Clinical Rheumatology

, Volume 37, Issue 8, pp 2187–2193 | Cite as

The association between the Mediterranean diet and magnetic resonance parameters for knee osteoarthritis: data from the Osteoarthritis Initiative

  • Nicola Veronese
  • Luciana La Tegola
  • Gaetano Crepaldi
  • Stefania Maggi
  • Domenico Rogoli
  • Giuseppe Guglielmi
Original Article


The Mediterranean diet appears to be beneficial for osteoarthritis (OA), but the few data available regarding the association between the diet and the condition are limited to X-ray and clinical findings. The current study aimed to investigate the association between adherence to the Mediterranean diet and knee cartilage morphology, assessed using magnetic resonance (MRI) in a cohort of North American participants. Seven hundred eighty-three participants in the Osteoarthritis Initiative (59.8% females; mean age 62.3 years) in possession of a MRI assessment (a coronal 3D FLASH with Water Excitation MR sequence of the right knee) were enrolled in our cross-sectional study. Adherence to the Mediterranean diet was evaluated using a validated Mediterranean diet score (aMED). The strength of the association between aMED and knee MRI parameters was gauged using an adjusted linear regression analysis, expressed as standardized betas with 95% confidence intervals (CIs). Using an adjusted linear regression analysis, each increase of one standard deviation (SD) in the aMED corresponded to a significant increase in the central medial femoral cartilage volume (beta = 0.12; 95%CI 0.09 to 0.15), in the mean central medial femoral cartilage thickness (beta = 0.13; 95%CI 0.01 to 0.17), in the cartilage thickness of the mean central medial tibiofemoral compartment (beta = 0.12; 95%CI 0.09 to 0.15), and in the cartilage volume of the medial tibiofemoral compartment (beta = 0.09; 95%CI 0.06 to 0.12). Higher adherence to a Mediterranean diet was found to be associated with a significant improvement in knee cartilage as assessed by MRI, even after adjusting for potential confounding factors.


Aged Healthy aging Knee osteoarthritis Lifestyle Mediterranean diet MRI 


With a worldwide prevalence estimated as 10% in men and 20% in women over 60, osteoarthritis (OA) is the most common cause of musculoskeletal disability in elderly individuals [1]. The prevalence of this degenerative joint disease is increasing because the proportion of aging individuals continues to grow. While plain radiographs have low sensitivity and specificity, they nevertheless continue to be the modality of choice for initial assessments of OA [1]. Recent studies have begun to take into consideration magnetic resonance imaging (MRI) to improve early detection based on both semi-quantitative and quantitative structural and compositional measurements delineating the three-dimensional distribution of articular cartilage thickness [2, 3].

Given the high cost and scarcity of this technology, MRI assessment is not routinely used in the clinical management of OA patients. It has become nevertheless an important imaging tool for OA research [4, 5] because it can investigate structures not visualized by radiography (e.g., articular cartilage, menisci, ligaments, synovium, capsular structures, fluid collection, and bone marrow) [6, 7, 8, 9] and assess pathologic alterations characteristic of pre-radiographic disease stages as well as the knee joint as an entity on its own. These are important considerations as it is well known that there are two primary types of risk factors for OA: systemic (e.g., age, gender, ethnicity, genetics, and dietary variables) and local factors (previous injury, occupational activities, obesity, mechanical loading) [4].

Although numerous studies have demonstrated the factors affecting joints, few have investigated the effect of diet on OA. One randomized controlled trial involving 399 overweight participants found that an intensive diet-induced weight loss improved knee OA symptoms even to a greater extent with respect to that in participants following an intensive exercise program [5]. Given that intensive weight loss programs are usually difficult to follow [6], other dietary patterns might be effective in reducing knee OA and its symptomatology. We recently reported that higher adherence to the Mediterranean diet was associated with a lower prevalence of knee OA, as detected by radiological and clinical instruments [7].

