- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
Intraarticular corticosteroid (IAC) injections are widely used in the management of juvenile idiopathic arthritis (JIA) to induce rapid relief of symptoms of active synovitis, particularly pain and functional impairment, and to obviate the need for regular systemic therapy ([1-3]). This mode of therapy is generally considered for the treatment of children with arthritis in a small number of joints, particularly large joints, or in children with a polyarticular course who are responding to systemic therapy but have a few joints that remain actively inflamed. Currently, it is thought that many pediatric rheumatologists are using IACs as their first approach in oligoarticular JIA ([4-9]). In children with this subtype, local injection therapy may facilitate correction of joint contractures, prevention of leg length discrepancy, resolution of Baker's cysts, improvement of tenosynovitis, and discontinuation of oral medications ([10, 11]).
In recent years, the strategy of performing IAC injections in multiple joints has been advocated in children with polyarticular JIA with the aim of inducing prompt remission of synovitis, while simultaneously initiating therapy with disease-modifying antirheumatic drugs (DMARDs) and/or biologic agents ([12, 13]). This approach is regarded as an alternative to systemic corticosteroids to pursue the so-called “bridge” effect, i.e., to achieve a quick control of inflammatory symptoms while awaiting the full therapeutic effect of a DMARD or biologic medication. Multiple IAC injections have the potential advantage of avoiding many side effects of systemic corticosteroids and of selectively targeting the inflamed joints. A recent survey among pediatric rheumatologists in the US and Canada has shown that 15% had performed more than 10 IAC injections in a single pediatric patient at one time ().
However, the therapeutic effectiveness of multiple IAC injections has seldom been investigated in JIA. Furthermore, to date there have been no controlled trials on IACs in childhood arthritis. As a result, the place of this treatment modality in disease management remains unclear and controversial. Notably, although the use of IACs has been incorporated in the recent American College of Rheumatology recommendations for the treatment of JIA (), the specific indications of multiple injections have not been addressed.
In the current biologic era, multiple IAC injections may be regarded as a somewhat old-fashioned therapeutic procedure. However, we previously found that a sizable proportion of 118 children who underwent multiple IAC injections, with or without concomitant systemic medications, experienced sustained remission of synovitis in injected joints (). Owing to its lower potential for corticosteroid-related toxicity, multiple IAC injection therapy may represent a suitable alternative to systemic corticosteroids in aggressive therapeutic protocols aimed at disease remission in severe polyarthritis, particularly in clinical settings where biologic agents are not available or affordable.
To gain insights into the potential utility of multiple IAC injections in the management of JIA, there is a need to obtain information about their potential to induce long-lasting remission of synovitis. Furthermore, it would be desirable to identify factors associated with the risk of flare of joint inflammation, in order to optimize the current therapeutic approaches. To address these issues, we investigated the effectiveness and sought outcome predictors of multiple IAC injections in 220 children with JIA who underwent this therapeutic procedure in our center over a 10-year period.
Box 1. Significance & Innovations
- This study describes the outcome and predicting factors of multiple intraarticular corticosteroid injections in children with juvenile idiopathic arthritis.
- Intraarticular corticosteroid therapy proved able to induce sustained remission of joint synovitis in a substantial percentage of patients, with a good safety profile.
- The information provided by this study helps define the place of intraarticular corticosteroid therapy in the current management of juvenile idiopathic arthritis.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
A total of 220 children with JIA who received their first multiple IAC injection were included in the study. These patients were part of a whole series of ∼900 JIA patients seen at the study center in the study period. The 220 patients had a total of 1,096 joints injected. The median and maximum number of injected joints was 4 (IQR 3–6) and 24, respectively. One hundred eighteen of the study patients were included in a previous analysis of IAC therapy in JIA ().
