The SAPHO syndrome revisited with an emphasis on ...

The SAPHO syndrome revisited with an emphasis on spinal manifestations

Antonio Leone, Victor N. CassarPullicino, Roberto Casale, Nicola Magarelli, Alessia Semprini & Cesare Colosimo Skeletal Radiology Journal of the International Skeletal Society A Journal of Radiology, Pathology and Orthopedics ISSN 0364-2348 Skeletal Radiol DOI 10.1007/s00256-014-2025-0

1 23

Your article is protected by copyright and all rights are held exclusively by ISS. This eoffprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23

Author's personal copy Skeletal Radiol DOI 10.1007/s00256-014-2025-0

REVIEW ARTICLE

The SAPHO syndrome revisited with an emphasis on spinal manifestations Antonio Leone & Victor N. Cassar-Pullicino & Roberto Casale & Nicola Magarelli & Alessia Semprini & Cesare Colosimo

Received: 12 July 2014 / Revised: 17 September 2014 / Accepted: 28 September 2014 # ISS 2014

Abstract The synovitis, acne, pustulosis, hyperostosis, and osteitis (SAPHO) syndrome includes a group of chronic, relapsing, inflammatory musculoskeletal disorders with similar manifestations, in particular synovitis, hyperostosis, and osteitis, which may or may not be associated with neutrophilic skin eruptions such as palmoplantar pustulosis and acne conglobata. The syndrome occurs at any age, can involve any skeletal site, and its imaging appearances are variable, depending on the stage/age of the lesion and imaging method. The diagnosis is difficult if there is no skin disease. Awareness of the imaging appearances, especially in the spine, may help the radiologist in avoiding misdiagnosis (e.g., infection, tumor) and unnecessary invasive procedures, while facilitating early diagnosis and selection of an effective treatment. In this article, we provide an overview of the radiological appearances of SAPHO syndrome, focusing on the magnetic resonance imaging findings of vertebral involvement, and present relevant clinical and pathological features that assist early diagnosis. Keywords SAPHO syndrome . Spondyloarthropathy . Synovitis . Osteitis . Hyperostosis

Introduction The association between musculoskeletal lesions and acne conglobata was originally documented by Windom et al. [1] A. Leone (*) : R. Casale : N. Magarelli : A. Semprini : C. Colosimo Department of Radiological Sciences, Catholic University, School of Medicine, Largo A. Gemelli 1, 00168 Rome, Italy e-mail: [email protected] V. N. Cassar-Pullicino Department of Diagnostic Imaging, The Robert Jones and Agnes Hunt Orthopaedic and District Hospital, Oswestry, Shropshire, England, UK

in 1961. Subsequently, several reports of patients with musculoskeletal and skin lesions appeared in the literature with up to 50 different names, including: “bilateral clavicular osteomyelitis with palmoplantar pustulosis (PPP)” [2], “subacute and chronic symmetric osteomyelitis” [3], “sterno-costo-clavicular hyperostosis”, [4] “chronic recurrent multifocal osteomyelitis” (CRMO) [5, 6], “inter-sterno-costo-clavicular ossification” [7], “pustulotic arthro-osteitis” [8], “acquired hyperostosis syndrome” [9], and pyogenic arthritis, pyoderma gangrenosum, and severe cystic acne [10]. The acronym CRMO was first coined in 1978 by Bjorksten et al. [5] to describe the association of multifocal osseous disease with PPP in children and adolescents with a predilection for metaphyses or metaphyseal equivalents of tubular long bones, and clavicles. However, most patients do not have direct temporal correlation between skin lesions and bony lesions as the skin lesions can occur several years after the first bone symptoms [11]. In CRMO, multiple osseous sites are usually involved, and the clinical course is characterized by exacerbations and remissions typically lasting years. However, CRMO is not necessarily chronic or recurrent, and there is great variability in the numbers of sites affected; unifocal involvement, especially of the clavicle, has been described, and the number of affected areas tends to decrease with age [12]. In 1987, Chamot et al. [13] performed a multicenter survey of 85 patients with ostearticular disorders associated with PPP and severe acne, and they first introduced the acronym SAPHO (synovitis, acne, pustulosis, hyperostosis, and osteitis) in order to bring together the heterogeneous descriptions as a single syndrome. According to Chamot et al. [13] and other authors [14, 15], we consider SAPHO as an “umbrella” acronym covering all the above-mentioned entities, including CRMO. The fundamental component of the SAPHO spectrum is a sterile inflammatory osteitis, which may or may not be associated with skin lesions. The main target sites are the anterior chest wall, spine, and peripheral skeleton. Diagnosing

Author's personal copy Skeletal Radiol

SAPHO syndrome is not difficult when typical bone lesions are located in characteristic target sites. The diagnosis is much more difficult if atypical sites are involved and the patient is free of skin disease. In this article, we present the broad spectrum of radiological appearances of SAPHO, focusing on the magnetic resonance (MR) imaging findings of vertebral involvement, and present relevant clinical and pathological features that assist early diagnosis.

Etiological hypotheses The etiopathogenesis of SAPHO syndrome still remains to be elucidated, but probably involves infectious, immunological, and genetic mechanisms [16]. It is thought that low-virulence pathogens may trigger an exaggerated autoimmune inflammatory response of the bone marrow in genetically susceptible individuals, leading to a form of “reactive osteitis.” Among the microorganisms isolated in bone biopsy specimens or synovial tissue from patients with SAPHO syndrome, the corynebacterium P. acnes, a member of normal flora of the skin and gastrointestinal tract, known to be involved in acne and possibly playing a role in PPP and pyoderma gangrenosum, is the most common implicated organism [16–20]. Trimble et al. [21] observed that intra-articular injection of inactivated P. acnes in rats can cause joint lesions and bone erosions. Assmann et al. [17] reported a positive bacterial culture for P. acnes in 14 of 21 (67 %) patients who had undergone a needle biopsy of osteitis lesions. Nevertheless, the bone lesions are often sterile [22, 23]. A possible explanation could involve the ability of P. acnes to persist in bone lesions in a form that is incompatible with culturing [24]. It has been hypothesized that SAPHO syndrome represents an autoimmune process triggered by infectious agents (i.e., P. acnes) inducing the production of cross-reacting antibodies (molecular mimicry) or the inflammatory second-hit mediated by Toll-like receptors (TLRs) [24]. The concept of molecular mimicry is based on a structural similarity of microbial and host antigens, and it has been proposed as a promoting factor for the expansion of the microorganism when infectious agents are not recognized and not completely eliminated [25, 26]. TLRs are transmembrane proteins, usually expressed in sentinel cells such as macrophages and dendritic cells, which recognize several components of microbial cell walls. Recognition of pathogens by TLRs leads to the induction of accessory signals (costimulators and cytokines) that are necessary for the activation of T cells [27]. P. acnes can possibly stimulate both the innate and the T-cell-mediated immune systems. The corynebacterium produces a number of molecules with chemoattractant properties that activate the innate immune response through TLR9 signaling [28]. In addition, the ability of P. acnes to trigger interleukin (IL)-1, IL-8, and IL-18 and tumor necrosis factor alpha (TNFα) release by monocytes,

keratinocytes, sebocytes, and dendritic cells through a paracrine auto-amplified feedback loop has been demonstrated [18]. These cytokines can be responsible for the maintenance of the clinical manifestations [29]. Regarding the involvement of TNFα in generating or perpetuating the rheumatic manifestations of SAPHO syndrome, Wagner et al. [30] documented TNFα overexpression in a focus of mandibular osteitis. Further support for involvement of TNFα comes from the promising benefits seen in SAPHO patients treated with TNFα antagonists [30–32]. The immunological response could be an attempt to eliminate the germ inducing the perpetuation of the inflammation [22, 33]. The presence of genetic susceptibility factors is probably required for the development of SAPHO syndrome. An hereditary basis of SAPHO has been suggested secondary to observation of disease in affected siblings sharing numerous human leukocyte antigens phenotypes, affected parents and children, and concordant monozygotic twins [29, 34]. The higher prevalence of inflammatory bowel disease in SAPHO suggests the possibility of a gene mutation associated with Crohn’s disease, such as NOD2/CARD15 (nucleotide-binding oligomerization domain protein 2/caspase recruitment domain 15), which is known to lead to exaggerated immune responses [23]. Furthermore, the Majeed syndrome and pyogenic sterile arthritis, pyoderma gangrenosum and acne (PAPA) syndrome, two other diseases having some manifestations similar to SAPHO, have been found to have genetic origins [35–38]. Majeed syndrome is an autosomal recessive disorder consisting of CRMO, neutrophilic dermatosis, and congenital dyserythropoietic anemia. This condition is caused by mutations in the LPIN2 gene, which encodes the lipin-2 protein thought to be involved in the apoptosis of polymorphonuclear leukocytes [35, 36]. PAPA syndrome is a rare autosomal dominant inherited autoinflammatory syndrome characterized by pyogenic sterile arthritis and accompanied by pyoderma gangrenosum and acne. It arises from mutations in the prolineserine-threonine phosphatase-interacting protein 1 gene (PSTPIP1) located on chromosome 15 [37]. The PSTPIP proteins are involved in regulating the immune response via several mechanisms mediated by T cells [38] and apoptosis [39]. The role of genetic factors in SAPHO syndrome is further supported by a murine model with lesions similar to those seen in SAPHO. The mouse shows mutations in the PSTPIP2 susceptibility gene, which is located on chromosome 18 (at a locus designated “cmo” for chronic multifocal osteomyelitis) [40, 41]. However, neither the PSTPIP1/PSTPIP2 nor the LPIN2 or NOD2/CARD15 gene mutations were found to increase in frequency in patients with SAPHO [33, 34]. An association with seronegative spondyloarthropathies has also been proposed because of a variety of common findings including a lack of association with rheumatoid factor, high incidence of spinal lesions/sacroiliitis, development of marginal or more commonly non-marginal


syndesmophytes, and association with inflammatory bowel disease in 8 % of patients [42, 43]. Compared to other spondyloarthropathies, human leukocyte antigen-B27 (HLAB27) may play a smaller role in the pathogenesis of SAPHO syndrome; although HLA-B27 has been reported in 4–30 % of patients [42–45], it is usually negative [7, 8, 42, 46, 47].

