The new england journal of medicine a prospective, randomized, controlled trial. J Neurosurg 2003; in cerebrospinal fluid of adult patients receiving adjunctive cor- 98:725-30. ticosteroids to treat pneumococcal meningitis: a prospective 2. Govender ST, Nathoo N, van Dellen JR. Evaluation of an multicenter observational study. Clin Infect Dis 2007;44:250-5. antibiotic-impregnated shunt system for the treatment of hydro- 5. Rybak M, Lomaestro B, Rotschafer JC, et al. Therapeutic cephalus. J Neurosurg 2003;99:831-9. monitoring of vancomycin in adult patients: a consensus review 3. Ratilal B, Costa J, Sampaio C. Antibiotic prophylaxis for sur- of the American Society of Health-System Pharmacists, the In- gical introduction of intracranial ventricular shunts. Cochrane fectious Diseases Society of America, and the Society of Infec- Database Syst Rev 2006;3:CD005365. tious Diseases Pharmacists. Am J Health Syst Pharm 2009;66: 4. Ricard JD, Wolff M, Lacherade JC, et al. Levels of vancomycin 82-98. Large Islets, Beta-Cell Proliferation, and a Glucokinase Mutation To the Editor: Rare, naturally occurring gene of control subjects or subjects with KATP-related mutations provide important insights into nor- hypoglycemia. The relative beta-cell area (the per- mal human physiology. We report on a young girl centage of the histologic section that stained posi- with severe neonatal hypoglycemia due to a novel tive for insulin) was 2.9% in the head of the pan- glucokinase mutation (V91L). Her father had a creas and 6.7% in the tail of the pancreas in the similar clinical course, but neither his DNA nor patient, as compared with 1.8% in the pancreases his pancreatic tissue was available for study. of the control subjects and 1.1% in the pancre- V91L showed a markedly increased affinity for ases of the subjects with KATP-related hypoglyce- glucose that was more than 8.5 times as high, an mia. In the patient’s pancreas, nine proliferating enzyme efficiency that was 7 times as high, and (Ki67-positive) beta cells were detected in 100 is- a relative-activity index that was 30 times as high lets (Fig. 1E) and apoptotic (terminal deoxynucleo- as that of the wild-type enzyme. The estimated tidyl transferase dUTP biotin nick end labeling threshold for glucose-stimulated insulin secretion [TUNEL]-positive) beta cells were observed clus- was markedly lower than that of the wild-type enzyme (0.96 vs. 5.00 mmol per liter). Diazoxide Figure 1 (facing page). Histologic Features of the Pan- and octreotide therapy did not control the pa- creas in the Patient and in Age-Matched Control Sub- tient’s hypoglycemia, and a subtotal pancreatecto- jects with Normal Pancreases. my was performed when she was 3 years of age. Panel A (hematoxylin and eosin) shows a large islet Quantitative histologic examination revealed (dotted line) near blood vessels (BV) in a biopsy speci- men obtained from the patient, and Panel B (hematox- abnormally large islets (Fig. 1A) with some beta ylin and eosin) shows a normal-size islet (dotted line) cells containing a large nucleus (Fig. 1C). The in a specimen obtained from a control subject; both mean islet area was 7705 μm2 in the head of the Panel A and Panel B are specimens shown at low magni- pancreas and 7048 μm2 in the tail of the pan- fication. Panel C shows abnormally large beta-cell nuclei creas. Both areas were significantly larger than (arrows) in the patient’s specimen at higher magnifica- tion than in Panel A, and Panel D shows beta-cell nuclei those in specimens obtained from five age-matched in a control subject’s specimen at the same magnifica- control subjects who did not have pancreatic dis- tion. Panel E (Cy5) shows Ki67-positive beta cells (reddish- ease (range, 1160 to 1997 μm2; P<0.001) and in purple nuclei, arrows) in a large islet in a specimen ob- three specimens obtained from two age-matched tained from the patient; Panel F shows beta cells in a subjects with diffuse hypoglycemia due to ABCC8 specimen obtained from a control subject. In Panels E and F, Cy2-labeled insulin shows green cytoplasmic stain- gene mutations (ATP-sensitive potassium-chan- ing. Panel G shows terminal deoxynucleotidyl transferase nel [KATP ]–related hypoglycemia) (range, 769 to dUTP biotin nick end labeling–positive cells (green nu- 859 μm2; P<0.001). All these subjects were in clei, arrows) in the patient’s specimen; these cells were subgroups of a population from a previously pub- not detected in any of the specimens obtained from con- lished study.1 Approximately 10% of the patient’s trol subjects (a specimen from one control subject is 2 shown in Panel H). In Panels G and H, Cy5-labeled insu- islets were larger than 13,000 μm ; this was lin shows red cytoplasmic staining. generally