Table 1 shows the main demographic and clinical features of the study patients at the time of IAC injection. Overall, the patient series had the typical characteristics of the Italian JIA population, which is overly represented by female patients with early disease onset, asymmetric arthritis, and positive ANA status (). The majority of patients (71.3%) had the ILAR categories of rheumatoid factor–negative polyarthritis or extended oligoarthritis; 6.8% of patients had systemic arthritis. The articular course in the 205 patients with nonsystemic arthritis was polyarticular in 169 patients (82.4%) and oligoarticular in 36 patients (17.6%). On average, the patients had a short disease duration (median 0.6 years) and were a young age (median 3.7 years). One hundred eighteen patients (53.6%) were age <4 years at the time of the procedures. After the IAC injection, 61.8% of patients received systemic medications, whereas 38.2% did not. The most frequently administered medication was methotrexate (56.8%), whereas only 9.5% of the patients were given biologic agents. Of the 21 patients who received biologic agents, 13 were given etanercept, 4 were given anakinra, 2 were given adalimumab, 1 was given infliximab, and 1 was given abatacept. Systemic corticosteroids were taken by 25 patients (11.4%), most of whom had systemic arthritis.
Table 1. Demographic and clinical features of 220 patients with juvenile idiopathic arthritis at the time of IAC injection*
| ||All patients (n = 220)||Patients with remission (n = 74)||Patients with flare (n = 146)||Pa|
|Female sex||175 (79.5)||60 (81.1)||115 (78.8)||0.69|
|Age at disease onset, median (IQR) years||2.4 (1.7–4.9)||2.6 (1.8–4.9)||2.3 (1.6–5.1)||0.65b|
|ILAR category|| || || || |
|Systemic arthritis||15 (6.8)||4 (5.4)||11 (7.5)|| |
|RF-negative polyarthritis||82 (37.2)||29 (39.2)||53 (36.3)|| |
|RF-positive polyarthritis||4 (1.8)||0 (0.0)||4 (2.7)|| |
|Persistent oligoarthritis||34 (15.5)||20 (27.0)||14 (9.6)|| |
|Extended oligoarthritis||75 (34.1)||20 (27.0)||55 (37.7)|| |
|Psoriatic arthritis||1 (0.5)||0 (0.0)||1 (0.7)|| |
|Enthesitis-related arthritis||4 (1.8)||1 (1.4)||3 (2.1)|| |
|Undifferentiated arthritis||5 (2.3)||0 (0.0)||5 (3.4)|| |
|Patients with positive ANAs, no./total (%)||175/205 (85.4)||66/71 (93.0)||109/134 (81.3)||0.025|
|Disease coursec|| || || ||0.009|
|Oligoarthritis||36 (16.4)||20 (27.0)||16 (11.0)|| |
|Polyarthritis||169 (76.8)||50 (67.6)||119 (81.5)|| |
|Systemic arthritis||15 (6.8)||4 (5.4)||11 (7.5)|| |
|Number of joints injected|| || || ||0.90|
|3 or 4||135 (61.4)||45 (60.8)||90 (61.6)|| |
|≥5||85 (38.6)||29 (39.2)||56 (38.4)|| |
|Disease duration, median (IQR) years||0.6 (0.2–2.5)||0.6 (0.2–1.2)||0.7 (0.2–3.1)||0.54b|
|Age, median (IQR) years||3.7 (2.3–9.1)||3.6 (2.6–7.5)||4.0 (2.2–9.6)||0.58b|
|Patients who underwent general anesthesia, no./total (%)||169/212 (79.7)||65/73 (89.0)||104/139 (74.8)||0.014|
|ESR (n = 201), median (IQR) mm/hour||40 (23–58)||38 (19–57)||41 (24–58.3)||0.27b|
|CRP level (n = 203), median (IQR) mg/dl||1.3 (0.46–3.12)||0.81 (0.45–2.13)||1.58 (0.59–3.24)||0.033b|
|ESR >20 mm/hour, no./total (%)||157/201 (78.1)||47/65 (72.3)||110/136 (80.9)||0.17|
|CRP level >0.46 mg/dl, no./total (%)||151/203 (74.4)||44/69 (63.8)||107/134 (79.9)||0.013|
|Therapy at the time of IAC injection||136 (61.8)||56 (75.7)||80 (54.8)||0.003|
|Ongoing methotrexate||49 (22.3)||20 (27.0)||29 (19.9)||0.23|
|Newly started methotrexate||76 (34.5)||32 (43.2)||44 (30.1)||0.053|
|Ongoing or newly started methotrexate||125 (56.8)||52 (70.3)||73 (50.0)||0.004|
|Ongoing or newly started biologic agents||21 (9.5)||6 (8.1)||15 (10.3)||0.61|
|Ongoing or newly started systemic corticosteroids||25 (11.4)||6 (8.1)||19 (13.0)||0.28|
Following the IAC injection, 146 patients (66.4%) experienced a flare of synovitis after a median of 0.5 years (IQR 0.3–1.3 years), whereas 74 patients (33.6%) had sustained remission of synovitis in all injected joints after a median of 0.9 years (IQR 0.6–1.9 years). The comparison of the demographic and clinical features between patients with remission and patients with flare of synovitis is shown in Table 1. As compared with patients with remission, patients who experience a flare had a lower frequency of positive ANA status and a greater frequency of CRP level elevation, had undergone general anesthesia less frequently, had a relatively greater frequency of a polyarticular or systemic disease course, and had received concomitant treatment with methotrexate less frequently. The proportion of patients who received methotrexate orally or subcutaneously was comparable in the 2 groups (data not shown). Of the 146 patients who experienced a flare of synovitis, 64 (43.8%) underwent a repeat IAC injection (which was administered as monotherapy in 37 patients and in association with a change in systemic therapy in 27 patients), whereas 82 patients (56.2%) had only a change in systemic treatment.
