A study recently published in Eye evaluated the association between hemoglobin levels and diabetic retinopathy (DR) risk in adults with type 2 diabetes mellitus (T2DM).
Give me some background.
While hyperglycemia is well established as the primary driver of DR, other modifiable risk factors— such as hemoglobin levels—may contribute to DR development as well.
Hemoglobin levels reflect the oxygen-carrying capacity of the blood and potentially have a significant impact on retinal microcirculation.
- Why: Reduced hemoglobin levels could theoretically compromise oxygen delivery to retinal tissues, exacerbating hypoxia-induced damage in diabetes-affected retinas.
In addition: A recent meta-analysis of 51 studies and over 26,000 patients found that the global prevalence of anemia among diabetics was ~35.45%, suggesting a substantial overlap between these two conditions.
Let’s dive into this new research.
In this retrospective cohort study, investigators utilized the TriNetX network to identify adults aged ≥45 years with T2DM from 2010 to 2022 and divided the cohort into two groups:
- Low hemoglobin (LHB): 8-12 g/dL
- Control: ≥12 g/dL
Propensity score matching (PSM) was used to address potential confounders and subgroup analyses examined the impact of:
- Hypertension status
- Glycemic control
- Sex
And the primary and secondary outcomes?
- Primary outcome: New-onset DR occurring 6-36 months after the index date
- Secondary outcomes: DR subtypes and diabetic ophthalmic complications
Findings?
After PSM (28,882 patients per group), patients in the LHB group showed significantly higher risk of:
- Overall DR (hazard ratio [HR]: 1.33, 95% confidence interval [CI]: 1.24-1.44)
- Proliferative DR (HR: 1.95, 95% CI: 1.62-2.35)
- Non-proliferative DR (HR: 1.19, 95% CI: 1.06-1.33)
In fact: Even mild hemoglobin reductions (10-12 g/dL) were associated with an increased risk of:
- Overall DR (HR: 1.28, 95% CI: 1.17-1.40)
- Proliferative DR (HR: 1.90, 95% CI: 1.79-2.42)
- Non-proliferative DR (HR: 1.17, 95% CI: 1.02-1.34)
Note: The stronger association with proliferative DR suggests that low hemoglobin levels may contribute to disease severity by promoting neovascularization, potentially via hypoxia-driven angiogenic pathways.
Anything else?
The association between DR and low hemoglobin levels was stronger in patients without hypertension (HR: 1.58, 95% CI: 1.40-1.78 vs. 1.21, 95% CI: 1.10-1.35).
- Why: Hypertension may contribute to DR through mechanisms, such as impaired autoregulation of retinal blood flow and endothelial dysfunction, potentially overshadowing the effects of low hemoglobin levels.
This relationship remained consistent regardless of HbA1c levels, suggesting that low hemoglobin level represents an independent risk factor for DR, even in patients with relatively well-controlled diabetes.
Go on …
Finally: In the sex-stratified subgroup analysis, the research team used sex-specific hemoglobin thresholds.
What this revealed: A stronger association in males (HR: 1.52, 95% CI: 1.37-1.68) than females (HR: 1.26, 95% CI: 1.15-1.38).
Expert opinion?
The study authors noted several clinical implications based on the results of this analysis, including:
- Routine hemoglobin monitoring should be considered as part of comprehensive diabetes management, even in patients without overt symptoms of anemia
- Maintaining optimal hemoglobin levels may represent a modifiable risk factor for DR prevention alongside glycemic control and blood pressure management
- Patients with T2DM and low hemoglobin levels may benefit from more frequent ophthalmologic screening to detect early DR changes
- Addressing the underlying causes of low hemoglobin levels, such as iron deficiency, chronic inflammation, or occult blood loss, may modify the risk of DR—though interventional studies are required to validate this hypothesis
Any notable limitations?
These included:
- Despite comprehensive PSM, residual confounding factors could not be completely eliminated
- Electronic health record (EHR) use may introduce misclassification bias due to coding inaccuracies
- There was a dearth of information on certain potential confounders, such as diabetes duration, socioeconomic status, and detailed medication adherence, which could influence about hemoglobin levels and DR risk
- The study could not differentiate between causes of low hemoglobin levels, which might have different implications for DR risk
- The TriNetX network may not be fully representative of all patient populations, potentially limiting its generalizability
- While the research team assessed healthcare utilization to address surveillance bias, this approach could not fully account for potential differences in care quality or comprehensiveness
Now the take home.
These findings demonstrate that low hemoglobin levels independently increased the risk of DR in adults with T2DM, with the strongest association for proliferative DR.
- This relationship was observed even with mild reductions in hemoglobin and remained consistent regardless of glycemic control.
As such: Routine hemoglobin monitoring and maintenance of optimal levels may represent a modifiable risk factor for DR prevention in addition to glycemic control and blood pressure management.
Last question: What are the next steps in this research?
Interventional studies targeting hemoglobin levels are needed to determine whether correcting low hemoglobin levels can modify DR risk and progression.