Given the potential benefits of the Mediterranean diet in connection to several diseases (including cardiovascular [8], metabolic [10], musculoskeletal [11], and neurological [12] ones) and since no study, to our knowledge, has assessed the potential association between the Mediterranean diet and knee OA using MRI assessment, the current one sets out to investigate the association between adherence to a Mediterranean diet and morphological parameters of knee joint cartilage in a large cohort of North Americans.

Patients and methods

Data source and subjects

The data that was analyzed by the current study was gathered from the Osteoarthritis Initiative (OAI) database, a publicly available platform ( containing information from participants residing in four cities in the USA (Baltimore, MD; Pittsburgh, PA; Pawtucket, RI; and Columbus, OH) who were enrolled February 2004 and May 2006. Individuals were eligible for registration in the initiative if they met at least two of the following requirements: (1) had knee OA with knee pain for a 30-day period in the past 12 months or (2) were at high risk of developing knee OA [13]. The data upon which our analyses were based were recorded during the baseline and screening evaluations carried out during the month of November 2008.

After being informed about the study’s aims and modality, the participants provided informed written consent. The OAI study was granted full ethical approval by the institutional review board of the OAI Coordinating Center at University of California, San Francisco.

Adherence to the Mediterranean diet (exposure)

The individual’s dietary pattern was analyzed using the Block Brief 2000 food frequency (FFQ) questionnaire [14] which assesses 70 food/beverage items. The frequency that foods/beverages were consumed was quantified at nine levels ranging from “never” to “every day.” The questionnaire devotes seven questions to variables such as food preparation methods and fat intake; there is one question on fiber intake, and 13 questions investigate vitamin and mineral intakes.

Adherence to the Mediterranean diet was evaluated using the Mediterranean diet score (aMED), as outlined by Panagiotakos et al. [15]. The score is based on a food frequency questionnaire that was administered during the baseline OAI visit. The aMED takes into consideration several food groups generally considered an integral part of the Mediterranean diet. Adherence to each food group has a possible score ranging from 0 (less adherent) to 5 (better adherence); the total score ranges from 0 to 55, with higher values indicating a stricter adherence to a Mediterranean diet. Cereals (e.g., bread, pasta, rice), potatoes, fruits, vegetables, legumes (e.g., peas, beans), and fish were quantified as to servings/month in the following manner: 0 = never; 1 = 1 to 4 servings per month; 2 = 5 to 8; 3 = 9 to 12; 4 = 13 to 18; 5 = more than 18 servings/month. Since no information regarding the consumption of whole vs. refined cereals was collected, all types of grains were placed under the same heading. The consumption of red meat, poultry, and full-fat dairy products (e.g., milk cheese, yogurt) was quantified in the following manner: 0 = more than 18 servings/month; 1 = 13 to 17 servings per month; 2 = 9 to 12; 3 = 5 to 8; 4 = 1 to 4; 5 = never. The use of olive oil was quantified as the times it was used a week: 0 = never; 1 = rare; 2 ≤ 1/weekly; 3 = 2 times/weekly; 4 = 3 to 6; 5 = daily. Alcoholic beverages were quantified as: 0 ≥ 700 ml/day; 1600 to 699 ml/day; 2 = 500 to 599 ml/day; 3 = 400 to 499 ml/day; 4 = 300 to 399 ml/day; 5 = < 300 ml/day. The scores were already utilized in the context of the OAI [7, 16].

The population was classified into quartiles using 25, 28, and 32 points.


The OAI examined coronal 3D FLASH images with Water Excitation MR sequence of the osteoarthritic right knee. It is a double oblique 3D fast low angle shot (FLASH) with water excitation, a slice thickness of 1.5 mm and an in-plane resolution of 0.31 mm × 0.31 mm [17] which was acquired at 3 T (Siemens Magnetom Trio, Erlangen, Germany) using quadrature transmit–receive knee coils (USA Instruments, Aurora, OH, USA).

