Probiotics for Modic Changes in Low Back Pain Patients

Probiotics for Modic Changes: a Randomised Controlled Study to Elucidate the Effect of Probiotics on Vertebral Inflammatory End-plate Changes, So-called Type 1 Modic Changes

Probiotics for spinal inflammatory end-plate changes, so-called Modic changes, a randomised controlled study Modic changes are inflammatory end-plate changes primarily affecting the lower lumbar levels. They are associated with disc degeneration, disc herniation and age, and are only seen on magnetic resonance imaging of the lumbar spine, not on X-ray. Three different types of Modic changes have been described, type 1, 2 and 3, the latter being rare [5]. It has been shown that type 1, but not type 2, is associated with more intense low back pain (LBP) and worse LBP outcome including a lower rate of return to work [7]. In addition, more inflammatory cells have been identified in this type of structural abnormality [25]. Modic changes may affect up to 20% of the general population and more than the double proportion in clinical populations, type 1 and 2 being about equally represented [6]. The investigators have estimated that at least 5,000 of the patients, who are referred to secondary health care in Denmark for LBP per year, will suffer from type 1 Modic changes. Based on a hypothesis of infection as the cause of Modic changes, a randomised clinical trial with one year follow-up has been performed [16]. This trial documented significant effect of long-term (100 days) broad-spectrum antibiotic treatment (amoxicillin/ clavulanic acid) in patients with type 1 Modic changes. However, infection in Modic changes has not been documented convincingly in this or in other studies, and almost all of the treatment effect occurred after the antibiotic treatment was stopped. As yet, the study has not been replicated. The investigators hypothesize that the treatment effect may have been caused by an effect on the gut microbiom possibly explaining the delay of the effect. Probiotics have been shown to influence interleukins in patients with inflammatory conditions and in healthy volunteers [32, 36, 37]. The investigators are therefore performing a randomised trial to study the clinical effect of probiotics on type 1 Modic changes as well as the effect on interleukin levels. The trial is designed as a randomised, clinical, double blind, placebo-controlled trial taking place at the Spine Center, Silkeborg Regional Hospital. Power calculation was based on a smaller treatment effect than the effect found in the study above and resulted in the requirement of including 94 patients. One-year follow-up is planned. The investigators use lactobacillus rhamnosus GG (Dicoflor®) for the trial, as this strain has been used in several clinical studies. The dietary supplement has been shown to influence relevant interleukins significantly, and it has been proven to be stable in action and safe, also in pregnancy. A positive result of such a trial may have significant consequences. The perspective of using probiotics instead of antibiotics in this common disorder is very attractive. At the time being, many type 1 Modic patients in Denmark and other western countries are treated with antibiotics. This is of great concern because of side effects and possible drug resistance development of bacteria. The Technology Transfer Office of Aarhus University is involved.

