The thrifty phenotype hypothesis and hearing problems

BMJ 2003; 327 doi: (Published 20 November 2003) Cite this as: BMJ 2003;327:1199
  1. Marie-Louise Barrenäs, associate professor of otorhinolaryngology (marie-louise.barrenas{at},
  2. Åsa Bratthall, medical doctor1,
  3. Jovanna Dahlgren, assistant professor of pediatrics1
  1. 1Göteborg Pediatric Growth Research Centre, Department of Pediatrics, Institute of the Health of the Woman and Child, Göteborg University, S416 85 Göteborg, Sweden
  1. Correspondence to: M-L Barrenäs
  • Accepted 18 July 2003


While looking for solutions to sensorineural hearing loss (SNHL) induced by age or noise, a serious incurable health problem, we became interested in the thrifty phenotype hypothesis because diseases related to this hypothesis are sometimes those linked to SNHL.1 According to the hypothesis, events during fetal life, such as malnutrition, may cause disease in adulthood. The malnourished fetus makes metabolic adaptations, which may become permanently programmed, persisting throughout life and causing disease later in life. For example, cardiovascular disease,2 hypertension, obesity, and hypercholesterolaemia are related to reduced fetal growth as reflected by a reduced birth size and to SNHL.

The mechanisms behind the thrifty phenotype hypothesis are unclear, but links to insulin-like growth factor I (IGF-I) have been suggested. During development, IGF-I is crucial for several organs. This includes the size of the cochlea and auditory neurones; the innervation of the auditory sensory cells; and the postnatal survival, differentiation, and maturation of auditory ganglion cells.3 Intrauterine growth retarded newborn babies and malnourished pregnancies have lower concentrations of IGF-I. Also, in Turner syndrome SNHL is associated with short stature and lower concentrations of IGF-I.4 To test the thrifty phenotype hypothesis on SNHL, we reanalysed data from two previous samples, assuming that people who are short in stature are over-represented among non-syndromic individuals with SNHL.

Participants, methods, and results

We assessed hearing with standard audiometry in 479 men aged 20 to 64, who were exposed to noise in their jobs, and 500 randomly selected 18 year old male conscripts born in 1974. We had data on body height, weight, exposure to noise, occurrence of hereditary taint for hearing loss, and other medical disorders including use of drugs.

Among the conscripts, using odds ratios, shortness was found twice as often in those with SNHL as in men with normal hearing. SNHL was also associated with a positive heredity for hearing loss but not with noise exposure. Short workers (less than two standard deviations below the mean) had worse hearing than expected by age (below the 10th centile given in ISO 7029), three times more often than taller workers and were 12 times more often taking drugs. To further test the thrifty phenotype hypothesis on hearing, we used multiple linear regression to model the high frequency hearing thresholds (the average of 3, 4, and 6 kHz bilaterally) among the noise exposed workers as a function of body height (cm), age (years), and hypertension (yes or no). Older short men with hypertension had significantly worse hearing (P ≥ 0.01), but among tall men, hypertension had no effect on hearing and the influence of age was less pronounced (R2 = 0.37, hypertension and height adding 9%).


Sensorineural hearing loss (SNHL) in adulthood may be programmed at birth: the thrifty phenotype hypothesis is applicable to SNHL. One mechanism in common to fetal growth retardation, cardiovascular disease, hypertension, and SNHL might be low IGF-I concentrations during fetal life (figure). Indeed, adults with low IGF-I concentrations have features of the metabolic syndrome, a combination of visceral obesity, insulin resistance, dyslipidaemia, and hypertension,5 and, probably, also SNHL. Since IGF-I is a powerful mitogen, a reduced capacity to maximise the speed of the cell cycle during development comes into focus.4 Consequences of a reduced number of cell cycles during development are fewer auditory sensory cells, ganglion cells, and neurones at birth and earlier hearing loss symptoms due to apotosis of the auditory sensory cells induced by age or noise.4 Hypertension and hypercholesterolaemia are commonly believed to cause SNHL; we now consider these confounding superficial markers.


Working hypothesis for sensorineural hearing loss (SNHL) because of fetal events

Shortness indicates that hearing problems in adulthood may be programmed at birth


We thank Alf Axelsson for supplying data from his previous studies. We also thank audiologists from the department of audiology at Sahlgrenska University Hospital, Göteborg, who assessed hearing thresholds.


  • Contributors MLB was principal investigator, designed the study, and abstracted data on workers. All authors collaborated on interpretation of the data and writing the manuscript. MLB is guarantor.

  • Funding No additional funding

  • Competing interests None declared

  • Ethical approval Not needed


View Abstract