The MR images were quality controlled and converted into a proprietary format at the image analysis center (Chondrometrics GmbH, Ainring, Germany). Images were read in pairs by seven technicians who were blinded to the acquisition order and experienced in segmenting femorotibial cartilage plates [18]. The total subchondral bone area and the cartilage joint surface area (AC) of the medial tibia (MT), the lateral tibia (LT), the central (weight-bearing) medial femoral condyle (cMF), and the central lateral femoral condyle (cLF) were drawn manually during the segmentation process. The weight-bearing region of the femoral condyles was analyzed between the intercondylar notch, and 60% of the distance to the posterior end of the femoral condyles used as the anatomical landmark. For total cartilage plates and for the central subregions, were also reported the medial (MFTC) and lateral (LFTC) femur-tibial compartments, by using summed values from MT and cMF, and LT and cLF, respectively. The quality of the segmentations were checked by a single expert [18].

The total subchondral bone area, the AC, the part of the subchondral bone covered with cartilage (cAB), the denuded subchondral bone area (dAB), the cartilage volume (VC), the mean cartilage thickness over the cAB (ThCcAB, not including denuded areas), and the mean cartilage thickness over the entire subchondral bone area (ThCtAB, including denuded areas as 0 mm cartilage thickness) were then computed [18]. Additional details concerning these procedures are available from the OAI imaging protocol [19].

The current study was particularly interested in the ThCtAB and the VC of the MT and LT, in cLF and cMF, and in the LFTC and MFTC since they seem to be the best predictors of greater volume and thickness of the bone–cartilage interface in the knee [18].


We identified ten potential self-reported confounders that we considered when assessing the relationship between aMED and knee MRI parameters. These included body mass index (BMI), physical activity evaluated using the Physical Activity Scale for the Elderly (PASE) [20], race, smoking habit, educational level, and yearly income (< or > $50,000 and missing data).

Validated general health measures of self-reported comorbidities were assessed using the modified Charlson comorbidity score [21].

Statistical analyses

Continuous variable data were normally distributed according to the Kolmogorov–Smirnov test. Data were presented as mean and standard deviation (SD) values for quantitative measures and frequency and percentages for all discrete variables. Levene’s test was used to test the homoscedasticity of variances, and if its assumption was violated, Welch’s ANOVA was used. P values were calculated using the Jonckheere–Terpstra test [22] for continuous variables and the Mantel–Haenszel chi-square test for categorical ones.

To assess the relationship between aMED and knee MRI parameters, a linear regression analysis was conducted using knee MRI parameters as the “outcomes” and aMED as the “exposure.” The basic model was not adjusted for any confounders. The fully adjusted model included the following adjustments: age (as continuous), sex, race (whites vs. others), body mass index (as continuous), education (degree vs. others), smoking habits (current/previous vs. others), yearly income (categorized as ≥ or < 50,000$ and missing data), Charlson comorbidity index, Physical Activity Scale for the Elderly (as continuous), total energy intake (as continuous). Multi-collinearity among covariates was assessed through variance inflation factor (VIF) [23], using a cut-off of two as reason of exclusion. No covariates were excluded, however, for this reason. The covariates were selected among those factors significantly associated with at least three outcomes in univariate analyses, using a P value < 0.10 for inclusion. Adjusted standardized betas with their 95% confidence intervals (CIs) were calculated to estimate the strength of the associations between aMED (increase in one standard deviation = 5 points) and knee MRI parameters.

Several sensitivity analyses were conducted evaluating the interaction between DII and selected factors (such as median age, sex, the presence of comorbidities, the presence of clinical/radiological knee OA, median BMI, education, income, smoking status, race) in the association with knee MRI parameters, but none emerged as a moderator of our findings (all P values > 0.10 for the interaction).

A P < 0.05 was considered statistically significant. All analyses were performed using SPSS software version 21.0 for Windows (SPSS Inc., Chicago, IL).


Sample selection

The OAI dataset is based upon a total of 4796 North American participants in the OAI. Of these, 906 had already undergone a MRI of the knee. Eighty-four were excluded because they did not have aMED data and 39 because their self-reported caloric intake was unreliable (less than 800/greater than 4200 for men; less than 500/greater than 3800 for women), and 783 were included in our analyses.