Background information on trial preparation The bacteria Lactobacillus rhamnosus GG is an anaerobic gram-positive rod that among other areas exists as a part of the body's normal intestinal flora. In the intestine the bacteria breaks down carbohydrates to form lactic acid. The bacteria is tolerant to low pH conditions. Lactobacillus rhamnosus are used in industrial processing of milk products, and medically as a probiotic in order to balance the natural intestinal flora. Lactobacillus rhamnosus is considered to be very safe and has been used during pregnancy in several studies [1, 2]. However, two cases of sepsis have been reported when using this probiotic [1, 2], one case in a patient with diabetes and one case in a heart-surgery patient. Purpose In a placebo-controlled randomized study To examine whether there are effects of prolonged treatment with probiotics in patients with prolonged back pain and type 1 Modic changes in the lumbar spine. To examine whether there can be measured a change in immunological activity in relation to treatment, and whether this change is related to the treatment with probiotics. To examine whether type 1 Modic changes transform differently on magnetic resonance imaging (MRI) during the first year in participants receiving probiotics as compared to participants who did not receive probiotics. Hypotheses This study will be able to show the effect of 100 days of continuous treatment with Lactobacillus rhamnosus GG in participants with back pain of more than 3 months duration and type 1 Modic changes. It can be shown that there is a difference in the activity of the immune system measured by Interleukin 6 (IL-6), interleukin 10 (IL-10) and TNF-alpha dependent on whether participants where treated with placebo or Lactobacillus rhamnosus GG. On MRI, Modic changes in the probiotic group more often will change from type 1 to type 2 as compared to the control group during the first year. Background Modic changes (MC) are inflammatory changes in the end plates of the vertebrae, also known as end plate edema [3]. MC occur at a median of 6% in the general population and a median of 43% in clinical populations in the secondary sector [4]. In patients on sick leave do to lumbar back pain MC is detected in 60% [5]. The changes can be seen on MR imaging, but not on X-ray [3]. MC occur more frequently in patients who have had a herniated disc and rarely occur when there are no degenerative changes in the discs [6, 7]. They usually occur at the level of L4-5 and L5-S1 and more frequently with age like degenerative changes [6]. MC are divided into types 1, 2 and 3, the latter being rare. The classification is based on histology and configuration on MR imaging [3, 7]. Histologically, fissures and granulation in the end plate characterize type 1. In type 2 the destruction of the end plate is more pronounced and there is a fatty degeneration extending into the bone marrow. In type 3 there is a sclerosing of the bone tissue, which is also visible in x-rays, but can also have another cause. On MRI, type 1 changes are low signalling on T1-weighted sequences and high signalling on T2-weighted sequences, whereas type 2 changes are high signalling in both sequences. Type 3 changes appear as low signalling on both T1 and T2-weighted sequences [8]. In addition, there are a so-called mixed type Modic changes with both type 1 and 2 in the same lesion. Finally, Type 1 and Type 2 may occur in different vertebrae in the same patient [9]. Type 2 are usually seen more often, but may also occur less frequent in some patient populations. According to the literature, about 40% of MC type 1 change into type 2, up to 20% may disappear again, and 40-50% remained stable or increase in size during 1-2 years. Type 2 and 3 may remain unchanged, disappear or change to a different type. Major changes remain more stable than small change [9]. It has been shown that type 1, including the mixed type, can result in severe degenerative changes [10]. In cross-sectional studies MC are more closely associated with pain than other degenerative changes in the spine, particularly type 1 [11, 12]. Patients with type 1 MC more often wake up at night because of pain and are often bothered by more than one hour of morning stiffness (inflammatory back pain). They also often experience pain in extension [13]. There have published three studies including patients with type 1 MC clearly showing that type 1 MC is associated with a poor one-year prognosis in regard to disability and pain [7, 11, 16], and Modic changes have been shown to be an independent risk factor for a severe episode of back pain [17]. Other degenerative manifestations hardly seem to adversely affect the prognosis [14]. Finally, the investigators have shown that type 1 MC is associated with increased risk of unsuccessful return to work also when adjusted for other established risk factors [5]. Another cohort study showed no prognostic significance of MC probably due to a low prevalence of MC in that study (14%) [15]. Previous treatment of MC Since MC represent areas of inflammation, it has been recommended that patients with MC should be less physically active and avoid weight load [16]. However, a controlled, prospective study did not show differences in pain and function after a program focusing on rest, compared with the usual active training program for lower back pain, which is why this advice is no longer relevant [17]. There are casuistic reports on temporary effect of treatment with corticosteroids, either as tablets or injection into the disc, but no controlled studies [13]. Some retrospective cohort studies [18, 19] have suggested a better effect of the back surgery in patients with type 1 compared with other patients, but other studies have suggested a lower rate of healing [9, 23]. There are no randomized prospective studies on this subject. Infection hypothesis There is no evidence of bacteria in MC [20], but there is evidence of an increased number of TNF-immunoreactive cells in type 1 changes as compared to type 2 changes [25]. Some studies have identified bacteriae in disc tissue in up to 50% of patients underwenting surgery for herniated disc, most frequently Proprionibacterium acnes, which is usually a benign skin bacteria [21]. These findings have led to the hypotheses of a low-grade infection as a possible cause of MC, since disc herniation has been shown to be a risk factor for subsequent occurrence of MC. An uncontrolled study suggested effect of Bioclavid treatment [22], a commonly used drug amoxicillin and clavulanic acid, which is commonly used for exacerbation of chronic obstructive pulmonary disease. January 2013, the results of a double-blind, randomized clinical trial with 1 year follow-up was published comprising of 160 patients with type 1 MC showing efficacy of treatment with Bioclavid, 1 tablet twice daily for 100 days. There were only moderate differences in pain and function during the treatment period with antibiotics, but increasing difference between intervention and control group during the follow-up period the following 9 months [16]. The treatment duration was chosen pragmatically from current practice in patients with spondylodiscitis. However, Modic patients do not have fever or elevated sedimentation rate. Included were only patients with verified herniated disc ½ - 2 years previously, patients with pain of more than 6 months duration and back pain intensity the previous 2 weeks ≥ 6 of 10. More than half of the patients had gastrointestinal side effects, but there were no serious adverse effects. The very modest impact of antibiotics during the first 100 days, but subsequently pronounced effect during the follow-up period, gave the present investigators reason to assume that the treatment effect was not caused by the antibiotic treatment itself, but rather