Oxygen That Is Carried but Not Released: A Two-Axis Glycohypoxia Hypothesis for the Excess Mortality of Tuberculosis with Diabetes
Amr Kamel Khalil Ahmed, Khulood Almutairi
Abstract
People treated for tuberculosis who also have diabetes die at roughly twice the rate of people treated for tuberculosis alone. The pooled odds ratio for death sits near 1.9, and it has barely moved across two decades of meta-analysis. The usual explanations, impaired phagocyte function, higher bacillary burden, slower culture conversion and drug interactions, are all real, but none of them explains why the excess persists after adjustment for age, cavitation, adherence and renal disease. We propose a different reading, built on a single physiological currency: the oxygen
the tissue actually receives. Tuberculosis creates a host that leans on two
compensations at once. The first is hematological. Anemia of chronic inflammation and chronic hypoxemia raise erythrocyte 2,3-Diphosphoglycerate (2,3-DPG), also known as 2,3-bisphosphoglycerate (2,3-BPG), which lowers hemoglobin oxygen affinity and preserves unloading at the low oxygen tensions found inside and around granulomas. Torrance and colleagues measured this adaptation at about
+0.30 mmHg of P50 for every 1 g/dL fall in hemoglobin. The second compensation is transcriptional. Granuloma hypoxia stabilizes HIF-1α, which drives interleukin-1β, inducible nitric oxide synthase and the glycolytic switch that macrophages need to restrict Mycobacterium tuberculosis.
Chronic hyperglycemia attacks both. Non-enzymatic glycation of the hemoglobin βchain N-terminal valine, the modification measured as HbA1c, raises oxygen affinity
and has recently been quantified at about -0.19 mmHg of P50 per 1% HbA1c, the
effect proposed as glycohypoxia. Methylglyoxal, the dicarbonyl by-product of
glycolytic flux that accumulates in diabetes, degrades HIF-1α through a CHIPdependent route and blocks the residual protein from recruiting p300Neither axis is new on its own. Our claim is about their conjunction: in tuberculosis with uncontrolled diabetes, the two compensations that matter most in a hypoxic lesion fail together, and they fail in a patient whose arterial oxygen tension and pulse oximetry look normal. We set out the quantitative form of the blood axis, show that it predicts erosion rather than reversal of the anemic right shift in most patients, address the older whole-blood studies that argue against a simple leftward shift, and list the measurements that would falsify the hypothesis.
Keywords: tuberculosis; diabetes mellitus; HbA1c; glycohypoxia; P50; 2,3-Diphosphoglycerate (2,3-DPG), 2,3-bisphosphoglycerate (2,3-BPG); HIF-1α; methylglyoxal; mortality