Descriptive characteristics

Four hundred sixty-eight out of the 783 participants (59.8%) were females (mean age = 62.3 years; ± 9.4 years; range 45–79). The participants’ mean aMED score was 28.3 points (4.9 points; range 11–44).

Table 1 illustrates the baseline characteristics in the subjects classified into aMED quartiles. Those in the highest quartile (reflecting higher adherence to Mediterranean diet) were significantly older (P for trend = 0.004), more likely to be female (P for trend = 0.04), whites (P for trend < 0.0001), and more likely to have a higher educational level (P for trend = 0.02) and income (P for trend = 0.02) than the other participants. Those in the highest aMED quartile had lower BMI values (P for trend = 0.002) and fewer medical morbidities (P for trend = 0.03) (Table 1).
Table 1

Descriptive findings of the participants by adherence to Mediterranean diet






P valuea

(n = 219)

(n = 175)

(n = 234)

(n = 155)

aMED ≤ 25

aMED 26–28

aMED 29–32

aMED ≥ 33

General characteristics

 Energy intake (Kcal/day)

1407 (584)

1394 (532)

1459 (561)

1428 (489)


 Age (years)

60.1 (9.0)

63.3 (9.4)

63.4 (9.6)

62.5 (9.4)


 PASE (points)

158 (78)

158 (77)

154 (75)

156 (80)


 Females (n, %)

122 (55.7)

102 (58.3)

140 (59.8)

104 (67.1)


 White race (n, %)

174 (79.5)

148 (84.6)

203 (86.8)

142 (91.6)

< 0.0001

 Smoking (previous/current) (n, %)

97 (44.9)

82 (47.1)

107 (45.9)

73 (47.4)


 Graduate degree (n, %)

53 (24.2)

40 (22.9)

75 (32.1)

51 (32.9)


 Yearly income (< 50,000 $) (n,%)

53 (24.2)

40 (22.9)

75 (32.1)

51 (32.9)


Medical conditions

 BMI (Kg/m2)

30.2 (4.9)

29.5 (4.5)

28.8 (4.5)

29.1 (4.9)


 Charlson comorbidity index (points)

0.5 (1.0)

0.5 (1.0)

0.3 (0.7)

0.4 (0.7)


The data are presented as means (with standard deviations) for continuous variables and number (with percentage)

PASE Physical Activity Scale for the Elderly, BMI body mass index

aP values for trends were calculated using the Jonckheere–Terpstra test for continuous variables and the Mantel–Haenszel chi-square test for categorical ones

Adherence to Mediterranean diet and knee MRI parameters

Table 2 outlines the association between aMED and knee MRI parameters. Using a linear regression analysis, adjusted for ten potential confounders, stricter adherence to Mediterranean diet (expressed as increase in one SD = 5 points) corresponded to a significant increase in the central medial femoral cartilage volume (beta = 0.12; 95%CI 0.09 to 0.15; P < 0.0001), in the mean central medial femoral cartilage thickness (beta = 0.13; 95%CI 0.01 to 0.17; P < 0.0001), in the mean cartilage thickness of the central medial tibiofemoral compartment (beta = 0.12; 95%CI 0.09 to 0.15; P < 0.0001), and in cartilage volume of the medial tibiofemoral compartment (beta = 0.09; 95%CI 0.06 to 0.12; P = 0.001) (Table 2).
Table 2

Association between adherence to Mediterranean diet (aMED) magnetic resonance parameters


Unadjusted linear regression

Fully adjusted1 linear regression



95%CI (beta)

P value


95%CI (beta)