by the consequent changes of the intestinal flora. Usually, the effect antibiotic treatment in infections occur more more rapidly. Activation of the immune system IL-6 is a marker of inflammation, and plasma IL-6 can be increased both by infections and by autoimmune activation [23]. Recent studies suggest that the IL-6 by disc degeneration plays a role in the interaction with other cytokines, particularly IL-1 and TNF-alpha [24], and in particular for the activation of macrophages [25]. A cross-sectional study in patients with low back pain has shown an association between IL-6 concentration, pain and poor sleep quality [26]. By measuring changes in IL-6 levels in relation to treatment with probiotics or placebo, the investigators will be able to see whether probiotics can reduce the overall IL-6 and TNF-alpha production in the body and increase the production of IL-10. The investigators suspect that IL-6 and TNF-alpha will decrease after the start of probiotic treatment, and that it will increase again after ending treatment with probiotics, in contrast to IL-10. Conclusion The investigators consider it important to clarify whether probiotic therapy has clinically relevant effect on disability and pain in patients with type 1 Modic changes, and whether this will result in detectable changes in inflammation markers and modifications to Modic changes over time. A positive result may open up for an alternative effective, low-risk and low-cost treatment. Selection of the probiotic The probiotic product Dicoflor ® (Pharma Force Aps) is used, as it contains Lactobacillus rhamnosus GG, the most commonly used bacterial strain in clinical trials. Each capsule contains 6 billion lactic acid bacteriae. The recommended dose is 1 capsule twice daily. The effect of this dietary supplement has been studied in patients with rheumatoid arthritis [32] and demonstrated as an effect on TNF-α and IL-8 [27-29]. The bacterial strain has been used for children with atopic dermatitis, with a demonstrated effect on IL-10 [30]. Furthermore, the effect on the immune system has been studied in healthy adults [31]. Methods Design Randomized double-blind clinically controlled study with 3, 6 and 12 month follow-up and 2 arms: Group 1: Probiotic Dicoflor ® twice daily for 100 days Group 2: Placebo twice daily for 100 days Randomization Web-based automated bloc randomization was established by the Data Manager Jakob Hjort and created in the database RedCap, belonging to the Department of Clinical Medicine, Aarhus University Department. Only one person, project assistant PhD Nanna Rolving, Diagnostic Center, Regional Hospital Silkeborg, knows the randomization code. Location Spine Center (The Research Unit), Diagnostic Center, Regional Hospital Silkeborg. Ole Kudsk Jensen, MD PhD is sponsor-investigator. A junior doctor (MHA) is in charge of the practical implementation of the project, which is planned as part of a PhD. Setting Study participants are recruited among the patients referred to the Spine Center, Silkeborg Regional Hospital, comprising the Surgical Unit (K-RYG), the Medical Unit (M-RYG) and the Research Unit (F-RYG). After having received the usual intervention in the Spine Center (surgical intervention not indicated), the patients are informed about the study and given written information with a telephone number allowing them to contact the Research Unit, if they are interested in participating in the study. Most participants are recruited from the Medical Unit taking care of the vast majority of the patients of the Spine Center. Usually, an MRI of the lumbar spine is available when a referral from a general practitioner is received. If MRI is not available, it usually will be ordered by a physician at the Spine Center to clarify the condition in relating to clinical findings. The usual interventions of M-RYG and F-RYG are very similar. The patients complete a questionnaire. A physician, chiropractor or physiotherapist, who subsequently conduct a clinical back examination, take a medical history. The clinical issues and MR images are discussed between physician, chiropractor and physiotherapist. If there is a need for further tests or imaging, this is ordered. In M-RYG and F-RYG patients receive general information about back pain, including both the general poor correlation between imaging and pain, the importance of exercising, general exercise or training, and finally that psychological factors can have an impact on back pain. Considerations are made whether medical pain treatment needs adjustment, and whether assessment by a surgeon should be considered. The MRI is assessed by the examining physician and discussed with an affiliated physician from the Department of Radiology at the Regional Hospital in Silkeborg [32]. When the patient has received the usual intervention in M-RYG, F-RYG or in K-RYG, and Modic type 1 or mixed Modic changes have been identified on MRI of the lumbar spine, the patient is told that this structural change may be a contributing factor to pain and may cause the pain and disability to be prolonged. The patient is then informed about the present Modic project and offered an opportunity to participate. If the patient expresses interest in participating they are given written information on the study. The information includes the phone number of the secretary at F-Ryg. The patient is advised to contact if the secretary after reading the information if still interested. Baseline When the patient phones the secretary of F-RYG, a visit at F-RYG is planned to finally decide whether in- and exclusion criteriae are fulfilled. When meeting at F-RYG, the participant-information is discussed. If the patient is still interested and there are no exclusion criteriae, a medical record and a clinical back examination are performed. Subsequently, a blood sample is taken. Afterwards, the patient is examined and guided by the physiotherapist regarding plan for physical activity. If lab tests are normal and there are no exclusion criteria, and all inclusion criteria are met, a consent form is signed, and the participant is included in the study. The physician ensures that the participant is given proper information in regard to back pain in general and specifically in relation to the patient's situation. The medical pain management is adjusted if necessary. Level of function is recorded (self-assessed walking distance, maximum time sitting down, nocturnal pain). The mobility of the spine is measured (modified Schober, side bending, bending backwards), and it is registered whether there are signs of nerve root compression. The patient is informed about the possibility of an assessment by a surgeon during the study period , if there is any worsening or no improvement of the condition. Indications for referral to a surgeon adheres to guidelines. If the patient is on sick leave, the person is advised to resume work when possible. No case manager is assigned. The physiotherapist registers self-rated loss of functioning in percent, and evaluates the pain response during testing in order to ensure that the patient is appropriately informed about specific exercises, exercise in general and "pacing" principles. A follow-up appointment with the physiotherapist is arranged after 14 days. Clinical data are recorded in schedules. Dispensing of capsules with dietary supplements / placebo After randomization, the capsules are handed out to the participant in blister packs, and compliance is recorded in a diary. The participants are advised to take the capsules at 08 and 18 at meals, and are recommended to save emptied blister packs. The capsules are recommended to be stored in a refrigerator. They can also be stored at room temperature. The capsules are designed so that placebo capsules are indistinguishable from active capsules. The participant is advised against