Body mass estimation from dimensions of the fourth lumbar vertebra in middle-aged Finns. (September 2019)

• Record Type: Journal Article
• Authors: Korpinen, Niina; Keisu, Asla; Niinimäki, Jaakko; Karppinen, Jaro; Niskanen, Markku; Junno, Juho-Antti; Oura, Petteri
• Is Part Of: Legal medicine. Volume 40(2019)
• Page Start: 5
• Page End: 16
• Publication Date: 2019-09
• Subjects: Body mass estimation -- L4 -- Vertebral dimensions -- Magnetic resonance imaging -- Forensic anthropology -- Population data; BMI Body mass index -- CSA Cross-sectional area -- ICC Intra-class correlation -- L1—L5 The lumbar vertebrae -- MRI Magnetic resonance imaging -- NFBC1966 Northern Finland Birth Cohort 1966 -- OLS Ordinary least squares regression -- PI Prediction interval -- R Pearson's correlation coefficient -- RMA Reduced major axis regression -- SD Standard deviation -- SEE Standard error of the estimate -- TEM Technical error of measurement; Medical jurisprudence -- Periodicals; Forensic Medicine -- Periodicals; Médecine légale -- Périodiques; Medical jurisprudence; Periodicals; 614.1
• DOI: 10.1016/j.legalmed.2019.06.008
• Languages: English
• ISSNs: 1344-6223

Abstract

Highlights: We studied the fourth lumbar vertebra (L4) in body mass estimation among Finns. Equations for body mass were generated for the full sample and for normal-weight individuals. In our data, body mass was clearly associated with all the L4 parameters. Vertebral cross-sectional area had the highest predictive value in body mass estimation. L4 dimensions are potentially useful in body mass estimation. Abstract: Although body mass is not a stable trait over the lifespan, information regarding body size assists the forensic identification of unknown individuals. In this study, we aimed to study the potential of using the fourth lumbar vertebra (L4) for body mass estimation among contemporary Finns. Our sample comprised 1158 individuals from the Northern Finland Birth Cohort 1966 who had undergone measurements of body mass at age 31 and 46 and lumbar magnetic resonance imaging (MRI) at age 46. MRI scans were used to measure the maximum and minimum widths, depths, and heights of the L4 body. Their means and sum were calculated together with vertebral cross-sectional area (CSA) and volume. Ordinary least squares (OLS) and reduced major axis (RMA) regression was used to produce equations for body mass among the full sample (n = 1158) and among normal-weight individuals (n = 420). In our data, body mass was associated with all the L4 size parameters (R = 0.093–0.582, p ≤ 0.019 among the full sample; R = 0.243–0.696, p ≤ 0.002 among the normal-weight sample). RMA regression models seemed to fit the data better than OLS, with vertebral CSA having the highest predictive value in body mass estimation. In the full sample, the lowest standard errors were 6.1% (95% prediction interval ±9.6 kg) and 7.1% (±9.1 kg) among men and women, respectively. In the normal-weight sample, the lowest errors were 4.9% (±6.9 kg) and 4.7% (±5.7 kg) among men and women, respectively. Our results indicate that L4 dimensions are potentially useful in body mass estimation, especially in cases with only the axial skeleton available.