P value

Mean cartilage thickness - medial tibia


− 0.01 to 0.09



0.06 to 0.13


Volume of cartilage - medial tibia

− 0.002

− 0.03 to 0.03



0.04 to 0.10


Volume of cartilage - lateral tibia

− 0.07

− 0.11 to − 0.03



− 0.004 to 0.04


Mean cartilage thickness—lateral tibia

− 0.03

− 0.07 to 0.006



− 0.004 to 0.08


Volume of cartilage - central lateral femur

− 0.10

− 0.14 to − 0.07


− 0.04

− 0.07 to 0.01


Mean cartilage thickness - central lateral femur

− 0.08

− 0.12 to − 0.04


− 0.04

− 0.07 to 0.009


Volume of cartilage - central medial femur


− 0.007 to 0.07



0.09 to 0.15

< 0.0001

Mean cartilage thickness - central medial femur


0.05 to 0.13



0.01 to 0.17

< 0.0001

Mean cartilage thickness - lateral tibial-femoral compartment

− 0.06

− 0.01 to 0.02



− 0.02 to 0.03


Cartilage volume - medial tibial-femoral compartment

− 0.08

− 0.11 to − 0.05



− 0.03 to 0.03


Mean cartilage thickness - central medial tibial-femoral compartment


0.04 to 0.12



0.09 to 0.15

< 0.0001

Cartilage volume - medial tibial-femoral compartment


− 0.03 to 0.07



0.06 to 0.12


Data are presented as standardized beta with 95% confidence intervals and corresponding P values and obtained through a linear regression analysis taking the increase in one standard deviation of adherence to Mediterranean diet as independent variable. Significant findings (P value < 0.05), after fully adjustment

1Fully adjusted model included as covariates: age (as continuous), sex, race (whites vs. others), body mass index (as continuous), education (degree vs. others), smoking habits (current and previous vs. others), yearly income (categorized as ≥ or < 50,000$ and missing data), Charlson comorbidity index, Physical Activity Scale for the Elderly (as continuous), total energy intake (as continuous)

The sensitivity analyses (i.e., stratification for median age, sex, presence of any comorbidities, the presence/absence of radiological/clinical knee OA, the median BMI, education, income, smoking status, race) did not show any significant interaction between aMED and these factors in predicting knee MRI parameters (details not shown, available upon request).


An analysis of data based on North American participants at risk of OA utilized by our study found that those who were more adherent to a Mediterranean diet had a significant increase in the central region of the medial femoral cartilage volume, the mean central medial femoral cartilage thickness, the mean cartilage thickness of the central medial tibiofemoral compartment, and the cartilage volume of the medial tibiofemoral compartment. All in all, these findings suggest that the Mediterranean diet is associated with better knee morphology even independently from several confounders.

A descriptive analysis of the data showed that the participants who had a stricter adherence to a Mediterranean diet had significantly lower BMI values and fewer medical morbidities as well as a higher education level and income with respect to the other participants. It can and has been argued that these factors themselves may have affected the knee joint cartilage volume and thickness values found in that subset of the population studied [24, 25]. But after adjustment of potential confounders (including severity of comorbidity and social and economic factors), the association between aMED and the MRI knee parameters assessed still remained statistically significant.

Study findings uncovered a significant association between a Mediterranean diet and knee joint cartilage thickness and volume, which suggests that it may have a protective effect on knee OA. In fact, as many studies have demonstrated [26, 27], cartilage thinning and loss are early features of OA that are typically detected by knee MRI scans. Baseline cartilage thickness and volume is, of course, are important factor in the onset and progression of knee osteoarthritis. The implication of these findings is that persons who have a higher risk of developing knee OA should adopt a Mediterranean diet which could delay the onset or attenuate the early phase of this condition.

The finding that the association with the Mediterranean diet was stronger in the medial with respect to the lateral compartment was an interesting, unexpected study upshot. We can hypothesize that the medial meniscus and compartment are larger than those on the lateral side, and from a biomechanical point of view, the forces applied on them are stronger leading to a more pronounced effect on that compartment.

Some hypotheses can be put forward to explain these findings; firstly, a stricter adherence to a Mediterranean diet has been linked to a decrease in inflammation [24] which plays an important role in destroying knee cartilage [28]. The anti-inflammatory properties typical of a Mediterranean diet may thus decrease inflammation and lead to a better conservation of knee structures [14]. Secondly, a Mediterranean diet seems to lower oxidative stress markers [26] which may influence the onset of OA by enhancing collagen type II and aggrecan expression levels and inhibiting apoptosis-related protein expression that have a chondroprotective effect [27, 28]. Lastly, a Mediterranean diet could play a role in remodeling the extracellular matrix (ECM) [29], thus promoting the effective repair of a structure which is frequently defective in those who develop OA. All these factors could play an important role in the development of knee OA and provide a physiological rationale for the findings outlined here.