the use of other preparations containing other lactic acid bacteria, for example Paraghurt or Idoform. Baseline data in addition to clinical findings Questionnaire data: Before clinical examination by the doctor and physiotherapist a comprehensive questionnaire is completed including questions of back pain and leg pain during the preceding 14 days [33], limitations of function at the examination day [34], possibly previous surgery for herniated disc, fear avoidance, work, possibly sick leave and duration, questions about perceived cause of the pain, questions about exercise during leisure time, type of housing, psychological symptoms, tobacco and alcohol consumption and the use of pain medication. How these questions are answered, have previously been shown to be important for prognosis [35, 36]. Follow-up Generally: If symptoms of nerve root irritation occur during the study, the participant may contact the Research Unit for an additional clinical evaluation. After 14 days: The physiotherapist makes adjustments to the exercise program. In addition, blood samples are taken. After 102 Days: The participant meets with project nurse and returns the emptied blister packs, complete a questionnaire (pain, function, possibly other treatment), and blood samples are taken. 6 months after initiation of the treatment the project nurse records the level of pain and function by phone or mail. 12 months after initiation of the treatment the participant attends for a new consultation including a questionnaire, blood tests and a new standardized clinical assessment. A new MRI scan is ordered of the lumbar spine to assess the extent to which the previously diagnosed Modic type 1 change is unchanged or has changed. After the project is ended, the MRI changes will be described standardized [38] at baseline and 1 year without knowledge of clinical information and intervention group (blinded description). Status with regards to sick leave is determined based on the DREAM database [37]. Blood tests: Blood samples are taken before starting treatment, partly as routine tests with regards to inclusion, partly as project samples, which are then frozen. Routine tests include: CRP, hemoglobin, white blood cells, platelets, creatinine, ALAT, alkaline phosphatase, albumin. Blood samples for the biobank: IL-6, IL-1 0 and TNF-alpha and an extra sample for future reference and possible additional analyzes. Similarly, additional blood samples are taken after 14 days, 102 days and 1 year. In total 4 x 7 ml of blood is taken for routine tests and 4 x 6 ml for project samples. Project samples are stored in the biobank and destroyed at the end of the full study including the research period. Registration of possible adverse effects Unwanted side effects or possible side effects are recorded as an adverse event in RedCap and reported to the Ethics Committee. There is no available product summary, since the product is registered as dietary supplement. There are no known side effects when using the product in people with a normally functioning immune system. Therefore, any potential side effect will be an unexpected side effect. Infections during the study are assessed and treated by the general practiioner as usual. If the participant is admitted at hospital for infection, the Research Unit should be contacted with regard to evaluating possibly precautions. If there is an infection and bacteraemia with intestinal bacteria or lactic acid bacteria, the medication will be discontinued and the randomization code broken. The case will be reported to the Ethics Committee. Possible causality should be assessed before unblinding. Side effects and possible discontinuation of the preparation is recorded in the journal and in RedCap, which is used as 'Case Report Form'. Data analysis Data are made anonymous, so that each patient can only be identified by a serial number. Calculation of power Power calculation is based on the findings of the placebo-controlled study [16] and on the assumption that treatment with probiotics will be less effective than treatment with antibiotics. Changing of the median values of function in the intervention group and the control group at 1 year were 8 and 1, respectively, in the previously mentioned study. It was assumed, that the mean values were not different from the median values. The standard deviations were not specified in the article [16], only the interquartiles that were comparable with data from a previous study made by the investigators [5]. The standard deviation was 6.3 in that study. Using a standard deviation of 6.5, a clinically relevant change was estimated to be 5 instead of 8. Strength calculation was made by STATA [38] (β = 0.8, α = 0.05): Requiring 42 patients in each group, to be able to detect a difference of '5' in functioning after 1 year compared to '1' in the control group. Assuming a drop-out rate at 10% will result in the need for a group size of 47. Statistical analysis Primary outcome: 1) The difference between 1-year function and baseline is analyzed by unpaired t-test, if the data are normally distributed, and with Wilcoxon's test for unpaired data, if the data are not normally distributed. Secondary outcome: Proportion with minimal loss of function is compared with Chi 2 test. And 3) are analyzed by unpaired t-test or Wilcoxon's test for unpaired data, depending on whether the data are normally distributed. 4) To be analyzed with Wilcoxon's test for unpaired data. 5) Compared by Chi 2 test. 6) Difference in number of operated is tested with Fishers exact test, the figures are assumed to be small. 7) Most significant change is the decrease or increase in IL-6 levels respectively after starting treatment and after treatment discontinuation, which is tested with paired t-test within the groups and unpaired t-test between groups, possibly with a Wilcoxon's test, if the data are not normally distributed. 8) And 9, 10) Compared by Chi 2 test. An 'Intention to treat' principle will be used, which means all randomized patients are included in the analysis, independent of their compliance with treatment or if operation was necessary. In the case of lacking follow-up data, baseline data will be entered (last observation carried forward). A significance level of 5% will be used. Ethical considerations It is considered consistent with good ethical standard to offer participation in a randomized trial, which deals with a preparation that is usually well tolerated and may have efficacy, and as an alternative to prolonged antibiotic therapy, which is used in some places and is not without risk. There is good evidence that patients with type 1 Modic changes are at higher risk than other back pain patients for persistent pain and disability and even problems with return to work. The latter is of special importance, not only for the quality of life but also economically, both for the individual patient and for society. It can be estimated that about 20% of patients referred to secondary health care have type 1 Modic changes, and it is estimated that at least 20,000 patients are referred to the hospitals in Denmark, due to low back pain. This may imply that a therapy with clinically relevant effect may be offered to about 4,000 patients annually in Denmark. Review and Approvals Application for approval by the Research Ethics Committee, Central Jutland, and the experiment reported to www.ClinicalTrials.gov and the Data Protection Agency. The Department of Health has announced that the trial should not be run as drug trials, and therefore there is no need for supervision from GCP unit. Time schedule Inclusion of the participants has started January 2016, and by now (end of March 2017) 91 participants have been included. One-year follow-up of all participants will end at April 2018. Data analysis and article writing are planned in 2018-19.

