Visuomotor Performance in KCNJ11-Related Neonatal Diabetes Is Impaired in Children With DEND-Associated Mutations and May Be Improved by Early Treatment With Sulfonylureas

All subjects were evaluated during a 2010 family forum in Chicago, as part of our Monogenic Diabetes Registry (http://monogenicdiabetes.uchicago.edu) (11). All parents provided informed written consent as approved by The University of Chicago institutional review board.

The Beery-Buktenica Developmental Test of Visual-Motor Integration (VMI) presents drawings of geometric forms in order of increasing difficulty to be copied with paper and pencil for subjects >2 years of age. The VMI is often administered to evaluate visual-motor and visual-perceptual deficits. It has a coefficient α of 0.82, and its validity has been established (12–14). The VMI was individually administered by a single clinician trained in the assessment. Raw scores are converted to age-appropriate standard scores.

Statistica (version 10.0; http://www.statsoft.com) was used for all analyses. Nonparametric analysis was performed using the Kruskal-Wallis ANOVA test (with value H) for group comparisons, as well as Spearman correlations.

Nineteen subjects 2.2–20.3 years of age participated in the study. Eight subjects had the R201H mutation characterized by isolated diabetes without neurodevelopmental concerns, eight had V59M or V59A (V59M/A) mutations associated with the intermediate DEND syndrome, and three had mutations that have an inconsistently reported neurodevelopmental phenotype (one each with R201C, Y330C, and E322K) (Table 1). All subjects were being successfully treated with oral sulfonylurea monotherapy. Overall, no group differences were found in age (Kruskal-Wallis ANOVA group comparison H [2, N = 19] = 1.9; P = 0.39), sex distribution (χ² [2] = 0.79; P = 0.68), age of diabetes diagnosis (H [2, N = 19] = 2.1; P = 0.35), or age of treatment initiation (H [2, N = 19] = 2.2; P = 0.33) among the three groups.

The three groups differed significantly (H [2, N = 19] = 12.78787; P = 0.0017) from each other in graphomotor constructional abilities necessary for accurate copying of the VMI geometric figures (Table 1 and Supplementary Fig. 1). Namely, in children with R201H, scores fell within the normal range (median = 107, lower quartile = 93, and upper quartile = 118.5), whereas children with V59M/A scored low to very low (median = 49, lower quartile = 49, upper quartile = 74.5), and scores of children with other mutations were intermediate (median = 89, lower quartile = 65, and upper quartile = 91).

Age at treatment initiation was significantly inversely correlated with VMI scores only in the V59M/A group (Spearman correlation = −0.79, P < 0.05) (Supplementary Table 1). All three children with V59M/A who had transitioned to sulfonylureas before their first birthday had standard scores >70, whereas the five children whose treatment started later scored <50 (greater than three SDs below the mean).

To our knowledge, our study represents the largest cohort of rare KCNJ11 neonatal diabetic patients undergoing an identical neurodevelopmental assessment at one time. Using a well-validated, age-standardized measure, we show that those with intermediate DEND-associated V59M/A mutations have significant impairment of eye-hand coordination, whereas those with the R201H mutation not associated with neurodevelopmental concerns performed in the normal range. Early treatment with sulfonylureas was associated with better VMI scores in the V59M/A group; however, this observation must be confirmed in larger numbers of patients with more comprehensive longitudinal assessments.

Why mutations such as R201H do not cause a similar level of impairment as seen in V59M/A is uncertain, especially since the diabetes caused by these mutations appears to be equivalent. Given that a few of the R201H and other mutation case subjects exhibited low normal scores, it may be that the VMI is not sensitive enough to quantify mild subclinical difficulty.

Other factors could contribute to the neurodevelopmental concerns in these patients, including diabetic ketoacidosis at diagnosis during a very young age critical for brain development, as well as longer-term metabolic control characterized by prolonged hyperglycemia and sometimes frequent episodes of severe hypoglycemia, which has been associated with a spectrum of visual-spatial, memory, attention, and executive dysfunctions (15).

The current study and previous reports of improved cognitive and motor symptoms after the change in treatment suggest the benefit of sulfonylurea blockade of activated glucose-responsive channels that have an unknown function in the brain (8–10). This implies that sulfonylureas cross the blood-brain barrier; however, the degree to which they do may be a critically important factor and warrants further study.

The three V59M/A subjects who started sulfonylurea therapy before 1 year of age had better VMI scores than the five subjects who did not; however, it remains to be determined whether there is a definitive critical age for treatment. This finding raises hope for prevention of at least some of the neurodevelopmental disability in those who start sulfonylureas at an early age. This bolsters consideration of an empirical trial of sulfonylureas in newly diagnosed neonatal diabetic patients before results of genetic testing are available, given that ATP-sensitive potassium channel mutations cause almost 50% of cases. However, the risk/benefit of sulfonylureas should be carefully considered and should in no way supplant mandatory genetic testing.

Notably, VMI scores in the three V59M/A children treated early were still low and developmental challenges are likely to persist, as is also likely in rare severe DEND cases not assessed in this study. Future research should clarify factors leading to differences in mutation phenotype and outcome. Efforts should include comprehensive specialty assessment and support of optimal progress and continued collection of data on greater numbers of patients tracking long-term neurodevelopmental outcome. In this regard, the VMI may be useful as a quantifiable marker of longitudinal progress.

The authors are grateful for generous support from the Lewis-Sebring Family Foundation and the Kovler Family Foundation and also acknowledge support from the Juvenile Diabetes Research Foundation (9-2008-177), the American Diabetes Association (1-11-CT-41), and the National Institutes of Health (NIH) (P60-DK-020595 and UL1-RR-024999). R.P.S. was supported in part as a Comer Scholar. R.P.S. and M.E.M. were supported in part by the Health Resources and Services Administration of the U.S. Department of Health and Human Services (T 73 MC11047 entitled Leadership Education in Neurodevelopmental Disabilities) and the NIH/the Eunice Kennedy Shriver National Institute of Child Health and Human Development (P30-HD-054275; John P. Kennedy Jr. Intellectual and Developmental Disabilities Research Center).

No potential conflicts of interest relevant to this article were reported.

R.P.S. contributed to study design; collected, analyzed, and interpreted data; and wrote the manuscript. K.S. provided biostatical analysis and interpretation of data. B.C.K. provided technical support and analyzed and interpreted data. S.A.W.G. designed the study; collected, analyzed, and interpreted data; provided administrative and material support; obtained funding; and supervised the study. M.E.M. designed the study, interpreted data, provided administrative and material support, obtained funding, and supervised the study. All authors reviewed and edited the manuscript, contributed to discussion, and approved the final manuscript. S.A.W.G. and M.E.M. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

The authors would like to thank all of the patients, family members, and clinicians who have participated in The University of Chicago Monogenic Diabetes Registry (http://monogenicdiabetes.org) and in this study in particular. The authors thank Graeme Bell and Louis Philipson (both at The University of Chicago, Chicago, IL) and all the staff and volunteers who made the first Monogenic Diabetes Family Meeting in Chicago in 2010 such a success.