Clinical features SAPHO is considered a rare disease, but its true prevalence remains unknown and is probably greatly underestimated because of the confusing symptoms and the possible lack of skin involvement [48]. It can occur at any age, rarely beyond the 6th decade of life [49], and is characterized by repeated episodes of remission and recurrences. There is a slightly higher incidence in the female population [42]. Patients mostly present with musculoskeletal complaints such as pain, tenderness, swelling, or limited range of motion referable to the involved skeletal sites, most commonly the anterior chest wall and spine in adults [15]. Conversely, in children and adolescents, the disease predominates in the long bones followed by the clavicle and spine [50]. In fact, the site of involvement seems to be age dependent [15, 50]. Although patients may present with a single site of involvement [51, 52], in most of them multiple different osteoarticular sites are often involved in a synchronous or metachronous manner [15, 43, 50], with a maximum number of 18 sites reported [53]. For this reason, a thorough search should be made to evaluate for the presence of subclinical foci due to the polyostotic nature of the disease [54]. Fever may be present. Laboratory tests are usually of little value. The white cell count, levels of Creactive protein, and erythrocyte sedimentation rate are usually normal or slightly elevated during exacerbations [42, 44]. The skin lesions are neutrophilic dermatoses, which comprise a heterogeneous but linked spectrum of disorders characterized by perivascular and diffuse neutrophilic infiltrates with no evidence of infection [55]. The most common are PPP and acne, affecting close to 60 % and 15–18 % of patients with SAPHO, respectively [13, 43, 49, 50, 56]. PPP is a chronic and recurrent skin condition thought to be a special variant of psoriasis and characterized by 2–4-mm yellowish intradermal sterile pustules, erythema, and hyperkeratosis on the palms and soles [14]. Acne is usually severe and can present as acne conglobata, acne fulminans, and hidradenitis suppurativa (Fig. 1). Other rare skin manifestations of SAPHO include Sweet’s syndrome, characterized by fever, peripheral leukocytosis, and erythematous painful skin lesions, and pyoderma gangrenosum, which is a rare chronic inflammatory skin disease characterized by the presence of nodules and pustules with progressively enlarging ulcers and or cutaneous necrosis [11, 42]. Most patients do not have direct temporal correlation between skin and osteoarticular lesions. Skin lesions may

Fig. 1 A 42-year-old man with SAPHO. a Hidradenitis suppurativa of the axilla. b Acne conglobata with multiple nodules and pustules

occur before, contemporary with, or after the onset of osteoarticular changes. In most cases, skin lesions occurred within an interval of 2 years before or after the onset of osteoarticular lesions [8, 57], but delays as long as 12 and 38 years have also been reported [8, 43]. However, it should be kept in mind that the absence of current skin lesions does not rule out the possibility of SAPHO. Approximately one third of the patients with osteoarticular lesions may never experience skin lesions [13, 58].

Imaging features Osteoarticular changes of SAPHO implicate synovitis, hyperostosis, and aseptic osteitis, but inflammatory enthesopathy leading to ligamentous ossification may also be a feature with potential development of bony bridging across joints [14, 15]. Synovitis, often due to osteitis extending to adjacent articular structures, generally manifests in oligo- or polyarthritis. It is most commonly encountered in the upper anterior chest wall (i.e., sternoclavicular, costoclavicular, and manubriosternal joints) and sacroiliac joints [14, 58]. Peripheral involvement with oligoarticular and asymmetric arthritis is also observed and most commonly involves the knees, hips, and ankles, but small joints of the hands and feet may also be involved [42]. Arthritis commonly leads to joint space narrowing, periarticular osteopenia, and bone erosions that mimic seronegative


spondyloarthropathies [14]. Ankylosis may occur and has been associated with considerable relief from pain [59]. The most important hallmarks of the SAPHO syndrome, however, are hyperostosis and osteitis. Hyperostosis refers to excessive bone formation (osteogenesis), which can occur within the medullary cavity or on both surfaces (endosteal and periosteal) of the cortex, leading to cortical thickening, bone hypertrophy, and narrowing of the medullary canal [14]. Osteitis refers to inflammation of bone, which may involve the cortex, medullary cavity, or both. The main target area is the anterior chest wall, with lesser involvement of the spine and peripheral skeleton. Hyperostosis and osteitis appear at radiography and multidetector computed tomography (MDCT) as osteosclerosis with or without areas of osteolysis [46, 50]. The sensitivity of radiography, however, is very low in the early stages of the disease (13 %); therefore, radiographic examinations of affected areas are often normal [60]. Maugars et al. [59] showed that whereas radiographs performed in the first 3 months of the disease course were normal in 80 %, all patients had abnormal radiographs at the end of follow-up. MDCT is thought to be the imaging modality of choice for demonstrating the various osteoarticular lesions and their extent [15]. This imaging modality is particularly useful and recommended in delineating involvement of the anterior chest wall because this area may be poorly demonstrated radiographically [61]. In the presence of active lesions, MR imaging with fluidsensitive fat-suppressed sequences, i.e., short-tau inversion recovery (STIR) or fat-saturated T2-weighted images, may depict bone marrow and soft tissue edema, a feature that helps differentiate active lesions from chronic ones [15, 50]. Active lesions demonstrate either focal or diffuse decreased signal intensity on T1-weighted images, and high signal intensity on both T2-weighted and fluid-sensitive images, whereas chronic sclerotic bone lesions appear hypointense on both T1- and T2weighted images [50, 61]. Intravenous contrast material may be administered for initial examination to increase morphologic evaluation of organ systems other than the musculoskeletal system. However, it may not be required for follow-up evaluation because fluid-sensitive images are highly sensitive for detection of osseous lesions [60]. In the series of Fritz et al. [60], all bone lesions showed abnormal hyperintensity on STIR images and post-contrast enhancement, which are features of edema-like lesions; furthermore, the conspicuity of osseous lesions was similar on STIR images and contrastenhanced fat-suppressed T1-weighted fast SE images. Targeted MR imaging, however, cannot be used to assess multifocality. Whole-body MR imaging, with coronal T1weighted and STIR sequences, is being increasingly used for evaluation of multifocal bone lesions and follow-up evaluation [60, 62, 63]. These techniques may visualize most of the osteoarticular changes of SAPHO in one examination and are

very well suited for assessing disease activity and therapeutic response [63]. Whole-body bone scintigraphy using 99mTclabeled diphosphonate is also highly sensitive to detect multifocal bone lesions, and the so-called “bull’s head” pattern of increased sternocostoclavicular inflammatory activity, with the manubrium sterni representing the upper skull and the inflamed sternoclavicular joint with the adjacent claviculae forming the horns, is considered highly specific for the SAPHO spectrum [64]. This finding can be important in patients without skin disease at the time of presentation to prevent unnecessary invasive procedures or treatment. In view of the possible fluctuation of the clinical symptoms, wholebody bone scintigraphy may be helpful to detect clinically occult sites of disease that would otherwise go unnoticed; but both active and chronic subclinical inflammatory lesions show increased tracer uptake [65–67]. Furthermore, compared with whole-body MR imaging, 99mTc-scintigraphy has the added disadvantage of using ionizing radiation. The radiation-free assessment of the entire axial skeleton and large portions of the appendicular skeleton becomes particularly important in the pediatric population and when repeated follow-up imaging is likely to be necessary. Several case reports have shown the utility of F-18 fluorodeoxyglucose-positron emission tomography (F-18 FDG-PET) alone or as the functional component of F-18 FDG-PET/CT to differentiate active from healed chronic inflammatory lesions, because F-18 FDG-PET shows increased uptake only in lesions with active inflammation, and to distinguish active inflammatory lesions in SAPHO syndrome from metastatic bone lesions [65–68]. However, distinguishing a malignant from benign inflammatory process in cases with a history of malignancy, or multiple bony lesions with no skin disease, is challenging because both conditions can show increased F-18 FDG uptake [66, 69]. Furthermore, there is considerable overlap between standardized uptake values (SUVs) at single time points of inflammatory and neoplastic disease processes causing difficulty in correctly interpreting F-18 FDG-PET data [70]. SUV is a semiquantitative value that provides an index of regional tracer uptake normalized to the administered dose of tracer. There are currently no major studies that define the rule of F-18 FDG-PET/ MR imaging in SAPHO, but they are likely to be employed in the future.