Inclusion criteriae Referral to one of the two spine centres in Central Denmark Region (Silkeborg and Aarhus) Sufficient competency in Danish to fill out questionnaires Age between 18 and 65 years, both included MRI verified type 1 Modic changes within the last 3 months, either exclusively, mixed or in combination with type 2 or type 3 changes No sign of activation of the immune system at inclusion (normal CRP) Back pain must dominate leg pain Pain duration >3 months and relevant disability (RM>5) Exclusion criteriae On beforehand, plans of or a wish for spinal surgery Spinal surgery for disc herniation or spinal stenosis within the last 6 months Previous spinal fusion surgery Plans of treatment with antibiotics for Modic changes elsewhere Antibiotic treatment for Modic changes within the preceding ½ year More than two weeks of antibiotic treatment within the past three months Immunosuppressive treatment at present Lacking ability to enter or complete the project due to either mental, social or geographical reasons Known intestinal pathology, autoimmune disease, immune deficiency, malabsorption, cancer or chronic infection

Zein EF, Karaa S, Chemaly A, Saidi I, Daou-Chahine W, Rohban R. [Lactobacillus rhamnosus septicemia in a diabetic patient associated with probiotic use: a case report]. Ann Biol Clin (Paris). 2008 Mar-Apr;66(2):195-8. doi: 10.1684/abc.2008.0210. French.