The effect of a Mediterranean diet on weight (confirmed by our own analysis detecting a lower BMI in those participants with a stricter adherence to a Mediterranean diet) could be secondary as far as structural damage of the knee is concerned, and this hypothesis was recently been proposed by the authors of a large trial involving 454 obese and overweight participants whose weight loss was not associated to any improvement in MRI knee parameters [29].

The study’s primary limitation was its cross-sectional nature which precluded any possibility of examining a potential causal relationship between a Mediterranean diet and knee MRI parameters. Second, it did not examine biohumoral markers (e.g., oxidative stress) and their effect on the association between a Mediterranean diet and knee status, but they may indeed be relevant. A third limitation is the self-reported assessment of medical conditions, another cause for possible bias. Finally, we used a slightly modified version of a previous Mediterranean diet adherence measurement [17], yet another decision that may have introduced bias. As far as study strengths are concerned, we can list the large sample size, and the fact this is the first epidemiological study reporting data on a Mediterranean diet’s impacts on knee status assessed by means of MRI.

To conclude, our findings confirm that a stricter adherence to a Mediterranean diet is associated with a significant increase in MRI-proven knee joint cartilage and volume, suggesting that it could play an important role in preventing OA, even after taking into consideration several important confounders. Further longitudinal studies are warranted to confirm/refute these results and explore potential pathophysiological mechanisms underlying these findings.


Statement of authorship

Analysis and interpretation of data: Veronese, La Tegola. Draft of the article: La Tegola, Crepaldi. Critical revision for important intellectual content: Maggi, Guglielmi. Statistical analysis: Veronese. All authors approved the version submitted.

Funding source

The OAI is a public–private partnership comprised of five contracts (N01-AR-2-2258; N01-AR-2-2259; N01-AR-2-2260; N01-AR-2-2261; N01-AR-2-2262) funded by the National Institutes of Health, a branch of the Department of Health and Human Services, and conducted by the OAI Study Investigators. Private funding partners include Merck Research Laboratories; Novartis Pharmaceuticals Corporation, GlaxoSmithKline; and Pfizer, Inc. Private sector funding for the OAI is managed by the Foundation for the National Institutes of Health.

Compliance with ethical standards

The OAI study was granted full ethical approval by the institutional review board of the OAI Coordinating Center at University of California, San Francisco.




This manuscript was prepared using an OAI public use dataset and does not necessarily reflect the opinions or views of the OAI investigators, the NIH, or the private funding partners. The sponsors had no role in the design, methods, subject recruitment, data collection, analysis, or preparation of this study.