Anterior chest wall (sterno-costo-clavicular junction) In adults, the most frequent site for the SAPHO syndrome is the anterior chest wall, accounting for 60–95 % of patients [13–15, 42, 43, 50]. Although any component of the anterior chest wall can be involved, the costochondral (52 %), sternoclavicular (48 %), manubriosternal (34 %), and costosternal (7 %) junctions are the most commonly affected


sites [15, 50] (Figs. 2 and 3). The most frequent radiological feature is hyperostosis, which can be associated with enthesopathy leading to ligamentous ossification and bony bridges between ribs and across adjacent joints [7, 51]. Inflammatory enthesopathy of the costoclavicular ligament (48 % of patients) and small hyperostotic foci of at least 5 mm diameter at the sternal end of the first pair of ribs are important early diagnostic findings [9]. Joint erosions, which are the result of a primary arthritis or an extension of the adjacent osteitis, frequently lead to ankylosis, particularly of the sternocostal and costoclavicular junctions (Fig. 3). Inflammatory involvement of adjacent soft tissue is often found and may be responsible for thoracic outlet syndrome and axillary vein, subclavian vein, and superior vena cava compression or obstruction [9, 43, 56, 59, 71]. Furthermore, soft tissue involvement may give the appearance of an aggressive process such as Ewing's sarcoma, lymphoma, or histiocytosis [72]. Osteoarticular changes usually develop in three stages [7, 15, 50]: stage 1 is localized to the area of the costoclavicular ligament and may be a primary enthesopathy; in stage 2, an arthropathy of the sternoclavicular joint develops with

Fig. 3 Sterno-costo-clavicular joint involvement in a 30-year-old man. a Reformatted coronal CT image shows sclerosis (asterisk) and erosive change (arrows) of the medial end of the right clavicle and adjacent manubrium. b, c Axial chest CT images obtained 1 year later show disease progression resulting in marked sternoclavicular erosive change (small arrows in b), hyperostosis, osteosclerosis, and ankylosis of the first sternocostal junctions (arrows in b and c)

osteolytic and osteosclerotic changes of the medial end of the clavicle, adjacent sternum, first rib, and costal cartilage (Fig. 2); stage 3 is a further progression of osteosclerosis, hyperostosis, and hypertrophy of these structures with arthritis and potential ankylosis in the adjacent joints (Fig. 3). When the disease is advanced, these changes are clearly identified by radiography, but early changes require the use of CT and MR imaging for a correct diagnosis. CT can demonstrate the extent of disease with lytic areas and surrounding sclerosis and hyperostosis (Fig. 3). MR imaging reveals bone marrow edema (Fig. 2) with surrounding periosteal reaction. In contrast to adults, the clavicle is the most common anterior chest wall bone involved by SAPHO in children. The sternoclavicular joint, sternum, and ribs are rarely affected [9, 15, 50]. Ligamentous ossification and bony bridging across the sternoclavicular joint have not been described. Clavicular disease typically manifests as an area of osteolysis in the medial end of the clavicle with a surrounding periosteal reaction as well as surrounding sclerosis and hyperostosis during remission [73]. Fig. 2 Sterno-costo-clavicular junction involvement in a 20-year-old man. a, b Coronal fat-saturated T2-weighted MR images showing marrow edema of the medial end of the clavicles (arrows in a) and sternal components of the first costosternal junctions (arrows in b). c Reformatted coronal CT image showing erosion of the medial end of the left clavicle (arrow) and sclerosis of the sternal components of the first costosternal junctions (small arrows)

Axial skeleton The second most common site of skeletal involvement in all ages is the spine, where abnormalities are found in 32–52 % of


adult patients [43, 59, 74]. Potentially affected are all vertebrae from the mid-cervical spine to the sacrum; however, the thoracolumbar spine is the most frequent site of involvement [55]. Pain and stiffness, usually localized, may draw attention. The affection is usually segmental, but two or more adjacent vertebrae can be involved [74, 75] (Fig. 4). The axial skeleton manifestations of SAPHO can take six forms and can occur in various combinations. These include vertebral body corner lesions, non-specific spondylodiscitis resembling infectious spondylodiscitis, and osteolytic lesions with variable degrees of vertebral body collapse, seen in adults and children, as well as osteosclerosis of one or more vertebral bodies with the development of hyperostosis, paravertebral ossification, and sacroiliitis, usually seen in adults [15, 61]. Vertebral body corner lesion This term is used to describe a focal cortical erosion involving one of the vertebral body corners, most commonly one of the anterior corners [75]. In 23 (96 %) of the 24 lesions identified by Laredo et al. [75] in 12 patients with SAPHO and investigated by using MR imaging, the erosion involved one anterior corner, whereas in the remaining lesion, it involved a posterior corner. These authors [75] suggested that the corner erosion seen at MR imaging may indicate enthesitis that is equivalent to the radiographic Romanus lesion found with Fig. 4 Sagittal MR and axial CT images of the thoracolumbar spine in an 18-year-old boy. a, b, c T1-weighted images show low signal intensity within T7 (arrow in a), T8, and T9 vertebral bodies (arrows in b) as well as the anterosuperior corner of the L4 and L5 vertebrae (arrows in c). d, e, f Corresponding fat-suppressed T2-weighted and g, h, i contrastenhanced fat-suppressed images show high-signal intensity and enhancement in the abovementioned vertebral sites. The intervertebral discs are preserved. l, m, n, o, p Axial CT images passing through the vertebral sites of abnormal signal intensity show cortical erosions and marginal sclerosis (arrow in l, m, n, o, p)

spondyloarthropathies, and with ankylosing spondylitis in particular [76]. Romanus lesions consist of erosions manifesting at the insertion of the outer fibers of the annulus fibrosus on the ring apophysis of the vertebral endplate. Because such a junction between bone and a ligamentous structure is an enthesis by definition, a Romanus lesion can be considered as an enthesitis of the anterior or posterior longitudinal ligamentous complexes. The condition undergoes several radiographic stages. At the first stage, nongranulomatous inflammatory tissue leads to the destruction of the attachment of ligament to bone resulting in a cortical erosion that can be seen on radiographs. Later in the course of the disease, the vertebral body corner undergoes a reactive sclerosis, a finding referred to as a “shiny corner.” Only at this stage is the lesion clearly depicted by radiography. The end stage consists of paravertebral ossification with formation of syndesmophytes [76–78]. Laredo et al. [75] believe that the corner erosion seen with SAPHO on MR images corresponds to the first stage of this process, which precedes the development of radiographic changes. MR imaging may effectively identify early erosive changes of vertebral corners and edema-like signal intensity changes of the adjacent bone marrow in radiographically normal vertebral sites [74]. In active disease, SAPHO-induced bone marrow edema and hyperemia manifest on MR imaging as a decrease in signal on T1-weighted images with a corresponding increase in signal on fluid-sensitive images and marked


enhancement on post-contrast fat-suppressed T1-weighted images [75, 77, 78] (Fig. 4). Reactive sclerosis will appear as low signal intensity corners on all MR sequences [75, 78]. When inflammation decreases, the “shiny corners” may become hyperintense rather than hypointense on T1-weighted MR images because of the post-inflammatory fatty degeneration of bone marrow. These changes, named fatty Romanus lesions by Bennett et al. [79], can be found in SAPHO [80] or ankylosing spondylitis [79]. Unlike spondyloarthropathies, the vertebral corner involvement in SAPHO progresses to the adjacent vertebral endplate and/or the anterior cortex of the vertebral body (79 %), and it also often leads to prevertebral soft-tissue thickness, which supports the diagnosis [75, 81] (Figs. 5 and 6). Furthermore, the involvement of two or more contiguous vertebrae is very uncommon in cases of spondyloarthropathy [81–83]. The involvement of multiple vertebral levels in a noncontiguous fashion may simulate the MR imaging pattern produced by multiple metastases [75]. In contrast to metastasis, in which tumoral deposits are randomly distributed in the vertebral bone marrow, the signal intensity abnormalities in SAPHO syndrome are characteristically centered on the vertebral body corner (Fig. 4c, f, i). This criterion can raise the suspicion and help in reaching a proper diagnosis. The vertebral involvement during the course of SAPHO, like that of spondyloarthropathies, may be asymptomatic [75, 78, 83]. Therefore, MR imaging of the spine or, even better, whole-body MR imaging may be useful in looking for asymptomatic changes that support a suspected diagnosis of SAPHO [75] (Figs. 7 and 8). Whole-body MR imaging allows the simultaneous visualization of inflammatory changes in axial and nonaxial sites, reducing imaging times. Spatial resolution is similar to targeted MR imaging [48, 60, 62, 63]. Although the earliest inflammatory changes are best observed with MR imaging, CT appears to be more sensitive for depicting chronic changes such as erosions, sclerotic changes, hyperostosis, and bone formations located at the same sites [84] (Fig. 9).