Kochan P, Chmielarczyk A, Szymaniak L, Brykczynski M, Galant K, Zych A, Pakosz K, Giedrys-Kalemba S, Lenouvel E, Heczko PB. Lactobacillus rhamnosus administration causes sepsis in a cardiosurgical patient--is the time right to revise probiotic safety guidelines? Clin Microbiol Infect. 2011 Oct;17(10):1589-92. doi: 10.1111/j.1469-0691.2011.03614.x. Epub 2011 Aug 16.

Modic MT, Steinberg PM, Ross JS, Masaryk TJ, Carter JR. Degenerative disk disease: assessment of changes in vertebral body marrow with MR imaging. Radiology. 1988 Jan;166(1 Pt 1):193-9. doi: 10.1148/radiology.166.1.3336678.

Jensen TS, Karppinen J, Sorensen JS, Niinimaki J, Leboeuf-Yde C. Vertebral endplate signal changes (Modic change): a systematic literature review of prevalence and association with non-specific low back pain. Eur Spine J. 2008 Nov;17(11):1407-22. doi: 10.1007/s00586-008-0770-2. Epub 2008 Sep 12.

Jensen OK, Nielsen CV, Sorensen JS, Stengaard-Pedersen K. Type 1 Modic changes was a significant risk factor for 1-year outcome in sick-listed low back pain patients: a nested cohort study using magnetic resonance imaging of the lumbar spine. Spine J. 2014 Nov 1;14(11):2568-81. doi: 10.1016/j.spinee.2014.02.018. Epub 2014 Feb 14.

Albert HB, Manniche C. Modic changes following lumbar disc herniation. Eur Spine J. 2007 Jul;16(7):977-82. doi: 10.1007/s00586-007-0336-8. Epub 2007 Mar 3.

Rahme R, Moussa R. The modic vertebral endplate and marrow changes: pathologic significance and relation to low back pain and segmental instability of the lumbar spine. AJNR Am J Neuroradiol. 2008 May;29(5):838-42. doi: 10.3174/ajnr.A0925. Epub 2008 Feb 13.

Albert HB, Briggs AM, Kent P, Byrhagen A, Hansen C, Kjaergaard K. The prevalence of MRI-defined spinal pathoanatomies and their association with modic changes in individuals seeking care for low back pain. Eur Spine J. 2011 Aug;20(8):1355-62. doi: 10.1007/s00586-011-1794-6. Epub 2011 May 5.

Jensen RK, Leboeuf-Yde C, Wedderkopp N, Sorensen JS, Jensen TS, Manniche C. Is the development of Modic changes associated with clinical symptoms? A 14-month cohort study with MRI. Eur Spine J. 2012 Nov;21(11):2271-9. doi: 10.1007/s00586-012-2309-9. Epub 2012 Apr 24.

Kerttula L, Luoma K, Vehmas T, Gronblad M, Kaapa E. Modic type I change may predict rapid progressive, deforming disc degeneration: a prospective 1-year follow-up study. Eur Spine J. 2012 Jun;21(6):1135-42. doi: 10.1007/s00586-012-2147-9. Epub 2012 Jan 17.