  1. 1.
    Hiligsmann M, Cooper C, Arden N, Boers M, Branco JC, Luisa Brandi M, Bruyère O, Guillemin F, Hochberg MC, Hunter DJ, Kanis JA, Kvien TK, Laslop A, Pelletier JP, Pinto D, Reiter-Niesert S, Rizzoli R, Rovati LC, Severens JL(H), Silverman S, Tsouderos Y, Tugwell P, Reginster JY (2013) Health economics in the field of osteoarthritis: an expert’s consensus paper from the European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO). Semin Arthritis Rheum 43:303–313CrossRefPubMedGoogle Scholar
  2. 2.
    Demehri S, Hafezi-Nejad N, Carrino JA (2015) Conventional and novel imaging modalities in osteoarthritis: current state of the evidence. Curr Opin Rheumatol 27:295–303CrossRefPubMedGoogle Scholar
  3. 3.
    Demehri S, Guermazi A, Kwoh CK (2016) Diagnosis and longitudinal assessment of osteoarthritis: review of available imaging techniques. Rheum Dis Clin N Am 42:607–620CrossRefGoogle Scholar
  4. 4.
    Palazzo C, Nguyen C, Lefevre-Colau MM, Rannou F, Poiraudeau S (2016) Risk factors and burden of osteoarthritis. Ann Phys Rehabil Med 59:134–138CrossRefPubMedGoogle Scholar
  5. 5.
    Messier SP, Mihalko SL, Legault C, Miller GD, Nicklas BJ, DeVita P, Beavers DP, Hunter DJ, Lyles MF, Eckstein F, Williamson JD, Carr JJ, Guermazi A, Loeser RF (2013) Effects of intensive diet and exercise on knee joint loads, inflammation, and clinical outcomes among overweight and obese adults with knee osteoarthritis: the IDEA randomized clinical trial. JAMA 310:1263–1273CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Montesi L, El Ghoch M, Brodosi L, Calugi S, Marchesini G, Dalle Grave R (2016) Long-term weight loss maintenance for obesity: a multidisciplinary approach. Diabetes Metab Syndr Obes 9:37–46PubMedPubMedCentralGoogle Scholar
  7. 7.
    Veronese N, Stubbs B, Noale M et al (2017) Adherence to a Mediterranean diet is associated with lower prevalence of osteoarthritis: data from the osteoarthritis initiative. Clin Nutr 36:1609–1614Google Scholar
  8. 8.
    Nissensohn M, Roman-Vinas B, Sanchez-Villegas A, Piscopo S, Serra-Majem L (2016) The effect of the Mediterranean diet on hypertension: a systematic review and meta-analysis. J Nutr Educ Behav 48:42–53 e41CrossRefPubMedGoogle Scholar
  9. 9.
    Schwingshackl L, Hoffmann G (2015) Adherence to Mediterranean diet and risk of cancer: an updated systematic review and meta-analysis of observational studies. Cancer Med 4:1933–1947CrossRefPubMedPubMedCentralGoogle Scholar
  10. 10.
    Schwingshackl L, Missbach B, Konig J, Hoffmann G (2015) Adherence to a Mediterranean diet and risk of diabetes: a systematic review and meta-analysis. Public Health Nutr 18:1292–1299CrossRefPubMedGoogle Scholar
  11. 11.
    Kelaiditi E, Jennings A, Steves CJ, Skinner J, Cassidy A, MacGregor AJ, Welch AA (2016) Measurements of skeletal muscle mass and power are positively related to a Mediterranean dietary pattern in women. Osteoporos Int 27:3251–3260CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Cao L, Tan L, Wang H-F, Jiang T, Zhu X-C, Lu H, Tan M-S, Yu J-T (2016) Dietary patterns and risk of dementia: a systematic review and meta-analysis of cohort studies. Mol Neurobiol 53:6144–6154Google Scholar
  13. 13.
    Felson DT, Nevitt MC (2004) Epidemiologic studies for osteoarthritis: New versus conventional study design approaches 30:783–797Google Scholar
  14. 14.
    Block G, Hartman AM, Naughton D (1990) A reduced dietary questionnaire: development and validation. Epidemiology 1:58–64CrossRefPubMedGoogle Scholar
  15. 15.
    Panagiotakos DB, Pitsavos C, Stefanadis C (2006) Dietary patterns: a Mediterranean diet score and its relation to clinical and biological markers of cardiovascular disease risk. Nutr Metab Cardiovasc Dis 16:559–568CrossRefPubMedGoogle Scholar
  16. 16.