The type and number of injected joints in the entire patient sample and by outcome of the IAC procedure are shown in Table 2. The knee was the most frequently injected joint, followed by the ankle, subtalar, elbow, wrist, and small hand and foot joints. The shoulder, hip, and temporomandibular joints were all injected in <5% of patients. Outcome assessment showed that the proportion of injected joints with sustained remission of synovitis was greater than the proportion of injected joints with synovitis flare for the knee, the subtalar joint, the elbow, and the small hand and foot joints. The only joints for which the frequency of remission and flare was comparable were the ankle and the wrist. The total number of injected joints was 1,096. Of the 1,079 individual joints for which the outcome could be assessed, 309 (28.6%) had a flare of synovitis, whereas 770 (71.4%) had sustained remission. Among the 146 patients with synovitis flare, the number of flared joints (n = 309) was much lower than the number of injected joints (n = 725). The frequency of synovitis flare was comparable among patients who were injected in ≤4 (n = 135) or >4 (n = 85) joints (66.7% and 65.9%, respectively).
Table 2. Type and number of injected joints*
| ||Patients who had the joint injected (n = 220)||No. of injected joints||Joints with remissiona||Joints with synovitis flarea||P|
|Knee||186 (84.5)||298||236 (79.2)||62 (20.8)||< 0.0001|
|Ankle||168 (76.4)||245||132/241 (54.8)||109/241 (45.2)||0.14|
|Subtalar||67 (30.5)||88||57/87 (65.5)||30/87 (34.5)||0.004|
|Elbow||64 (29.1)||84||71 (84.5)||13 (15.5)||< 0.0001|
|Proximal interphalangeal||58 (26.4)||128||100/127 (78.7)||27/127 (21.3)||< 0.0001|
|Wrist||57 (25.9)||83||49/82 (59.8)||33/82 (40.2)||0.08|
|Metacarpophalangeal||51 (23.2)||109||79/107 (73.8)||28/107 (26.2)||< 0.0001|
|Metatarsophalangeal||14 (6.4)||16||12/14 (85.7)||2/14 (14.3)||0.008|
|Foot interphalangeal||14 (6.4)||21||18/20 (90.0)||2/20 (10.0)||0.0003|
|Shoulder||8 (3.6)||11||5/7 (71.4)||2/7 (28.6)||–|
|Intertarsal||8 (3.6)||10||9 (90.0)||1 (10.0)||–|
|Hip||2 (0.9)||2||1/1 (100.0)||0/1 (0.0)||–|
|Temporomandibular||1 (0.5)||1||1 (100.0)||0 (0.0)||–|
|Total number of injected joints||–||1,096||770/1,079 (71.4)||309/1,079 (28.6)||< 0.0001|
The survival analysis with flare of synovitis as the event of interest is shown in Figure 1. The cumulative probability of survival without synovitis flare was 50%, 31.5%, and 19.5% at 1, 2, and 3 years, respectively. Looking at the slope of the curve, it appears that the probability of flare reached the maximum value in the first year after the IAC injection and decreased steadily thereafter.
Figure 1. Cumulative probability of survival without synovitis flare in the entire patient sample, obtained by the Kaplan-Meier method.
