Aim Remote conditioning, whereby intermittent non-critical ischaemia of remote peripheral muscle can protect cardiac muscle from ischaemia-reperfusion (I-R) injury even when applied after the onset of cardiac ischaemia, has emerged as a potential therapeutic manoeuvre for reducing I-R injury. Whether such protection is mediated via serum factors and whether the development of the protection is impaired in disease states is unclear. We sought to develop a cell based model for assessing I-R injury, and utilised this to screen sera from both control healthy volunteers and type 1 diabetic patients for blood borne cardioprotective signals capable of mediating remote conditioning protection.

Methods Subjects studied included healthy volunteers and patients with type 1 diabetes with microvascular complications (retinopathy, microalbuminuria). Control (unconditioned) blood samples were taken prior to remote conditioning. An upper limb was then occluded with a tourniquet [30 mm Hg suprasystolic, 5 min] and released to reperfuse the arm [5 min] three times. Conditioned blood was collected immediately afterwards from the contralateral arm. Serum was separated and stored at −80°C until assay. Serum was screened for its protective capacity using a cellular model of I-R. Myocardial ischaemia was simulated by centrifugation of freshly isolated rat cardiac ventricular myocytes into a pellet [30 min]. Gaseous diffusion was prevented by an impermeant layer of mineral oil. Reperfusion injury was simulated through dispersal of the pellet in oxygenated saline solution. Cell viability was determined by propidium iodide staining for necrosis and calcein-AM counterstaining for viability.

Results In healthy subjects (n=21) unconditioned serum resulted in necrosis of 23.7%±4.9% (mean±SEM) of cells. Conditioned serum resulted in a significant reduction of necrosis to 6.4%±5.1% (p<0.05). However conditioned serum from diabetics showed no evidence of protection compared with unconditioned diabetic serum (51.1%±4.6% vs 45.7%±6.4% respectively, n=14, p>0.05, NS). The difference in necrosis rates between conditioned serum from healthy subjects and diabetics was highly significant (p<0.001). Results were similar whether the serum was applied before centrifugation of the cells (mimicking preconditioning in vivo) or on reperfusion (mimicking postconditioning).

Conclusion We report the development of a novel bioassay that can be used to quantitatively assess the strength of the serum-based signal that mediates remote conditioning. We further show that this signal is significantly impaired in patients with advanced diabetes. The later finding has important implications for the interpretation of clinical trials of remote conditioning, which include diabetics.