    Veronese N, Stubbs B, Noale M, Solmi M, Luchini C, Maggi S (2016) Adherence to the Mediterranean diet is associated with better quality of life: data from the Osteoarthritis Initiative. Am J Clin Nutr 104:1403–1409CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Glaser C, Faber S, Eckstein F, Fischer H, Springer V, Heudorfer L, Stammberger T, Englmeier KH, Reiser M (2001) Optimization and validation of a rapid high-resolution T1-w 3D FLASH water excitation MRI sequence for the quantitative assessment of articular cartilage volume and thickness. Magn Reson Imaging 19:177–185CrossRefPubMedGoogle Scholar
  18. 18.
    Eckstein F, Kunz M, Schutzer M, Hudelmaier M, Jackson RD, Yu J, Eaton CB, Schneider E (2007) Two year longitudinal change and test-retest-precision of knee cartilage morphology in a pilot study for the osteoarthritis initiative. Osteoarthr Cartil 15:1326–1332CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Peterfy CG, Schneider E, Nevitt M (2008) The osteoarthritis initiative: report on the design rationale for the magnetic resonance imaging protocol for the knee. Osteoarthr Cartil 16:1433–1441CrossRefPubMedPubMedCentralGoogle Scholar
  20. 20.
    Washburn RA, McAuley E, Katula J, Mihalko SL, Boileau RA (1999) The physical activity scale for the elderly (PASE): evidence for validity. J Clin Epidemiol 52:643–651CrossRefPubMedGoogle Scholar
  21. 21.
    Katz JN, Chang LC, Sangha O, Fossel AH, Bates DW (1996) Can comorbidity be measured by questionnaire rather than medical record review? Med Care 34:73–84CrossRefPubMedGoogle Scholar
  22. 22.
    Jonckheere AR (1954) A distribution-free k-sample test against ordered alternatives. Biometrika 41:133–145CrossRefGoogle Scholar
  23. 23.
    Miles J (2014) Tolerance and variance inflation factor. In: Balakrishnan N, Colton T, Everitt B, Piegorsch W, Ruggeri F, Teugels JL (eds) Wiley StatsRef: Statistics Reference Online.
  24. 24.
    Hung A, Sayre EC, Guermazi A, Esdaile JM, Kopec JA, Thorne A, Singer J, Wong H, Nicolaou S, Cibere J (2016) Association of body mass index with incidence and progression of knee effusion on magnetic resonance imaging and on knee examination. Arthritis Care Res 68:511–516CrossRefGoogle Scholar
  25. 25.
    Chanchek N, Gersing AS, Schwaiger BJ et al (2018) Association of diabetes mellitus and biochemical knee cartilage composition assessed by T2 relaxation time measurements: data from the osteoarthritis initiative. J Magn Reson Imaging 47:380–390Google Scholar
  26. 26.
    Favero M, Ramonda R, Goldring MB, Goldring SR, Punzi L (2015) Early knee osteoarthritis. RMD Open 1:e000062CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Guermazi A, Niu J, Hayashi D, Roemer FW, Englund M, Neogi T, Aliabadi P, McLennan CE, Felson DT (2012) Prevalence of abnormalities in knees detected by MRI in adults without knee osteoarthritis: population based observational study (Framingham Osteoarthritis Study). BMJ 345:e5339CrossRefPubMedPubMedCentralGoogle Scholar
  28. 28.
    Rainbow R, Ren W, Zeng L (2012) Inflammation and joint tissue interactions in OA: implications for potential therapeutic approaches. Arthritis 2012:741582CrossRefPubMedPubMedCentralGoogle Scholar
  29. 29.
    Hunter DJ, Beavers DP, Eckstein F, Guermazi A, Loeser RF, Nicklas BJ, Mihalko SL, Miller GD, Lyles M, DeVita P, Legault C, Carr JJ, Williamson JD, Messier SP (2015) The Intensive Diet and Exercise for Arthritis (IDEA) trial: 18-month radiographic and MRI outcomes. Osteoarthr Cartil 23:1090–1098CrossRefPubMedGoogle Scholar

Copyright information

© International League of Associations for Rheumatology (ILAR) 2018

Authors and Affiliations

  1. 1.National Research CouncilNeuroscience InstitutePadovaItaly
  2. 2.Ambulatory of Clinical NutritionNational Institute of Gastroenterology-Research Hospital, Saverio de BellisCastellana GrotteItaly
  3. 3.Department of RadiologyUniversity of FoggiaFoggiaItaly
  4. 4.Department of RadiologyScientific Institute “Casa Sollievo della Sofferenza” HospitalFoggiaItaly

Personalised recommendations