Download figure to PowerPoint
Table 3 shows, for each category of risk factors tested for their association with flare of synovitis, the number of patients with flare of synovitis in the risk factor category of the total number of patients, the time to synovitis flare expressed in terms of person-years, and the incidence rate of flare. Factors associated with a greater risk of flare and a shorter survival time were a negative ANA status, a positive CRP value, a polyarticular (versus an oligoarticular) disease course, and the lack of concomitant treatment with methotrexate. When the predictive role of variables was tested in a Cox proportional hazards regression model, the results showed that a positive CRP value, a negative ANA status, the lack of concomitant treatment with methotrexate, and a polyarticular (versus an oligoarticular) disease course were the only parameters associated with the risk of synovitis in injected joints (Table 4). A sensitivity analysis performed after the exclusion of the 15 patients with systemic arthritis led us to obtain similar findings (data not shown).
Table 3. Risk factors tested for associations with flare of synovitis*
| ||E/N (%)||Person-years||Incidence rate × 100 person-years||Pa|
|Sex|| || || ||0.45|
|Female||115/175 (65.7)||207.64||55.38|| |
|Male||31/45 (68.9)||48.29||64.19|| |
|Age at disease onset, years|| || || ||0.76|
|<2.5||80/117 (68.4)||138.77||57.65|| |
|≥2.5||66/103 (64.1)||117.16||56.30|| |
|Disease duration at IAC injection, years|| || || ||0.95|
|<1||83/132 (62.9)||143.50||57.84|| |
|≥1||63/88 (71.6)||112.43||56.04|| |
|ILAR category|| || || ||0.23|
|Nonsystemic||135/205 (65.9)||242.78||55.60|| |
|Systemic||11/15 (73.3)||13.15||83.65|| |
|Disease courseb|| || || ||0.018|
|Polyarthritis||119/169 (70.4)||191.14||62.26|| |
|Oligoarthritis||16/36 (44.4)||51.64||30.98|| |
|Systemic arthritis||11/15 (73.3)||13.15||83.65|| |
|Antinuclear antibody status|| || || ||0.004|
|Positive||109/175 (62.3)||219.10||49.75|| |
|Negative||25/30 (83.3)||24.41||102.42|| |
|General anesthesia|| || || ||0.29|
|Yes||104/169 (61.5)||192.42||54.05|| |
|No||35/43 (81.4)||53.41||65.53|| |
|Erythrocyte sedimentation rate, mm/hour|| || || ||0.66|
|≤20||26/44 (59.1)||44.75||58.10|| |
|>20||110/157 (70.1)||182.41||60.30|| |
|C-reactive protein level, mg/dl|| || || ||0.0004|
|≤0.46||27/52 (51.9)||82.88||32.58|| |
|>0.46||107/151 (70.9)||149.89||71.39|| |
|Ongoing or newly started methotrexate|| || || ||0.022|
|Yes||73/125 (58.4)||148.93||49.02|| |
|No||73/95 (76.8)||107.00||68.22|| |
|Ongoing or newly started biologic agents|| || || ||0.73|
|Yes||15/21 (71.4)||25.16||59.62|| |
|No||131/199 (65.8)||230.77||59.77|| |
|Ongoing or newly started systemic corticosteroids|| || || ||0.20|
|Yes||19/25 (76.0)||26.52||71.64|| |
|No||127/195 (65.1)||229.41||55.36|| |
|Number of joints injected|| || || ||0.15|
|3 or 4||90/135 (66.7)||179.11||50.25|| |
|≥5||56/85 (65.9)||76.82||72.90|| |
|Injection in specific joints|| || || || |
|Elbow|| || || ||0.49|
|Yes||46/64 (71.9)||75.71||60.76|| |
|No||100/156 (64.1)||180.22||55.49|| |
|Wrist|| || || ||0.59|
|Yes||42/57 (73.7)||67.88||61.87|| |
|No||104/163 (63.8)||188.05||55.30|| |
|Metacarpophalangeal|| || || ||0.27|
|Yes||25/51 (49.0)||50.78||49.23|| |
|No||121/169 (71.6)||205.15||58.98|| |
|Proximal interphalangeal|| || || ||0.61|
|Yes||29/58 (50.0)||51.54||56.27|| |
|No||117/162 (72.2)||204.39||57.24|| |
|Knee|| || || ||0.45|
|Yes||127/186 (68.3)||216.98||58.53|| |
|No||19/34 (55.9)||38.95||48.78|| |
|Ankle|| || || ||0.27|
|Yes||116/168 (69.0)||193.15||60.06|| |
|Subtalar|| || || ||0.78|
|Yes||44/67 (65.7)||71.22||61.78|| |
|No||102/153 (66.7)||184.71||55.22|| |
Table 4. Cox regression model*
|Risk factor||HR (95% CI)||Pa|
|ANA status: negative (reference category: positive)||2.97 (1.71–5.18)||0.0004|
|CRP level >0.46 mg/dl (reference category: ≤0.46 mg/dl)||2.40 (1.48–3.88)||0.0001|
|Ongoing or newly started methotrexate: no (reference category: yes)||1.91 (1.30–2.81)||0.0011|
|Disease course|| ||0.017|
|Polyarthritis (reference category: oligoarthritis)||2.04 (1.12–3.71)|| |
|Systemic arthritis (reference category: oligoarthritis)||1.13 (0.41–3.06)|| |
