Conflicting findings emerge on the link between HbA1c and mortality in diabetics on dialysis.
Measured by glycosylated hemoglobin A1c (HbA1c), poor diabetic metabolic control correlates with the complications such as nephropathy, as demonstrated in two important studies, the Diabetes Control and Complications Trial and the U.K. Prospective Diabetes Study.
Therefore, it should come as no surprise that inadequate glycemic control in diabetic patients not on dialysis is associated with greater macrovascular complications, including MI, stroke, and peripheral vascular disease. Now the question is whether poor glycemic control influences vascular complications and mortality in diabetics on hemodialysis. We have some answers from recent reports.
Continue Reading
First, Mark Williams, MD, of the Joslin Diabetes Center in Boston, and his colleagues analyzed glycemic control and survival in an end-stage renal disease database of 23,504 diabetes patients (95% of whom were type 2 diabetics) who had a recorded HbA1c in the three months preceding the study baseline date of January 1, 2003 (Kidney Int. 2006;70:1503-1509). The mean HbA1c levels were 6.7% in type 2 and 7.5% in type 1 diabetics.
Overall, 35% had an HbA1c above 7%. For those pa-tients with HbA1c above 6%, 37% were not being treated with insulin or hypoglycemic agents. Survival rates at 12 months ranged from 80%-85% between groups. Survival rates did not differ significantly for the five HbA1c subgroups, but the study was limited by its relatively short follow-up of only one year.
Investigators at Osaka City University Graduate Medical School in Japan followed 114 diabetic patients on hemodialysis for seven years (Diabetes Care. 2006;29: 1496-1500). The researchers categorized patients by mean HbA1c level in the three months preceding study entry. They defined good control as HbA1c less than 6.5%, fair control as 6.5%-8.0%, and poor control as 8.0% or higher. Survival rates were not different at one year. Over seven years, cumulative survival rates were lower in the poor-control group, but not different between the good- and fair-control groups. Neither study included HbA1c recorded after study entry, so the effect of changes in glycemic control were not included in either analysis.
More microvascular problems
A team at Queen’s University in Ontario studied vascular complications in 100 diabetic patients on hemodialysis (Diabetes Care. 2006;29:2247-2251). They divided the population into two groups based on HbA1c under or above 7.0%. There was no difference in macrovascular complications between the groups; retinopathy and neuropathy were twice as common in subjects with HbA1c above 7%, but the difference in the proportion of patients with these conditions was not statistically different, probably because of the relatively small number of subjects in the study overall.
It remains unclear whether diabetic patients on hemodialysis should be treated to improve glycemic control. No clinical study has looked at the effect of intervention. Studies of nondiabetics suggest that achieving glycemic control of HbA1c of 7% or less reduces complications and possible mortality. The metabolic sit-uation in hemodialyzed patients, however, is complicated, and control of other risk factors affected by dietary intake may dominate the treatment strategy. Some patients may be able to improve glycemic control with appropriate dietary instruction. Some guidance on that topic follows.
A major barrier to improving glycemic control through diet has been the complexity of the instructions. The older of the two current methods is the exchange system in which each food has a count of how many units of fat, carbohydrate, and protein. Patients are prescribed a dietary pattern and choose foods to match that pattern. A newer method is to count carbohydrates because it is the major food component that affects blood glucose. Although simpler than the exchange method, the carbohydrate counting method has been criticized for ignoring differences in the rate and amount of rise in blood glucose after different foods with similar total carbohydrate.
Labeling each food item
A newer method is to use the glycemic index (GI) of a food. GI is a ratio of how much rise in blood glucose a food causes compared with the same amount of glucose ingested. Implementation of this dietary treatment consists of substituting similar foods with lower GI. An example would be the substitution of Kellogg’s Special K (GI = 56) for Rice Krispies (GI = 82). Thousand of foods have now been analyzed and values are available on the web at www.glycemicindex .com. Shorter lists of common foods are available in the series of books entitled “Glucose Revolution” with Jennie Brand-Miller as first author. Some lists have categorized foods as good, marginal, and bad based on GI. The implementation is less demanding, because nothing is counted. Results of large scale clinical trials using this dietary treatment have not yet been published, but smaller trials show encouraging results, researchers say.
