte zinc protoporphyrin, soluble transferrin receptor, hepcidin, and superconducting quantum interference devices.

2

Older laboratory biomarkers: bone marrow iron stores, serum iron, transferrin saturation, iron-binding capacity, and ferritin.