The most common side effect was skin hypopigmentation or subcutaneous atrophy at the site of the injection. This adverse event was seen in 0.9% of injected joints. A few patients developed flushing or redness of the cheeks 24–48 hours after the procedure as a result of systemic corticosteroid absorption. However, no case of overt or symptomatic Cushingoid syndrome was observed. Side effects of sedation most commonly encountered included confused state with irritability, drowsiness, and transient decrease in blood pressure or oxygenation. No serious adverse events or deaths occurred.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
We investigated the outcome of multiple IAC injections in a large sample of children with JIA, the majority of whom (76.8%) had a polyarticular disease course. Our patients had, on average, early disease, as shown by their median disease duration of 0.6 years. Therefore, our study provides information about the role of this therapeutic intervention as early treatment. The majority of children were candidates for multiple IAC injections as bridge therapy, aimed at achieving a quick control of inflammatory symptoms while awaiting the full effect of methotrexate, as an alternative to systemic corticosteroid administration or because etanercept could not be prescribed since the patients were ages <4 years.
One-third of the patients experienced sustained remission of synovitis in all injected joints after a median of 0.9 years from the injection procedure. Furthermore, the cumulative probability of survival without synovitis flare was 50% at 1 year and 31.5% at 2 years. These figures are in the range of the remission rates reported in JIA patients treated with etanercept in standard clinical practice or with abatacept in the context of a long-term extension phase of the registrative trial ([19-22]). Notably, only 56.8% of the patients had received concomitant methotrexate therapy and less than 10% were given biologic medications. These findings underscore the strong remission-inducing potential of this therapeutic modality.
However, although multiple IAC injection therapy proved highly effective over the short to medium term, most patients experienced a flare of synovitis over time. Nevertheless, considering only the 146 patients who experienced synovitis flare, the number of flared joints (n = 309) was much lower than the number of injected joints (n = 725). This finding suggests that patients who experience a synovitis flare after the IAC injection may require a less aggressive therapeutic intervention to regain complete disease control. In the case of synovitis flare, the decision to repeat an IAC injection in flaring joints versus changing systemic therapy depended on the number and type of the flared joints. In case of involvement of 1 or fewer large joints, a repeat IAC injection was often performed, whereas if the flare affected multiple large and small joints, a change in systemic therapy was usually made. Another important observation made in our study was that the risk of synovitis flare reached the maximum value in the first year after the IAC injection and decreased steadily thereafter. This indicates that the first year after the injection constitutes a therapeutic window in which the combination of IAC injection with a systemic therapeutic intervention (with methotrexate or biologic medications) may help maximize the long-term benefit.
On Cox regression analysis, the strongest predictors of synovitis flare were a positive CRP value, a negative ANA status, a polyarticular (versus an oligoarticular) disease course, and the lack of concomitant treatment with methotrexate. Another relevant flare predictor, identified only in univariate analysis, was the lack of general anesthesia, which suggests that carrying out the procedure in the operation room increases its effectiveness. This potential advantage should be weighed against the inherent risk of general anesthesia, however. Unlike our previous analysis of IAC therapy in JIA (), concomitant methotrexate administration entered multivariate analysis, whereas injection in the ankle joint did not. This disparity may depend on the fact that most patients included in the previous study were injected in 1 or 2 joints, whereas the current investigation was only focused on patients who underwent injection in ≥3 joints.
