Prenatal famine exposure → Insulin resistance (HOMA-IR)

Mechanism

The authors treat the Hunger Winter as a shock to what the fetus received. Midlife oral glucose tests in the Dutch Famine Birth Cohort show higher 120-minute glucose after prenatal exposure. de Rooij, Painter, Roseboom and colleagues gave an OGTT to 702 people at age 50 and 699 at age 58, all born as term singletons just before, during, or after the 1944-1945 famine. At 58, famine-exposed people had 0.4 mmol/l higher 120-minute glucose than unexposed peers (95 percent CI 0.1 to 0.7), after sex and BMI. The authors note that the same cohort already had worse glucose tolerance and higher insulin at 50. Glucose and insulin both rose from 50 to 58. The rise was not faster in the exposed group (p=0.28 for glucose, p=0.09 for insulin). A 94-person intravenous glucose tolerance subsample asked whether that pattern is insulin resistance, a secretion defect, or both. Glucose tolerance (Kg) was 21 percent lower after prenatal famine (95 percent CI -41 to -4). People exposed in mid-gestation had a 53 percent lower disposition index, the product of insulin sensitivity and first-phase insulin response (95 percent CI -126 to -3). Early-gestation exposure pointed the same way and did not reach significance. The authors read the mid- and early-gestation pattern as an insulin secretion defect. Neither abstract reports HOMA-IR. The measures are 120-minute glucose and insulin, IVGTT Kg, and disposition index. Those sit on this page as the atlas home for glucose-insulin homeostasis.

Caveats

The stamp is smoke because nobody randomized a wartime siege. Observation alone cannot close the backdoor, so the stamp stays smoke. More papers of the same design do not settle the claim. We do not put a robust-causal stamp on this page without a feeding trial. That is a rule about design, not a claim that prenatal hunger is harmless. This is the Roseboom, Ravelli, and de Rooij file. The Hunger Winter bundled food shortage with cold, infection, bombardment, and who was able to conceive or survive to midlife. A siege is not a randomized diet. Both papers on this page are available here only as abstracts. The OGTT file is 702 people at 50 and 699 at 58. The IVGTT file is 94 normoglycemic adults from the same cohort. Do not read 94 intravenous tests as a population HOMA-IR. The IVGTT authors concluded a secretion defect after mid-gestation famine. This page is insulin resistance as this atlas uses it. The abstracts do not publish a HOMA-IR number. Models adjust for sex and BMI. Adult BMI can sit on the path from prenatal famine to midlife glucose. Residual confounding remains. Education and later income still sit on who reached the exam and how they lived after the war. Omitted: Ravelli 1999 (BMI and waist at age 50, not glucose or insulin) and Stein 1995 (third-trimester weight gain and birth size). Those are other pages. We do not open a second page here. This is a journal entry, not medical advice.

Effect

qualitative. de Rooij 2006: +0.4 mmol/l 120-min glucose at 58 (n=699). IVGTT Kg -21%; midgestation disposition index -53% (n=94). Worse glucose after famine. Design does not identify.

Nodes

Named confounders

• Educational attainment
• Disposable income

Papers

• Glucose tolerance at age 58 and the decline of glucose tolerance in comparison with age 50 in people prenatally exposed to the Dutch famine.

  SR de Rooij, RC Painter, TJ Roseboom, DI Phillips, C Osmond, DJ Barker, MW Tanck, RP Michels · 2006 · Diabetologia

• Impaired insulin secretion after prenatal exposure to the Dutch famine.

  SR de Rooij, RC Painter, DI Phillips, C Osmond, RP Michels, IF Godsland, PM Bossuyt, OP Bleker · 2006 · Diabetes care