SEARCH uncovers significantly higher rates of nephropathy in young people with type 2 diabetes

Although the absolute numbers remain low, incidence rates of young people (YP) with type 2 diabetes (T2D) in some ethnic groups are similar to those seen for type 1 diabetes (T1D) in white (Europid) populations (Mayer-Davis et al, 2017). These data from the SEARCH investigators in the US confirm an increase in the annual incidence rate of 1.4% for T1D, but a five-fold greater increase for T2D from 2002/3 to 2011/12. In the study summarised alongside, the same investigators report the rate of microvascular and other complications in YP with diabetes.

SEARCH recruited 2018 newly diagnosed YP (1746 with T1D and 272 with T2D) aged <20 years from five diabetes registries. A subset with a diabetes duration of >5 years (mean, 7.9 years) were reviewed for the presence of nephropathy (defined as a single albumin:creatinine ratio measurement of ≥30 mg/g [~3 mg/mmol]); retinopathy (graded mild, moderate or proliferative from digital photography); or neuropathy (defined as a score >2 using the Michigan Screening Instrument). Cardiac autonomic neuropathy, arterial stiffness and hypertension were also assessed.

The age-adjusted prevalence of nephropathy was almost four times higher in those with T2D (19.9% vs 5.8%; P<0.001). The differences were less for retinopathy and neuropathy, but the overall odds ratios for YP with T2D were 2.58, 2.24 and 2.53, respectively. The odds ratios were not significant for autonomic neuropathy, arterial stiffness or hypertension. Overall, 72% of those with T2D had at least one diabetes complication, compared to 32% with T1D.

For T1D, the rates are similar to the 2003 DCCT/EDIC study (DCCT/EDIC Research Group, 2003). This is disappointing but not surprising as the mean HbA_1c at the time of review was 77 mmol/mol (9.2%), almost identical to the conventional control arm of the DCCT. For T2D, the data are more worrying. In the UKPDS, the prevalence of nephropathy 15 years after diagnosis was 28% (Bilous, 2008). SEARCH has a rate of nearly 20% after 7.9 years.

What might explain this disparity? Of the clinical features associated with nephropathy development, YP with T2D were more overweight; had higher blood pressures; were more likely to be black, Hispanic or Native American; and more likely to come from poorer socioeconomic backgrounds. Obviously, all of those with T1D were on insulin, but only half of those with T2D were, and 27% were on no blood glucose-lowering therapy. Despite these differences, HbA_1c levels were similar.

There are some caveats. First, the numbers with T2D were relatively low and these individuals may represent a particularly vulnerable population. It may not be possible to extend the results to all YP with T2D globally. Second, albuminuria was assessed only once and there is a recognised large day-to-day variability. Moreover, the albumin:creatinine ratio is increased in obese people and in those with hypertension, and it is uncertain that all of those identified in this study will progress to renal impairment.

Nevertheless, these data show we are still falling well short of target glycaemic control in YP, and that those with T2D seem particularly vulnerable to potentially life-shortening complications. Some of the causative factors are non-modifiable, but glycaemia, blood pressure and obesity can be addressed and should be targeted at least as vigorously in YP with T2D as they are for T1D. After witnessing a steady decline in rates of end-stage renal disease in people with T1D (US Renal Data System, 2016), we may now be confronting a much greater increase in YP with T2D.

To read the article summaries, please download the PDF