Kjaer P, Korsholm L, Bendix T, Sorensen JS, Leboeuf-Yde C. Modic changes and their associations with clinical findings. Eur Spine J. 2006 Sep;15(9):1312-9. doi: 10.1007/s00586-006-0185-x. Epub 2006 Aug 9.

Kjaer P, Leboeuf-Yde C, Korsholm L, Sorensen JS, Bendix T. Magnetic resonance imaging and low back pain in adults: a diagnostic imaging study of 40-year-old men and women. Spine (Phila Pa 1976). 2005 May 15;30(10):1173-80. doi: 10.1097/01.brs.0000162396.97739.76.

Bailly F, Maigne JY, Genevay S, Marty M, Gandjbakhch F, Rozenberg S, Foltz V. Inflammatory pain pattern and pain with lumbar extension associated with Modic 1 changes on MRI: a prospective case-control study of 120 patients. Eur Spine J. 2014 Mar;23(3):493-7. doi: 10.1007/s00586-013-3036-6. Epub 2013 Sep 25.

Steffens D, Hancock MJ, Maher CG, Williams C, Jensen TS, Latimer J. Does magnetic resonance imaging predict future low back pain? A systematic review. Eur J Pain. 2014 Jul;18(6):755-65. doi: 10.1002/j.1532-2149.2013.00427.x. Epub 2013 Nov 26.

Keller A, Boyle E, Skog TA, Cassidy JD, Bautz-Holter E. Are Modic changes prognostic for recovery in a cohort of patients with non-specific low back pain? Eur Spine J. 2012 Mar;21(3):418-24. doi: 10.1007/s00586-011-1964-6. Epub 2011 Aug 12.

Albert HB, Sorensen JS, Christensen BS, Manniche C. Antibiotic treatment in patients with chronic low back pain and vertebral bone edema (Modic type 1 changes): a double-blind randomized clinical controlled trial of efficacy. Eur Spine J. 2013 Apr;22(4):697-707. doi: 10.1007/s00586-013-2675-y. Epub 2013 Feb 13.

Jensen RK, Leboeuf-Yde C, Wedderkopp N, Sorensen JS, Manniche C. Rest versus exercise as treatment for patients with low back pain and Modic changes. A randomized controlled clinical trial. BMC Med. 2012 Feb 29;10:22. doi: 10.1186/1741-7015-10-22.

Blondel B, Tropiano P, Gaudart J, Huang RC, Marnay T. Clinical results of lumbar total disc arthroplasty in accordance with Modic signs, with a 2-year-minimum follow-up. Spine (Phila Pa 1976). 2011 Dec 15;36(26):2309-15. doi: 10.1097/BRS.0b013e31820f7372.

Esposito P, Pinheiro-Franco JL, Froelich S, Maitrot D. Predictive value of MRI vertebral end-plate signal changes (Modic) on outcome of surgically treated degenerative disc disease. Results of a cohort study including 60 patients. Neurochirurgie. 2006 Sep;52(4):315-22. doi: 10.1016/s0028-3770(06)71225-5.

Wedderkopp N, Thomsen K, Manniche C, Kolmos HJ, Secher Jensen T, Leboeuf Yde C. No evidence for presence of bacteria in modic type I changes. Acta Radiol. 2009 Jan;50(1):65-70. doi: 10.1080/02841850802524485.

Albert HB, Lambert P, Rollason J, Sorensen JS, Worthington T, Pedersen MB, Norgaard HS, Vernallis A, Busch F, Manniche C, Elliott T. Does nuclear tissue infected with bacteria following disc herniations lead to Modic changes in the adjacent vertebrae? Eur Spine J. 2013 Apr;22(4):690-6. doi: 10.1007/s00586-013-2674-z. Epub 2013 Feb 10.

Albert HB, Manniche C, Sorensen JS, Deleuran BW. Antibiotic treatment in patients with low-back pain associated with Modic changes Type 1 (bone oedema): a pilot study. Br J Sports Med. 2008 Dec;42(12):969-73. doi: 10.1136/bjsm.2008.050369. Epub 2008 Aug 21.

Munoz-Canoves P, Scheele C, Pedersen BK, Serrano AL. Interleukin-6 myokine signaling in skeletal muscle: a double-edged sword? FEBS J. 2013 Sep;280(17):4131-48. doi: 10.1111/febs.12338. Epub 2013 Jun 18.

