Table 3.: Selected contemporary observations and studies associating DISH with obesity.

Author [Year] No. Subjects DISH Criteria* Obesity Criteria# Percentage Obese Comment
Forestier et al. [1950] [11] 9 M Not stated Not stated 66% Important clinical observation
No controls
Boulet et al. [1954] [29] 8 M / 4 F “Forestier” Not stated 17% No controls
Schilling et al. [1965] [24] 66 M / 14 F
80 controls		 “Forestier” % above German norms 66% >10% overweight, 25%>25% overweight; controls 34% and 9% Obesity strongly associated with DISH
Schoen et al. [1969] [30] 507 patients, > 30yrs with abnormal GT NS – lateral Xray thoracic spine NS “Increased” Compared to 347 patients, >50 yrs with normal GT
Beardwell [1969] [31] 4 M / 4F “Forestier” Not stated “Majority” Familial study
Julkunen et al. [1971] [3] 12,858 unselected persons Julkunen Weight-height ratio, triceps skinfold thickness Significantly higher in both sexes [all age groups] with DISH compared to non-DISH subjects. Obesity strongly associated with DISH
Harris et al. [1974] [4] 17 M / 17 F Harris Not stated 88% No controls
Utsinger et al. [1976] [32] 25 M / 5 F Resnick Not stated 30% No controls
Resnick [1978] [33] 21 M Resnick Not stated 22% No controls
Boachie-Adjei et al. [1987] [34] 21 of 75 autopsy spines “Forestier” Average weight DISH =85kg, non-DISH =65kg Obesity strongly associated with DISH
Troillet and Gerster [1993] [35] 17 M / 8 F, matched for age, sex, BMI
 Resnick BMI Mean BMI = 29.5 All obese
Daragon et al. [1995] [36] 50 persons -DISH >60yrs
50 controls - matched sex, age, weight, height		 Resnick Weight-height index No difference between DISH and controls Study controlled for weight
Mata et al. [1997] [37] DISH -56, Spondylosis -43,
Healthy controls -31		 Resnick Weight, BMI, WC DISH significantly higher weight at marker state** and at study time, higher BMI, higher WC
[p< 0.001] 		Obesity strongly associated with DISH
Coaccioli et al. [2000] [38] Obese subjects- [1] 32 without T2DM, [2] 30 with T2DM
Control - 30 subjects		 Utsinger BMI>30
BMI <30		 [1] 37.5% DISH
[2] 40% DISH
2.5% DISH		 High prevalence DISH in obese subjects
Kiss et al. [2002] [25] 69 M / 62 F with DISH; 69 M/62 F with spondylosis Resnick BMI DISH BMI 27.8, controls 26.0, [p<0.001] Obesity strongly associated with DISH
Miyazama and Akiyama [2006] [39] DISH 35 M / 10 F
Control [spondylosis] 70 M /20 F		 Resnick Weight DISH mean= 62.3 kg, Controls mean =58.33 kg
[p<0.01]		 Obesity strongly associated with DISH
Mader et al. [2009] [27] 13 M / 34 F DISH; 48 age-sex matched controls Resnick WC
BMI > 30 		WC significantly elevated in DISH, BMI > 30 in 39% DISH versus 26% controls Obesity strongly associated with DISH
Zincarelli et al. [2012] [40] DISH 105 M / 27 F
Non-DISH 231 M / 73 F		 Resnick BMI>30 35.6% DISH obese compared with 23% non-DISH
[p< 0.009]		 Obesity strongly associated with DISH-all had cardiovascular disease
Haddad et al. [2013] [28] 57 M / 21 F with DISH and PsA; 171 M / 62 F control patients with PsA only Modified Utsinger [thoracic spine only, no extra spinal assessment included] BMI
Obesity definition not stated		 DISH BMI =32.9%, control BMI =28.7%, p<0.0001
DISH obesity=31%, control obesity=50%. [p<0.0001]		 Higher BMI strongly associated with DISH, large variance in obesity groups

* M = males, F = females, # Weight-height ratio= weight in kilograms/height in centimetres-100, WC = waist circumference, BMI = Body Mass Index, GT = glucose tolerance, PsA =psoriatic arthritis, * “Forestier” indicates that diagnosis based on typical features of Forestier et al. description [11], ** marker state =weight at time of first marriage.