Fig. 5 The same patient as in Fig. 1. a Sagittal T1-weighted and b corresponding T2-weighted MR images showing bone marrow edema in the anterior aspects of the T6 and T7 vertebral bodies. Note the prevertebral soft-tissue swelling (white arrow in a and b) and the presence of subcutaneous pustules (black arrow in a and b)

Non-specific spondylodiscitis Involvement of two or more contiguous vertebrae with cortical erosions and underlying subchondral sclerosis in the endplates, on either side of an intervertebral disc, can be present in patients with SAPHO and may resemble infectious spondylodiscitis. These lesions are usually localized to the central or anterior portions of the discovertebral junction in up to 29–32 % of cases [74, 75] and may be the initial manifestation of SAPHO [85, 86]. The adjacent intervertebral disc space is usually well maintained but may be reduced. Radiography and CT reveal erosions with sclerosing remodeling of the end plates and narrowing of the adjacent disc space if present. MR imaging, in addition, shows and defines the extent of the surrounding bone marrow edema, which may be focally centered on the endplate erosions or diffuse appearing as high signal on fluid-sensitive sequences [15, 80]. In most of these lesions, the low signal intensity of the disc on fluid-sensitive images and the absence of disc space enhancement on post-contrast images aid greatly in arriving at the correct diagnosis. However, the presence of both high signal on T2-weighted images and disc space enhancement on post-contrast images can be seen in up to 30 % of cases [52, 74, 80, 87], rendering the differentiation of the SAPHO spectrum from infectious spondylodiscitis more difficult. The differential diagnosis of SAPHO syndrome versus infectious spondylodiscitis may be further complicated if a prevertebral soft-tissue swelling coexists (Fig. 6e). Laredo et al. [75] found a soft-tissue thickening, always less than 1 cm in thickness, with gadolinium enhancement in 8 (33 %) of the 24 vertebral lesions of their series. However, absence of an abscess or of epidural involvement, foci of spondylodiscitis at different consecutive or non-consecutive spinal levels in the same patient, and earlier productive changes are uncommon in infection [50, 74, 80, 88]. Furthermore, the presence of typical vertebral corner erosion in another spinal segment is suggestive of SAPHO [75]. Biopsy of the involved disc space in patients with SAPHO syndrome reveals a chronic nonspecific inflammatory process with mild fibrosis, and cultures are classically sterile [52, 74]. The disc space involvement is equivalent to an aseptic spondylodiscitis or Andersson lesion. This condition, first described in 1937 by Andersson [89], is an inflammatory disorder of the discovertebral region which occurs in up to 18 % of patients with ankylosing spondylitis [90]. The exact etiology of Andersson lesions is unknown; however, they are characterized by inflammation (Andersson A) and stress fractures (Andersson B), resulting in the development of pseudoarthrosis [91]. In cases where a biopsy was taken, reactive bone tissue was usually found associated with chronic sterile nonspecific inflammation, in part with replacement of bone by fibrous tissue or associated with the picture of chronic osteomyelitis with an aspecific cellular infiltrate composed of


Fig. 6 MR images of the lumbar spine in a 29-year-old man before and after anti-TNFα treatment. a,b Sagittal T1-weighted MR images show low signal intensity of the L4 vertebral body (arrow in a), spinous process (small arrow in a), and right pedicle (arrow in b) of the L3 vertebra. c,d Corresponding T2-weighted fat-suppressed MR images show high signal intensity in the same vertebral sites, consistent with bone marrow edema

(arrow and small arrow in c, arrow in d). e Axial contrast-enhanced fatsuppressed T1-weighted MR image passing through the L4 vertebra shows right prevertebral soft-tissue thickening and enhancement (arrow). This feature can be mistaken for spread of osteomyelitis or discitis. f,g Sagittal T2-weighted fat-suppressed MR images obtained 6 months later (after anti-TNFα treatment) do not show persistence of existing lesions

lymphocytes, plasma cells, and, less often, macrophages [74, 75, 91].

patients with complete vertebral body collapse (vertebra plana) may undergo kyphoscoliotic deformity and occasionally spinal canal stenosis and spinal cord injury [45, 46]. Vertebral deformity due to vertebra plana is rare in the adult population; it is much more likely to be seen during childhood [93–95]. Unlike that seen in the vertebra plana caused by eosinophilic granuloma, reconstitution of vertebral height is not a feature in SAPHO [92]. Absence of associated paravertebral soft-tissue mass lesions may be important for distinguishing SAPHO involvement from malignancy.

Osteolytic lesions The radiological findings of spinal involvement may include an osteolytic lesion with partial or complete vertebral body collapse with no history of important trauma. Patients with partial collapse undergo gradual healing of the lesion with minor sequelae apart from minimal kyphosis [92]. In contrast,

Fig. 7 Asymptomatic vertebral involvement in a 14-year-old boy with acne conglobata of the face, buttock pain, bilateral sacroileitis, and multiple symptomatic areas of bone marrow edema of the pelvis, right acetabulum, and left femoral greater trochanter (see Fig. 8). a Sagittal fatsuppressed T2-weighted MR image showing high signal intensity of the anterosuperior corner of L2 and L3 (arrows), and spinous process of L1 and L2 vertebrae (small arrows), consistent with bone marrow edema. b

Coronal fat-suppressed T2-weighted MR image showing bone marrow edema of the posteroinferior corners of T9 (arrows) and left posteroinferior corner of T7 and T8 vertebral bodies (small arrows). c,d Coronal fat-suppressed T2-weighted MR images showing bone marrow edema of the sternal end of the fifth left rib (arrow in c) and the lateral end of the clavicles (arrows in d)


Fig. 8 a Coronal oblique STIR and b corresponding T1-weighted MR images showing bilateral subchondral bone marrow edema in the sacral aspect of the sacro-iliac joints (arrows in a and b). c Axial STIR and d corresponding T1-weighted MR images showing bone marrow edema

within the posterior column of the right acetabulum (arrow in c and d). e Axial and f coronal STIR MR images showing high-signal intensity of the left greater trochanter (arrow in e) and iliac crests (arrows in f), consistent with bone marrow edema

Anderson et al. [95] described MR imaging findings of a subchondral fracture-like line adjacent to endplates, surrounded by bone marrow edema, and associated with vertebra plana or partial vertebral body collapse in three young patients. This line is thought to represent microtrabecular impaction secondary to insufficiency fracture [95], but there are no data on the specificity of this finding for a diagnosis of SAPHO [73].

ossifications to be enthesophytes rather than true syndesmophytes. In chronic cases, anterior bony bridging secondary to florid hyperostosis can be seen across the disco-vertebral junction at single or multiple levels [48, 50]. This evolves with the years toward ankylosis and associated kyphosis [52]. Once ankylosis is complete, hyperostosis has been noted to disappear [42]. These changes can be seen on MR imaging, but are best demonstrated on MDCT with reformatted images (Fig. 9). Developm ent of usually sym metrical paravertebral fatty masses have also been described in chronic cases [50, 66].

Osteosclerosis Pure osteosclerosis of one or more vertebral bodies may be encountered in patients with SAPHO [52, 67, 75]. Vertebral body sclerosis is usually seen adjacent to endplate or cortical erosive changes, and it may progress to produce diffuse and generalized sclerosis, similar to “an ivory vertebra,” with development of hyperostosis in more chronic cases (Fig. 9) [15]. The meaning of this finding is not yet fully elucidated; however, it is probably due to reactive inflammatory changes triggered by the endplate or cortical erosion [75]. The appearance may resemble blastic metastasis or Paget’s disease on radiography. These changes are best appreciated on radiography and CT, although associated bone marrow edema may still be demonstrated on MR imaging (Fig. 9). Paravertebral ossifications Paravertebral ossifications in SAPHO syndrome are marginal syndesmophytes or more commonly nonmarginal and asymmetrical syndesmophytes similar to, but not identical with, the syndesmophytes seen in psoriasis [42, 50, 57, 61]. Sugimoto et al. [57] found that the radiographic appearance of such ossifications may initially simulate syndesmophytes; however, their progression was found to be indicative of new periosteal ossification. Other authors [42] never observed true syndesmophytes in their series and considered these

Fig. 9 Multiple contiguous SAPHO discovertebral involvement. a Sagittal T2 fat-suppressed MR image and b corresponding MDCT reformatted image of the thoracic spine showing bone marrow edema, cortical erosions, and vertebral body hyperostosis with bony bridging. Note the erosive process extending along the anterior body cortices (arrows in b)


Sacroiliitis Patients with SAPHO are at increased risk of developing spondyloarthropathy with unilateral sacroiliitis; therefore, close attention should be paid to the sacroiliac joints. Sacroiliitis is seen in up to 50 % of patients and is characterized by erosions, extensive sclerosis, and hyperostosis primarily along the iliac side of the joint [42, 50] (Fig. 10). The association between a sacroiliitis and extensive sclerosis of the adjacent iliac bone is highly suggestive of the SAPHO spectrum and helps to differentiate it from other spondyloarthropathies [15, 45, 50]. These lesions may be subtle on radiography; however, CT clearly demonstrates joint erosions, joint space widening, surrounding osteosclerosis, and hyperostosis (Fig. 10). MR imaging shows edema in active lesions (Fig. 8a, b) and associated soft-tissue inflammation.

