Severe hypoglycemia can lead to fatal coma or severe braindamage.
Severe hypoglycemia can lead to fatal coma or severe brain damage. Profound hypoglycemia typically affects diabetic patients receiving treatment with oral hypoglycemic agents or insulin as a complication of therapy. Radiologists seldom come across patients with severe hypoglycemia, except in rare instances when abdominal CT is requested to rule out insulinoma as the cause. The clinical cause of the coma is usually known, and the value of MRI of the brain to diagnose hypoglycemia or predict outcome is not backed up by evidence- based studies.
Physicians began to see clusters, however, of nondiabetic men admitted to hospitals across Singapore with severe, unexplained hypoglycemia in February 2008.1 Several of these patients suffered severe coma and neurological sequelae. At least four subsequently died. This wave coincided with a sudden increase in requests for MRI examinations of coma patients who had severe brain damage.
All patients underwent a series of toxicological tests and were questioned about their drug-taking history. This questioning revealed that their drop in blood sugar was due to consumption of an illegal product sold for sexual enhancement that contained glibenclamide or glyburide, an oral hypoglycemic agent commonly used for diabetes treatment.
The majority of cases were associated with Power 1 Walnut, an illegal preparation that claims to be an herbal medicine. These preparations had been purchased from unauthorized itinerant peddlers and makeshift stalls. Forensic chemistry tests showed that the supposedly natural tablets were adulterated with glibenclamide and small traces of sildenafil, the active ingredient in Viagra.
Most patients who were poisoned by up to five times the total daily recommended dose of glibenclamide were treated successfully in the emergency department without significant sequelae. Eight patients suffered prolonged hypoglycemia and severe complications.2,3
Patients were examined on 1.5T and 3T MRI systems using conventional and advanced pulse sequences. Diffusionweighted imaging and T2-weighted MRI showed abnormalities in the hippocampus, cerebral cortex, and basal ganglia, sparing the subcortical white matter and cerebellum (Figure 1). These findings are consistent with previous case reports and animal studies that describe the selective vulnerability of the brain to hypoglycemic damage and characteristic lesion distribution in the cerebral cortex, insular cortex, and hippocampus.
The MRI distribution and appearance of these lesions are not specific, and ischemia, hypoglycemia, and epilepsy may have similar MRI findings. The underlying neurochemistry, time course, and selective neuronal necrosis are, however, fundamentally different and distinct for each of them.4
Some patients also had abnormalities in the splenium of the corpus callosum and internal capsules. These may represent a distinctive pattern of limited involvement on diffusion-weighted imaging (Figure 2).5,6 The abnormalities were not seen on T2-weighted images in some of these reported cases. They may normalize rapidly on follow-up imaging and probably represent fluid movement rather than irreversible cell necrosis. Further follow-up studies may still be needed to better understand the mechanisms of these unusual findings.
Three patients had unilateral cortical involvement, a finding that had not been reported before (Figure 3). This pattern of unilateral involvement could be misdiagnosed as cerebral infarction. Investigators noted that it did not conform to the typical cerebral arterial territories and that these patients had normal MR angiography. Perfusion MRI in one patient showed a slight increase in relative cerebral blood volume, and no evidence of abnormal lactate was detected on MR spectroscopy. It is still not known why generalized hypoglycemia to the whole brain caused unilateral lesions in these patients.
Previous reports of hypoglycemic brain damage have been concerned with isolated case reports or animal experiments. This “outbreak” allowed researchers to collate imaging findings on hypoglycemic brain damage from a much larger patient group.1
The investigation highlighted the importance of multidisciplinary collaboration among radiologists and referring clinicians from different services, including endocrinology, general medicine, neurology, and the emergency department. Formal case conferences and informal “curbside” consultations helped to highlight clusters and/or trends in patients. The Health Sciences Authority of Singapore was instrumental in coordinating the distribution of timely information, and sent frequent clinical updates and alerts about the outbreak to hospital doctors.7,8
The initial outbreak described was associated with an adulterated “herbal” product labeled Power 1 Walnut. Three other illegal preparations have since been found to contain high levels of glibenclamide and have been linked to cases of severe hypoglycemia. These include a preparation falsely labeled as Cialis and two other allegedly herbal preparations, Santi Bovine Penis Erecting Capsule and Zhong Hua Niu Bian.9
The recreational use of phosphodiesterase- 5 inhibitors such as sildenafil (Viagra) to enhance sexual performance is well known. The fast-growing parallelimport, counterfeit, and imitation herbal supplement market complicates matters. It is possible that manufacturers of the products linked to this outbreak used the wrong active pharmaceutical ingredient. It is still not clear, however, why illegal sexual enhancement products should be adulterated with an unrelated drug for the treatment of diabetes, and whether this was intentional or accidental. The number of new cases of severe hypoglycemia has decreased as a result of rigorous law enforcement. It has not reached zero, however, and patients are still suffering hypoglycemic brain damage that is potentially fatal yet completely preventable.
The source of these illegal drugs has not been found. Similar cases of hypoglycemia have been reported in the lay press in China and Hong Kong.10 This seems to implicate an imported source from factories overseas. In February 2009, a new outbreak of poisoning by glibenclamide in fake diabetes drugs was reported.11
There are few reports about this phenomenon in the scientific literature. The real and dangerous threat that counterfeit drugs and pseudoherbal products might enter the pharmaceutical supply chain via dubious online pharmacies is seldom discussed.
Radiologists need to recognize the typical and atypical MRI findings of severe hypoglycemia, including cortical, hippocampal, and basal ganglion abnormalities, with additional involvement of the corpus callosum and internal capsule in some cases.
Unilateral cortical diffusion-weighted MRI lesions may be distinguishable from acute ischemic infarction by the nonvascular pattern of involvement and normal MR angiography, perfusion, and spectroscopy findings.
Suspicious clusters of such complications in nondiabetic patients should alert physicians to coordinate with clinical and regulatory colleagues to investigate the possibility of illegal and adulterated health products. Cooperation among networks of radiologists across different hospitals and, perhaps more important, across international borders may also be essential in expediting discussion and image analysis and spotting new patterns of abnormalities in these unusual situations.
We urge readers of Diagnostic Imaging Asia Pacific who come across similar cases of unexplained hypoglycemia that may be linked to illegal drug use to contact us by e-mail at firstname.lastname@example.org.