Studer RK, Vo N, Sowa G, Ondeck C, Kang J. Human nucleus pulposus cells react to IL-6: independent actions and amplification of response to IL-1 and TNF-alpha. Spine (Phila Pa 1976). 2011 Apr 15;36(8):593-9. doi: 10.1097/BRS.0b013e3181da38d5.

Hamamoto H, Miyamoto H, Doita M, Takada T, Nishida K, Kurosaka M. Capability of nondegenerated and degenerated discs in producing inflammatory agents with or without macrophage interaction. Spine (Phila Pa 1976). 2012 Feb 1;37(3):161-7. doi: 10.1097/BRS.0b013e31821a874b.

Heffner KL, France CR, Trost Z, Ng HM, Pigeon WR. Chronic low back pain, sleep disturbance, and interleukin-6. Clin J Pain. 2011 Jan;27(1):35-41. doi: 10.1097/ajp.0b013e3181eef761.

Zhang L, Li N, Caicedo R, Neu J. Alive and dead Lactobacillus rhamnosus GG decrease tumor necrosis factor-alpha-induced interleukin-8 production in Caco-2 cells. J Nutr. 2005 Jul;135(7):1752-6. doi: 10.1093/jn/135.7.1752.

Lopez M, Li N, Kataria J, Russell M, Neu J. Live and ultraviolet-inactivated Lactobacillus rhamnosus GG decrease flagellin-induced interleukin-8 production in Caco-2 cells. J Nutr. 2008 Nov;138(11):2264-8. doi: 10.3945/jn.108.093658.

Donato KA, Gareau MG, Wang YJJ, Sherman PM. Lactobacillus rhamnosus GG attenuates interferon-gamma and tumour necrosis factor-alpha-induced barrier dysfunction and pro-inflammatory signalling. Microbiology (Reading). 2010 Nov;156(Pt 11):3288-3297. doi: 10.1099/mic.0.040139-0. Epub 2010 Jul 23.

Pessi T, Sutas Y, Hurme M, Isolauri E. Interleukin-10 generation in atopic children following oral Lactobacillus rhamnosus GG. Clin Exp Allergy. 2000 Dec;30(12):1804-8. doi: 10.1046/j.1365-2222.2000.00948.x.

Schultz M, Linde HJ, Lehn N, Zimmermann K, Grossmann J, Falk W, Scholmerich J. Immunomodulatory consequences of oral administration of Lactobacillus rhamnosus strain GG in healthy volunteers. J Dairy Res. 2003 May;70(2):165-73. doi: 10.1017/s0022029903006034.

Solgaard Sorensen J, Kjaer P, Jensen ST, Andersen P. Low-field magnetic resonance imaging of the lumbar spine: reliability of qualitative evaluation of disc and muscle parameters. Acta Radiol. 2006 Nov;47(9):947-53. doi: 10.1080/02841850600965062.

Manniche C, Asmussen K, Lauritsen B, Vinterberg H, Kreiner S, Jordan A. Low Back Pain Rating scale: validation of a tool for assessment of low back pain. Pain. 1994 Jun;57(3):317-326. doi: 10.1016/0304-3959(94)90007-8.

Albert HB, Jensen AM, Dahl D, Rasmussen MN. [Criteria validation of the Roland Morris questionnaire. A Danish translation of the international scale for the assessment of functional level in patients with low back pain and sciatica]. Ugeskr Laeger. 2003 Apr 28;165(18):1875-80. Danish.

Jensen OK, Nielsen CV, Stengaard-Pedersen K. One-year prognosis in sick-listed low back pain patients with and without radiculopathy. Prognostic factors influencing pain and disability. Spine J. 2010 Aug;10(8):659-75. doi: 10.1016/j.spinee.2010.03.026. Epub 2010 May 5.

Jensen OK, Stengaard-Pedersen K, Jensen C, Nielsen CV. Prediction model for unsuccessful return to work after hospital-based intervention in low back pain patients. BMC Musculoskelet Disord. 2013 Apr 19;14:140. doi: 10.1186/1471-2474-14-140.

Hjollund NH, Larsen FB, Andersen JH. Register-based follow-up of social benefits and other transfer payments: accuracy and degree of completeness in a Danish interdepartmental administrative database compared with a population-based survey. Scand J Public Health. 2007;35(5):497-502. doi: 10.1080/14034940701271882.