Appendicular skeleton

75 % of all lesions in the series of Mandell et al. [53], followed by the diaphysis, diaphyseal-metaphyseal region, and metaphyseal-epiphyseal region. Multifocality with symmetrical predominance of lower extremity lesions is almost always present at presentation or later in the course of the disease, but monostotic involvement has been described [53, 60] (Figs. 11 and 12). Changes in the appendicular skeleton include osteolysis, osteosclerosis, and periosteal new bone formation. At the beginning of the disease, the typical finding is a round or column-shaped osteolytic metaphyseal process mimicking neoplasm. However, this becomes surrounded by a thin, sclerotic rim within 1–2 weeks. With time, progressive sclerosis is seen around the lytic lesion producing a mixed lytic/sclerotic picture on radiography. As the inflammatory process extends into the cortex, periosteal new bone formation with enlargement of the bone develops, so that the radiographic appearance of chronic lesions may be predominantly sclerotic with associated hyperostosis [15, 50, 61, 73] (Figs. 11 and 12). Except for the sequestra and abscesses, which are characteristically absent in SAPHO, these features are those of a chronic osteomyelitis. The extent of the periosteal reaction

The appendicular skeleton is not frequently involved in adult patients. In contrast, long bones represent the most common site of disease in younger patients (CRMO) where the distal and proximal tibia are most frequently affected, followed by the proximal and distal femur (Fig. 8e) [53, 96]. The fibula, small tubular bones of the feet, humerus, radius, and ulna can also be involved [53] (Figs. 11 and 12). The metaphysis adjacent to the growth plate or metaphyseal equivalents are the most common locations, accounting for approximately

Fig. 10 Right sacroiliitis. a,b Axial MDCT images of the sacroiliac joints showing marginal erosions (arrows in a), sclerosis at the iliac side of the right joint (arrow in b), and mild changes on the left sacroiliac joint (small arrow in b)

Fig. 11 A 16-year-old boy presenting with pain and swelling of the right elbow. a Anteroposterior radiograph and b the corresponding image obtained 14 months later show the progression from an early pathological process with lysis and periosteal reaction (arrow in a) to sclerosis and hyperostosis (arrow in b). c Axial post-gadolinium spin-echo image at presentation showing diffuse enhancement of edematous muscle (short arrow) and periosteal reaction of the radius (long arrow). d Axial CT scan shows the periosteal reaction, which has become a hyperostosis (arrow)


Pathological features

Fig. 12 Chronic osteitis. a Lateral radiograph of the left elbow and b axial chest CT image show ulnar deformity due to sclerosis and hyperostosis (arrow in a) along with sternocostal hyerostosis and fusion (arrows in b)

depends on both the duration of disease and the size of the involved bone, generally being more pronounced in smalldiameter bones such as the fibula and metacarpals-metatarsals [73]. Manson et al. [97] suggested that small-diameter bones demonstrate extensive periosteal reaction, often associated with soft-tissue inflammation, owing to earlier extension of the inflammatory process from the medullary space into the cortex. MR imaging is useful to assess the extent and activity of the lesions. In active lesions, MR imaging shows typical findings of marrow edema, which appears hypointense on T1-weighted images and hyperintense on fluid-sensitive images (Fig. 8e). Frequently, associated periostitis, soft tissue inflammation, and joint effusions adjacent to the bone lesions can also be demonstrated on MR imaging.

Flat bones The ilium and mandible may be involved in up to 11 % of cases [15, 43]. The ilium is usually involved in association with adjacent sacroiliitis (Fig. 8). The disorder originally called diffuse sclerosing osteomyelitis of the mandible is now recognized to represent a form of SAPHO syndrome affecting the posterior body and ramus [42, 61, 73, 98]. Mandibular lesions, which can be isolated or accompanied by disease at other sites in the body, rarely spread to the temporomandibular joint where ankylosis may result [99, 100]. In the early stages, mandibular lesions are characterized by an osteolytic process with associated variable amounts of sclerosis on radiography. MR imaging often demonstrates an associated extensive organized periosteal reaction, and painful inflammation of the overlying soft tissue as well, without evidence of abscess formation [61, 73]. With time, increasing sclerosis, hyperostosis, and progressive enlargement of the mandible develop [73].

The histological features of the bone lesions are nonspecific and change over the course of the disease. In the initial acute phase, there is neutrophil-predominant inflammation with edema and both osteoclastic bone resorption and reactive bone formation indistinguishable from ordinary bacterial osteomyelitis [101–103]. Subsequently, lesions show lymphocyte-predominant inflammation [101]. In the late phase, there is only mild chronic inflammation with markedly sclerotic bone trabeculae and prominent marrow fibrosis [101–103]. Biopsy specimens may show acute, subacute, and chronic changes mixed in the same lesion [103].

Diagnosis The definite diagnosis of SAPHO can be hard to establish. This is mainly due to the variable clinical presentation and the broad spectrum of osteoarticular manifestations covering several diseases. Diagnosis is based on a combination of clinical and radiological findings. Inclusion and exclusion criteria for the diagnosis have been proposed by Benhamou et al. [58]. The presence of any of the following inclusion criteria establishes the diagnosis of SAPHO: (1) osteoarticular manifestations of acne conglobata, acne fulminans, or hidradenitis suppurativa, (2) osteoarticular manifestations of PPP, (3) hyperostosis involving the anterior chest wall, spine, or limbs with or without dermatosis, and (4) CRMO with or without dermatosis. However, there are no validated diagnostic criteria designed specifically for SAPHO syndrome. The diagnosis is not difficult when typical skeletal lesions are characteristically located and associated with PPP or acne; it is much more difficult if the sites of involvement or radiological findings are atypical, especially if skin manifestations are absent. The diagnosis becomes impossible in patients with a solitary symptomatic skeletal lesion in the absence of both skin manifestations and hyperostosis because of a wide radiological differential diagnoses including infection, bone tumors, lymphoma, metastases, eosinophilic granuloma, fibrous dysplasia, and Paget’s disease [14, 46, 61]. In such circumstances, the radiologist needs to consider SAPHO by enquiring about a history of skin disorders, obtaining previous imaging of other skeletal lesions and recommending whole-body MR imaging or nuclear medicine techniques to reveal other subclinical sites of osteoarticular involvement. Detection of multiple lesions, which may include the joints, axial and appendicular skeleton, along with no history of inflammatory arthropathy or primary cancer can be helpful for the diagnosis, and the scintigraphic “bull’s head” pattern at the sternoclavicular region is a


highly specific manifestation of this condition [64]. In the other cases, biopsy may be needed. It is important, however, to appreciate that the diagnosis can never be done by histological results alone, and the benefit of biopsy is to exclude other diagnoses [15, 46, 103].

Clinical course The course of the disease is unpredictable with remissions and relapses. A series of 71 patients has been described by Colina et al. [42]. Thirty-seven patients (52 %) had a chronic course characterized by fluctuating intermittent periods of exacerbation and short improvement, requiring almost continuous treatment. Twenty-five patients (35 %) reported multiple remission and exacerbations of the disease after a first, limited course of less than 6 months. Only nine patients (13 %) had a limited single-phase disease course, lasting less than 6 months, which afterward faded away. Female gender, peripheral arthritis, anterior chest wall involvement, the coexistence of skin lesions, and high inflammatory markers at onset were correlated with a chronic disease course. Involvement of additional sites during the disease course is a frequent observation [42, 43, 58, 61]. In the long-term SAPHO is not particularly debilitating as its rheumatic manifestations show minimal and slow progression [15, 43, 45]. Nonetheless, the peripheral arthritis may become erosive in some patients [42, 43, 61].

Treatment A number of different therapies have been reported to be useful in patients with SAPHO. However, there have been no randomized, controlled trials studying the effectiveness of the various therapeutic agents. The treatment is mainly symptomatic with nonsteroidal anti-inflammatory drugs (NSAIDs), analgesics, and corticosteroids as the first-line therapeutic agents. However, they often fail to control the disease, and second-line therapies, mainly antibiotics, bisphosphonates, and disease-modifying antirheumatic drugs (DMARDs), are required [42, 43]. The use of antibiotics, instituted based on the isolation of P. acnes in bone biopsies from several subjects with SAPHO, has not shown convincing results, although some patients have been responsive to this therapy [42, 104–106]. Assmann et al. [17] found that a 16-week course of antimicrobial therapy was effective in controlling disease activity of both the osteitis and the dermatologic lesions for the period of drug administration. After discontinuation of the antibiotics, however, this effect was nullified. Over the last decade, intravenous bisphosphonates, including pamidronate and zolendronic acid, have been found effective to promote long-term remission in a considerable portion of patients

refractory to NSAIDs and corticosteroids [12, 19, 61, 107, 108]. However, a number of treatment failures [108, 109], drug-induced osteopetrosis in a child [110], and renal toxicity [111] have also been reported. Bisphosphonates act by suppressing osteoclastic bone resorption and turnover [112] with an added anti-inflammatory effect that suppresses the production of interleukin 1β, interleukin 6, and TNF-α [108]. The use of DMARDs, particularly methotrexate, sulfasalazine, or azathioprine, has been reported with contradictory results [34, 43]. Conversely, treatment with biologicals, TNF-α antagonists (infliximab, adalimumab, and etanercept), has recently shown promising results and should be considered in the therapeutic strategy of refractory cases [30, 32, 42, 113–116] (Fig. 6). Biological therapy is supported by the efficacy of TNF-α blocking therapy in spondyloarthopathies [117] and the description of high concentrations of TNF-α in bone biopsy specimens of patients with SAPHO [116]. The response is generally rapid, with remission of bone pain in the majority of patients, although it is not clear whether this treatment is permanently effective. Infliximab induces remission of skin lesions in some patients but cause exacerbation in others [118]. Adalimumab, a human monoclonal antibody that specifically binds TNF-α, has been suggested as a possible alternative therapy in such situations [113]. However, also in TNF-α blockade no double-blind randomized placebo-controlled trials have been performed to confirm these observations. In 1999, two groups simultaneously [119, 120] described an autosomal recessive disorder—deficiency of IL-1-receptor antagonist (DIRA)—that presents in the neonatal period with pustulosis, sterile multifocal osteomyelitis, and periostitis. It is a newly recognized autoinflammatory disorder that looks similar to SAPHO [121, 122]. Institution of empirical treatment with the recombinant interleukin-1receptor antagonist anakinra resulted in a good shortterm outcome and prompted investigators to evaluate the effects of anakinra in SAPHO syndrome [121–124]. However, the long-term outcome of anakinra-treated SAPHO has yet to be established.