larger than any islet in the pancreases 1348 n engl j med 362;14 nejm.org april 8, 2010 The New England Journal of Medicine Downloaded from www.nejm.org at UNIV MIAMI MILLER SCH MED on October 13, 2010. For personal use only. No other uses without permission. Copyright © 2010 Massachusetts Medical Society. All rights reserved. correspondence A Patient B Age-matched control subjects BV 50 µm 50 µm C D 10 µm 10 µm E F 10 µm 10 µm G H 10 µm 10 µm n engl j med 362;14 nejm.org april 8, 2010 1349 ICM AUTHOR Cuesta-Munoz RETAKE 1st The New England Journal of Medicine REG F FIGURE 1a-h 2nd Downloaded from www.nejm.org at UNIV MIAMI MILLER SCH MED on October 13, 2010.3rd For personal use only. No other uses without permission. CASE TITLE Copyright © 2010 Massachusetts Medical Society.Revised All rights reserved. EMail Line 4-C SIZE Enon ARTIST: mst H/T H/T FILL Combo 33p9 AUTHOR, PLEASE NOTE: Figure has been redrawn and type has been reset. Please check carefully. JOB: 36214 ISSUE: 4-8-10 corrections tering within some islets (Fig. 1G). Neither Ki67- Jacques Rahier, M.D., Ph.D. positive nor TUNEL-positive cells were seen in Université Catholique de Louvain any of the pancreases of the control subjects Brussels, Belgium (Fig. 1F and 1H), suggesting that increased intra- Pablo Rodríguez-Bada, M.Sc. cellular glucose flux stimulates both proliferation Nadia Cobo-Vuilleumier, M.Sc. and apoptosis pathways. Antonio Luis Cuesta-Muñoz, M.D., Ph.D. Normal histologic findings in four previous Carlos Haya University Hospital cases of glucokinase-related hypoglycemia have Málaga, Spain [email protected] been reported, but none of these patients under- Supported by grants (MICINN, P.I3, SAF2006-12863, SAS/ went detailed quantitative morphometric analy- PI-024/2007, and SAS/PI-0236/2009, to Dr. Cuesta-Muñoz) from 2-4 sis. In a previously reported case that included the Junta de Andalucía, and a grant (to Dr. Glaser) from the Rus- quantitative histologic analysis, a similar increase sell Berrie Foundation and D-Cure, Diabetes Care in Israel. Disclosure forms provided by the authors are available with in the mean islet profile was confirmed (it was the full text of this letter at NEJM.org. 2.5 times larger than that of control subjects and 8.0 to 10.0 times larger than that of patients with 1. Kassem SA, Ariel I, Thornton PS, et al. Beta-cell prolifera- 5 tion and apoptosis in the normal fetal and neonatal human pan- a KATP-channel deficiency). In both that patient creas and in persistent hyperinsulinemic hypoglycemia of in- and our patient, the routine pathology report did fancy (PHHI). Diabetes 2000;49:1325-33. not indicate any abnormality in islet size; this 2. Gloyn AL, Noordam K, Willemsen MA, et al. Insights into the biochemical and genetic basis of glucokinase activation emphasizes the importance of quantitative mor- from naturally occurring hypoglycemia mutations. Diabetes phometric analysis to determine islet size. 2003;52:2433-40. Thus, histologic findings in infants with hy- 3. Sayed S, Langdon DR, Odili S, et al. Extremes of clinical and enzymatic phenotypes in children with hyperinsulinism caused perinsulinemic hypoglycemia may differ accord- by glucokinase activating mutations. Diabetes 2009;58:1419-27. ing to the genetic cause of the condition. Further- 4. Wabitsch M, Lahr G, Van de Bunt M, et al. Heterogeneity in more, intracellular glucose f lux appears to regulate disease severity in a family with a novel G68V GCK activating mutation causing persistent hyperinsulinaemic hypoglycaemia proliferation and apoptosis in human beta cells; of infancy. Diabet Med 2007;24:1393-9. this is consistent with previous findings in mu- 5. Cuesta-Muñoz AL, Huopio H, Otonkoski T, et al. Severe per- rine models. Small-molecule activators of glu- sistent hyperinsulinemic hypoglycemia due to a de novo glu- cokinase mutation. Diabetes 2004;53:2164-8. cokinase are currently being developed for the Correspondence Copyright © 2010 Massachusetts Medical Society. clinical management of diabetes. Although ob- servations in this young child with a congenital glucokinase mutation may not be directly appli- cable to adults with diabetes, the effect of these corrections glucokinase activators on human beta-cell mass may be of interest. Sudden Death in Myotonic Dystrophy (October 9, 2008;359: Sameer Kassem, M.D., Ph.D. 1626-9). In Panel B of Figure 1 (page 1627), under the Controls Maayan Heyman, M.L.T. heading, the value for N should have been 14 rather than 7. We regret the error. The article has been corrected at NEJM.org. Benjamin Glaser, M.D. Case 13-2009: A 54-Year-Old Woman with Respiratory Failure Hadassah–Hebrew University Medical Center and a Cavitary Lesion in the Lung (April 23, 2009;360:1770-9).