Based on the results of the present study, some speculations can be made regarding the potential place of multiple IAC injections in the current management of JIA. In general, this therapeutic modality represents a powerful remission-inducing intervention that can be applied to any child with active synovitis in large and small joints, and may be more advantageous for the control of arthritis, particularly in terms of risk of side effects, than systemic corticosteroid therapy.
The specific therapeutic indications may be different depending on the disease subset. Children with oligoarthritis who are injected in 3 or 4 joints have the greatest likelihood of experiencing long-lasting remission of synovitis, particularly if they have a positive ANA status and a negative CRP value. It is unclear, however, whether these patients deserve the sole administration of IAC injections or should be given concomitant methotrexate therapy. A response to this question will likely be provided by a controlled trial that is being conducted by the Italian Pediatric Rheumatology Study Group and is aimed at comparing the efficacy of IAC injections alone or in association with methotrexate in children with oligoarticular JIA (Agenzia Italiana del Farmaco project code FARM7Y279L; online at http://www.agenziafarmaco.gov.it/it/content/ricerca-e-sviluppo).
Children with polyarthritis may warrant the simultaneous administration of methotrexate, irrespective of the disease features or the number of injected joints, because this treatment may increase the probability of remission of joint synovitis or prolong its duration. The prescription of a biologic medication should be considered in case of a suboptimal response to IAC therapy in combination with methotrexate or short-term flare of synovitis. It remains to be established whether children with arthritis in the ankle or wrist joints, which may be less susceptible to the efficacy of IACs, should already be started with biologic therapy at the time of the injection.
IAC injections are also suited to induce remission of joint synovitis in children with systemic JIA. However, in our experience, the effect is usually short lived. In this patient subgroup, multiple IAC injection therapy may have a role as rescue therapy for severe or refractory polyarthritis or as a method to spare systemic corticosteroids. However, it has a secondary role with respect to traditional DMARDs and biologic medications.
Our study should be viewed in light of some potential limitations, primarily its retrospective nature. A retrospective analysis is subject to missing and possibly erroneous data. Because the study was not controlled, the results regarding therapeutic effectiveness and outcome predictors should be regarded with caution. We should acknowledge that the choice of censoring the followup at the time when patients had a relapse of arthritis in uninjected (i.e., previously unaffected) joints while synovitis was still in remission in injected joints might have led to an increased relapse rate and a shortened time to flare. Very few patients underwent injection in the hip and temporomandibular joints. Therefore, our results are of no value for IAC administration in these joints. However, it is our policy to treat early children with arthritis in the temporomandibular joints with methotrexate and children with hip disease with methotrexate and biologic medications, often in combination. Notably, active arthritis in the temporomandibular joints has been detected in a sizeable proportion of children with JIA in spite of systemic therapy with methotrexate or tumor necrosis factor antagonists (). This observation suggests that IAC injection in these joints may help prevent damage or growth disturbance. Very few joints were injected under imaging guidance, which does not allow us to compare the time to flare between injections made with or without radiographic or ultrasound assistance. Better success with radiographic guidance of IAC therapy for subtalar arthritis has been reported (). Our choice of using methylprednisolone acetate instead of triamcinolone hexacetonide for small and difficult to access joints may hamper comparability of our findings with those of investigators who inject these joints with triamcinolone hexacetonide.
In summary, our study showed that multiple IAC injection therapy was able to induce sustained remission of joint synovitis in a substantial percentage of patients, with a good safety profile. A positive CRP value, a negative ANA status, a polyarticular (versus an oligoarticular) disease course, and the lack of concomitant administration of methotrexate were the strongest predictors of flare of synovitis in injected joints. A controlled trial comparing multiple IAC injection therapy in combination with methotrexate versus methotrexate in combination with a tumor necrosis factor antagonist in children with polyarthritis is warranted.
- Top of page
- PATIENTS AND METHODS
- AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Ravelli had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Papadopoulou, Lanni, Consolaro, Martini, Ravelli.
Acquisition of data. Papadopoulou, Kostik, Gonzalez-Fernandez, Nieto-Gonzalez.
Analysis and interpretation of data. Papadopoulou, Bohm, Pistorio, Lanni, Consolaro, Martini, Ravelli.