Conclusion Clinical presentation of SAPHO syndrome is heterogeneous, resulting in diagnostic difficulties. Diagnosis requires a team approach among the rheumatologist, dermatologist, orthopedic surgeon, radiologist, and pathologist. The general radiologist needs to be familiar with the imaging findings because they may be the first to suggest the correct diagnosis. In such circumstances, cases of suspected SAPHO should be sent to dedicated musculoskeletal radiologists ideally working in tertiary referral centers where there is a multidisciplinary team approach. This approach ensures the correct diagnosis,


avoiding unnecessary biopsy, and aids in establishing the true prevalence of SAPHO.

Conflict of interest The authors declare that they have no conflict of interest.

References 1. Windom RE, Sandford JP, Ziff M. Acne conglobata and arthritis. Arthritis Rheum. 1961;4:632–5. 2. Kato T, Kambara H, Hoshi E. Case of bilateral clavicular osteomyelitis with palmar and plantar pustulosis. Seikei Geka. 1968;19: 590–3. 3. Giedion A, Holthusen W, Masel LF, et al. Subacute and chronic “symmetrical” osteomyelitis. Ann Radiol (Paris). 1972;15:329–42. 4. Köhler H, Uehlinger E, Kutzner J, et al. Sterno-costo-clavicular hyperostosis: a hitherto undescribed entity (author’s translation). Dtsch Med Wochenschr. 1975;100:1519–23. 5. Björkstén B, Gustavson KH, Eriksson B, et al. Chronic recurrent multifocal osteomyelitis and pustulosis palmoplantaris. J Pediatr. 1978;93:227–31. 6. Bergdahl K, Björkstén B, Gustavson KH, et al. Pustulosis palmoplantaris and its relation to chronic recurrent multifocal osteomyelitis. Dermatologica. 1979;159:37–45. 7. Sonozaki H, Azuma A, Okai K, et al. Clinical features of 22 cases with “inter-sterno-costo-clavicular ossification”. A new rheumatic syndrome. Arch Orthop Trauma Surg. 1979;95:13–22. 8. Sonozaki H, Mitsui H, Miyanaga Y, et al. Clinical features of 53 cases with pustulotic arthro-osteitis. Ann Rheum Dis. 1981;40:547–53. 9. Dihlmann W, Dihlmann SW. Acquired hyperostosis syndrome: spectrum of manifestations at the sternocostoclavicular region. Radiologic evaluation of 34 cases. Clin Rheumatol. 1991;10:250–63. 10. Lindor NM, Arsenault TM, Solomon H, Seidman CE, McEvoy MT. A new autosomal dominant disorder of pyogenic sterile arthritis, pyoderma gangrenosum, and acne: PAPA Syndrome. Mayo Clin Proc. 1997;72:611–5. 11. Tlougan BE, Podjase JO, O’Haver J, et al. Chronic recurrent multifocal osteomyelitis (CRMO) and synovitis, acne, pustulosis, hyperostosis, and osteitis (SAPHO) syndrome with associated neutrophilic dermatoses: a report of seven cases and review of the literature. Pediatr Dermatol. 2009;26(5):497–505. 12. Siau K, Laversuch CJ. SAPHO syndrome in an adult with ulcerative colitis responsive to intravenous pamidronate: a case report and review of the literature. Rheumatol Int. 2010;30:1085–8. 13. Chamot AM, Benhamou CL, Kahn MF, et al. Acne–pustulosis– hyperostosis–osteitis syndrome. Results of a national survey. 85 cases. Rev Rhum Mal Osteoartic. 1987;54:187–96. 14. Boutin RD, Resnick D. The SAPHO syndrome: an evolving concept for unifying several idiopathic disorders of bone and skin. AJR Am J Roentgenol. 1998;170(3):585–91. 15. Depasquale R, Kumer N, Lalam RK, et al. SAPHO: what radiologists should know. Clin Radiol. 2012;67:195–206. 16. Assmann G, Simon P. The SAPHO syndrome—are microbes involved? Best Pract Res Clin Rheumatol. 2011;25(3):423–37. 17. Assmann G, Kueck O, Kirchhoff T, et al. Efficacy of antibiotic therapy for SAPHO syndrome is lost after its discontinuation: an interventional study. Arthritis Res Ther. 2009;11:R140. 18. Govoni M, Colina M, Massara A, Trotta F. SAPHO syndrome and infections. Autoimmun Rev. 2009;8:256–9.

19. Colina M, La Corte R, Trotta F. Sustained remission of SAPHO syndrome with pamidronate: a follow-up of fourteen cases and review of the literature. Clin Exp Rheumatol. 2009;27:112–5. 20. Edlund E, Johnsson U, Lidgren L, et al. Palmoplantar pustulosis and sternocostoclavicular arthro-osteitis. Ann Rheum Dis. 1988;47:809–15. 21. Trimble BS, Evers CJ, Ballaron SA, Young JM. Intraarticular injection of Propionibacterium acnes causes an erosive arthritis in rats. Agents Actions. 1987;21:281–3. 22. Colina M, Lo Monaco A, Khodeir M, Trotta F. Propionibacterium acnes and SAPHO syndrome: a case report and literature review. Clin Exp Rheumatol. 2007;25:457–60. 23. Hayem G. Valuable lessons from SAPHO syndrome. Joint Bone Spine. 2007;74:123–6. 24. Amital H, Govoni M, Maya R, et al. Role of infectious agents in systemic rheumatic diseases. Clin Exp Rheumatol. 2008;26(1 Suppl 48):S27–32. 25. Rozin AP. From molecular mimicry to cross-reactivity or pathogen expansion? A hypothesis. Clin Rheumatol. 2007;26:285–8. 26. Rozin AP. SAPHO syndrome: is a range of pathogenassociated rheumatic diseases extended? Arthritis Res Ther. 2009;11(6):131. 27. Barton GM, Medzhitov R. Toll-like receptor signaling pathways. Science. 2003;300(5625):1524–5. 28. Kalis C, Gumenscheimer M, Freudenberg N, et al. Requirement of TLR9 in the immunomodulatory activity of P. acnes. J Immunol. 2005;174:4295–300. 29. Hurtado-Nedelec M, Chollet-Martin S, Chapeton D, Hugot JP, Hayem G, Gérard B. Genetic susceptibility factors in a cohort of 38 patients with SAPHO syndrome: a study of PSTPIP2, NOD2, and LPIN2 genes. J Rheumatol. 2010;37:401–9. 30. Wagner AD, Andresen J, Jendro MC, Hulsemann JL, Zeidler H. Sustained response to tumor necrosis factor alpha-blocking agents in two patients with SAPHO syndrome. Arthritis Rheum. 2002;46: 1965–8. 31. Carpenter E, Jackson MA, Friesen CA, Scarbrough M, Roberts CC. Crohn’s-associated chronic recurrent multifocal osteomyelitis responsive to infliximab. J Pediatr. 2004;144:541–4. 32. Olivieri I, Padula A, Ciancio G, Salvarani C, Niccoli L, Cantini F. Successful treatment of SAPHO syndrome with infliximab: report of two cases. Ann Rheum Dis. 2002;61:375–6. 33. Jappe U, Boit R, Farrar MD, Ingham E, Sandoe J, Holland KT. Evidence for diversity with Propionibacterium acnes: a comparison of the T cell stimulatory activity of isolates from inflammatory acne, endocarditis and the laboratory. J Eur Acad Dermatol Venereol. 2004;18:450–4. 34. Jansson A, Renner ED, Ramser J, et al. Classification of nonbacterial osteitis: retrospective study of clinical, immunological and genetic aspects in 89 patients. Rheumatology (Oxford). 2007;46:154–60. 35. Ferguson PJ, Chen S, Tayeh MK, et al. Homozygous mutations in LPIN2 are responsible for the syndrome of chronic recurrent multifocal osteomyelitis and congenital dyserythropoietic anaemia (Majeed syndrome). J Med Genet. 2005;42:551–7. 36. El-Shanti HI, Ferguson PJ. Chronic recurrent multifocal osteomyelitis: a concise review and genetic update. Clin Orthop Relat Res. 2007;462:11–9. 37. Yeon HB, Lindor NM, Seidman JG, Seidman CE. Pyogenic arthritis, pyoderma gangrenosum, and acne syndrome maps to chromosome 15q. Am J Hum Genet. 2000;66:1443–8. 38. Li J, Nishizawa K, An W, Hussey RE, Lialios FE, Salgia R, et al. A cdc15-like adaptor protein (CD2BP1) interacts with the CD2 cytoplasmic domain and regulates CD2-triggered adhesion. EMBO J. 1998;17:7320–36.

Author's personal copy Skeletal Radiol 39. Baum W, Kirkin V, Mateus Fernandez SB, et al. Binding of the intracellular Fas ligand (FasL) domain to the adaptor protein PSTPIP results in a cytoplasmic localization of FasL. J Biol Chem. 2005;280:4012–24. 40. Ferguson PJ, Bing X, Vasef MA, et al. A missense mutation in pstpip2 is associated with the murine autoinflammatory disorder chronic multifocal osteomyelitis. Bone. 2006;38:41–7. 41. Byrd L, Grossmann M, Potter M, Shen-Ong G. Chronic multifocal osteomyelitis, a new recessive mutation on chromosome 18 of the mouse. Genomics. 1991;11:794–8. 42. Colina M, Govoni M, Orzincolo C, Trotta F. Clinical and radiologic evolution of synovitis, acne, pustulosis, hyperostosis, and osteitis syndrome: a single center study of a cohort of 71 subjects. Arthritis Rheum. 2009;61:813–21. 43. Hayem G, Bouchaud-Chabot A, Benali K, et al. SAPHO syndrome: a long-term follow-up study of 120 cases. Semin Arthritis Rheum. 1999;29:159–71. 44. Mann B, Shaerf DA, Sheeraz A, et al. SAPHO syndrome presenting as widespread bony metastatic disease of unknown origin. Rheumatol Int. 2012;32(2):505–7. 45. Sallés M, Olivé A, Perez-Andres R, et al. The SAPHO syndrome: a clinical and imaging study. Clin Rheumatol. 2011;30:245–9. 46. Kahn MF, Khan MA. The SAPHO syndrome. Baillieres Clin Rheumatol. 1994;8:333–62. Semin Arthritis Rheum 1999;293:159–171. 47. Takigawa T, Tanaka M, Nakahara S, et al. SAPHO syndrome with rapidly progressing destructive spondylitis: two cases treated surgically. Eur Spine J. 2008;17 Suppl 2:S331–7. 48. Magrey M, Khan MA. New insights into synovitis, acne, pustulosis, hyperostosis, and osteitis (SAPHO) syndrome. Curr Rheumatol Rep. 2009;11:329–33. 49. Van Doornum S, Barraclough D, McColl G, Wicks I. SAPHO: rare or just not recognized? Semin Arthritis Rheum. 2000;30:70–7. 50. Earwaker JWS, Cotten A. SAPHO: syndrome or concept? Imaging findings. Skelet Radiol. 2003;32:311–27. 51. Karadag-Saygi E, Gunduz OH, Gumrukcu G, et al. SAPHO syndrome: misdiagnosed and operated. Acta Reumatol Port. 2008;33: 460–3. 52. Court C, Charlez C, Molina V, et al. Isolated thoracic spine lesion: is this the presentation of a SAPHO syndrome? A case report. Eur Spine J. 2005;14:711–5. 53. Mandell GA, Contreras SJ, Conard K, Harcke HT, Maas KW. Bone scintigraphy in the detection of chronic recurrent multifocal osteomyelitis. J Nucl Med. 1998;39:1778–83. 54. Khanna L, El-khoury GY. SAPHO syndrome—a pictorial assay. Iowa Orthop J. 2012;32:189–95. 55. Callen JP. Neutrophilic dermatoses. Dermatol Clin. 2002;20:409–19. 56. Cotten A, Flipo RM, Mentre A, et al. SAPHO syndrome. Radiographics. 1995;15:1147–54. 57. Sugimoto H, Tamura K, Fujii T. The SAPHO syndrome: defining the radiological spectrum of disease comprising the syndrome. Eur Radiol. 1998;8:800–6. 58. Benhamou CL, Chamot AM, Kahn MF. Synovitis-acne-pustulosishyperostosis-osteomyelitis syndrome (SAPHO). A new syndrome among the spondyloarthropathies? Clin Exp Rheumatol. 1988;6: 109–12. 59. Maugars Y, Berthelot JM, Ducloux JM, et al. SAPHO syndrome: a follow-up study of 19 cases with special emphasis on enthesis involvement. J Rheumatol. 1995;22:2135–41. 60. Fritz J, Tzaribatchev N, Claussen CD, et al. Chronic recurrent multifocal osteomyelitis: comparison of whole-body MR imaging with radiography and correlation with clinical and laboratory data. Radiology. 2009;252:842–51. 61. Nguyen MT, Borchers A, Selmi C, Naguwa SM, Cheema G, Gershwin ME. The SAPHO syndrome. Semin Arthritis Rheum. 2012;42:254–65.

62. Guérin-Pfyffer S, Guillaume-Czitrom S, Tammam S, Koné-Paut I. Evaluation of chronic recurrent multifocal osteitis in children by whole-body magnetic resonance imaging. Joint Bone Spine. 2012;79(6):616–20. 63. Weckbach S. Whole-body MRI for inflammatory arthritis and other multifocal rheumatoid diseases. Semin Musculoskelet Radiol. 2012;16(5):377–88. 64. Freyschmidt J, Sternberg A. The bullhead sign: scintigraphic pattern of sternocostoclavicular hyperostosis and pustulotic arthroosteitis. Eur Radiol. 1998;8:807–12. 65. Pichler R, Weiglein K, Schmekal B, Sfetsos K, Maschek W. Bone scintigraphy using Tc-99m DPD and F18-FDG in a patient with SAPHO syndrome. Scand J Rheumatol. 2003;32:58–60. 66. Inoue K, Yamaguchi T, Ozawa H, et al. Diagnosing active inflammation in the SAPHO syndrome using 18FDG-PET/CT in suspected metastatic vertebral bone tumors. Ann Nucl Med. 2007;21(8):477–80. 67. Takeuchi K, Matsusita M, Takagishi K. A case of SAPHO (synovitis-acne-pustulosis-hyperostosis-osteomyelitis) syndrome in which [18F]fluorodeoxyglucose positron emission tomography was useful for differentiating from multiple metastatic bone tumors. Mod Rheumatol. 2007;17(1):67–71. 68. Patel CN, Smith JT, Rankine JJ, et al. F-18 FDG PET/CT can help differentiate SAPHO syndrome from suspected metastatic bone disease. Clin Nucl Med. 2009;34:254–7. 69. Canbaz F, Gonullu G, Baris S, Selcuk MB, Bicakci N. SAPHO syndrome without dermatologic manifestations: multifocal uptake mismatch on 99mTc-MDP and 18F-FDG-PET/CT imaging. Hell J Nucl Med. 2010;13(1):73–5. 70. Adams MC, Turkington TG, Wilson JM, Wong TZ. A systematic review of the factors affecting accuracy of SUV measurements. AJR Am J Roentgenol. 2010;195(2):310–20. 71. van Holsbeeck M, Martel W, Dequeker J, et al. Soft tissue involvement, mediastinal pseudotumor, and venous thrombosis in pustulotic arthroosteitis. A study of eight new cases. Skelet Radiol. 1989;18:1–8. 72. Girschick HJ, Krauspe R, Tschammler A, Huppertz HI. Chronic recurrent osteomyelitis with clavicular involvement in children: diagnostic value of different imaging techniques and therapy with non-steroidal anti-inflammatory drugs. Eur J Pediatr. 1998;157: 28–33. 73. Khanna G, Sato TS, Ferguson P. Imaging of chronic recurrent multifocal osteomyelitis. Radiographics. 2009;29(4):1159–77. 74. Toussirot E, Dupond JL, Wendling D. Spondylodiscitis in SAPHO syndrome. A series of eight cases. Ann Rheum Dis. 1997;56:52–8. 75. Laredo JD, Vuillemin-Bodaghi V, Boutry N, Cotton A, Parlier-Cuau C. SAPHO syndrome: MR appearance of vertebral involvement. Radiology. 2007;242(3):825–31. 76. Romanus R, Yden S. Destructive and ossifying spondylitic changes in rheumatoid ankylosing spondylitis (pelvospondylitis ossificans). Acta Orthop Scand. 1952;22:88–99. 77. Hermann KG, Althoff CE, Schneider U, et al. Spinal changes in patients with spondyloarthritis: comparison of MR imaging and radiographic appearances. RadioGraphics. 2005;253:559–69. 78. Jevtic V, Kos-Golja M, Rozman B, McCall I. Marginal erosive discovertebral “Romanus” lesions in ankylosing spondylitis demonstrated by contrast enhanced Gd-DTPA magnetic resonance imaging. Skelet Radiol. 2000;29:27–33. 79. Bennet AN, Rehman A, Hensor EM, Marzo-Ortega H, Emery P, McGonagle D. The fatty Romanus lesion: a non-inflammatory spinal MRI lesion specific for axial spondyloarthropathy. Ann Rheum Dis. 2010;69(5):891–4. 80. Nachtigal A, Cardinal E, Bureau NJ, Sainte-Marie LG, Milette F. Vertebral involvement in SAPHO syndrome: MRI findings. Skelet Radiol. 1999;28:163–8.

Author's personal copy Skeletal Radiol 81. Remedios D, Natali C, Saifuddin A. Case report: MRI of vertebral osteitis in early ankylosing spondylitis. Clin Radiol. 1998;53:534–6. 82. Jevtic V, Rozman B, Kos-Golja M, Watt I. MR imaging in seronegative spondyloarthritis. Radiologe. 1996;36:624–31. 83. Marc V, Dromer C, Le Guennec P, Manelfe C, Fournie B. Magnetic resonance imaging and axial involvement in spondylarthropathies: delineation of the spinal entheses. Rev Rhum (Engl Ed). 1997;64: 465–73. 84. Lacout A, Rousselin B, Pelage JP. CT and MRI of spine and sacroiliac involvement in spondyloarthropathy. Am J Roentgenol. 2008;191:1016–23. 85. Kotilainen P, Gullichsen RE, Saario R, Manner I, Kotilainen E. Aseptic spondylitis as the initial manifestation of the SAPHO syndrome. Eur Spine J. 1997;6(5):327–9. 86. Sweeney SA, Kumar VA, Tayar J, et al. Case 181: synovitis acne pustulosis hyperostosis osteitis (SAPHO) syndrome. Radiology. 2012;263(2):613–7. 87. Nikolaisen C, Nossent H. Early histology in ankylosing spondylitis related spondylodiscitis supports its inflammatory origin. Scand J Rheumatol. 2005;34(5):396–8. 88. Tohme-Noun C, Feydy A, Belmatoug N, Krainik A, Fantin B, Menu Y. Cervical involvement in SAPHO syndrome: imaging findings with a 10-year follow-up. Skelet Radiol. 2003;32:103–6. 89. Andersson O. Röntgenbilden vid spondylarthritis ankylopoetica. Nord Med Tidskr. 1937;14:2000–2. 90. Langlois S, Cedoz JP, Lohse A, Toussirot E, Wendling D. Aseptic discitis in patients with ankylosing spondylitis: a retrospective study of 14 cases. Joint Bone Spine. 2005;72(3):248–53. 91. Bron JL, de Vries MK, Snieders MN, van der HorstBruinsma IE, van Royen BJ. Discovertebral (Andersson) lesions of the spine in ankylosing spondylitis revisited. Clin Rheumatol. 2009;28(8):883–92. 92. Yu L, Kasser JR, O’Rourke E, Kozakewich H. Chronic recurrent osteomyelitis. Association with vertebra plana. J Bone Joint Surg Am. 1989;71:105–12. 93. Jurik AG, Egund N. MRI in chronic recurrent multifocal osteomyelitis. Skelet Radiol. 1997;26:230–8. 94. Demharter J, Bohndorf K, Michl W, Vogt H. Chronic recurrent multifocal osteomyelitis: a radiological and clinical investigation of five cases. Skelet Radiol. 1997;26:579–88. 95. Anderson SE, Heini P, Sauvain MJ. Imaging of chronic recurrent multifocalosteomyelitis of childhood first presenting with isolated primary spinal involvement. Skelet Radiol. 2003;32(6):328–36. 96. Huber AM, Lam PY, Duffy CM, et al. Chronic recurrent multifocal osteomyelitis: clinical outcomes after more than 5 years of followup. J Pediatr. 2002;141:198–203. 97. Manson D, Wilmot DM, King S, Laxer RM. Physeal involvement in chronic recurrent multifocal osteomyelitis. Pediatr Radiol. 1989;20:76–9. 98. Beretta-Piccoli BC, Sauvain MJ, Gal I, et al. Synovitis, acne, pustulosis, hyperostosis, osteitis (SAPHO) syndrome in childhood: a report of ten cases and review of the literature. Eur J Pediatr. 2000;159:594–601. 99. Kodama Y, Tanaka R, Kurokawa A. Severe destruction of the temporomandibular joint with complete resorption of the condyle associated with synovitis, acne, pustulosis, hyperostosis, and osteitis syndrome. Oral Surg Oral Med Oral Pathol Oral Radiol. 2013;116(2):e128–33. 100. Müller-Richter UD, Roldán JC, Mörtl M, Behr M, Reichert TE, Driemel O. SAPHO syndrome with ankylosis of the temporomandibular joint. Int J Oral Maxillofac Surg. 2009;38(12):1335–41. 101. Björkstén B, Boquist L. Histopathological aspects of chronic recurrent multifocal osteomyelitis. J Bone Joint Surg (Br). 1980;62:376–80.

102. Reith JD, Bauer TW, Schils JP. Osseous manifestations of SAPHO (synovitis, acne, pustulosis, hyperostosis, osteitis) syndrome. Am J Surg Pathol. 1996;20:1368–77. 103. Chow LT, Griffith JF, Kumta SM, Leung PC. Chronic recurrent multifocal osteomyelitis: a great clinical and radiologic mimic in need of recognition by the pathologist. APMIS. 1999;107:369–79. 104. Rozin AP, Nahir AM. Is SAPHO syndrome a target for antibiotic therapy? Clin Rheumatol. 2007;26:817–20. 105. Ballara SC, Siraj QH, Maini RN, Venables PJ. Sustained response to doxycycline therapy in two patients with SAPHO syndrome. Arthritis Rheum. 1999;42:819–21. 106. Schilling F, Wagner AD. Azithromycin: an anti-inflammatory effect in chronic recurrent multifocal osteomyelitis? A preliminary report. Z Rheumatol. 2000;59:352–3. 107. Just A, Adams S, Brinkmeier T, et al. Successful treatment of primary chronic osteomyelitis in SAPHO syndrome with bisphosphonates. Dtsch Dermatol Ges. 2008;6(8):657–60. 108. Amital H, Applbaum YH, Aamar S, Daniel N, Rubinow A. SAPHO syndrome treated with pamidronate: an open-label study of 10 patients. Rheumatology (Oxford). 2004;43(5):658–61. 109. Solau-Gervais E, Soubrier M, Gerot I, et al. The usefulness of bone remodelling markers in predicting the efficacy of pamidronate treatment in SAPHO syndrome. Rheumatology (Oxford). 2006;45:339–42. 110. Whyte MP, Wenkert D, Clements KL, McAlister WH, Mumm S. Bisphosphonate-induced osteopetrosis. N Engl J Med. 2003;349: 457–63. 111. Chang JT, Green L, Beitz J. Renal failure with the use of zoledronic acid. N Engl J Med. 2003;349:1676–9. 112. Rogers MJ, Gordon S, Benford HL, et al. Cellular and molecular mechanisms of action of bisphosphonates. Cancer. 2000;88(12 Suppl):2961–78. 113. Arias-Santiago S, Sanchez-Cano D, Callejas-Rubio JL, et al. Adalimumab treatment for SAPHO syndrome. Acta Derm Venereol. 2010;90:301–2. 114. Deutschmann A, Mache CJ, Bodo K, et al. Successful treatment of chronic recurrent multifocal osteomyelitis with tumor necrosis factor-alpha blockage. Pediatrics. 2005;116:1231–3. 115. Ben Abdelghani K, Dran DG, Gottenberg JE, et al. Tumor necrosis factor alpha blockers in SAPHO syndrome. J Rheumatol. 2010;37: 1699–704. 116. Moll C, Hernández MV, Cañete JD, et al. Ilium Osteitis as the main manifestation of the SAPHO syndrome: response to infliximab therapy and review of the literature. Semin Arthritis Rheum. 2008;37(5):299–306. 117. Braun J, Baraliakos X, Brandt J, Listing J, Zink A, Alten R, et al. Persistent clinical response to the anti-TNF-alpha antibody infliximab in patients with ankylosing spondylitis over 3 years. Rheumatology. 2005;44:670–6. 118. Wollina U, Hansel G, Koch A, Schönlebe J, Köstler E, Haroske G. Tumor necrosis factor-alpha inhibitor-induced psoriasis or psoriasiform exanthemata: first 120 cases from the literature including a series of six new patients. Am J Clin Dermatol. 2008;9:1–14. 119. Aksentijevich I, Masters SL, Ferguson PJ, et al. An autoinflammatory disease with deficiency of the interleukin-1receptor antagonist. N Engl J Med. 2009;360:2426–37. 120. Reddy S, Jia S, Geoffrey R, et al. An autoinflammatory disease due to homozygous deletion of the IL1RN locus. N Engl J Med. 2009;360:2438–44. 121. Burgermeister LT, Baeten DL, Tas SW. Biologics for rare inflammatory diseases: TNF blockade in the SAPHO syndrome. Neth J Med. 2012;70(10):444–9. 122. Ferguson PJ, Sandu M. Current understanding of the pathogenesis and management of chronic recurrent multifocal osteomyelitis. Curr Rheumatol Rep. 2012;14(2):130–41.

Author's personal copy Skeletal Radiol 123. Rech J, Manger B, Lang B, Schett G, Wilhelm M, Birkmann J. Adult-onset Still’s disease and chronic recurrent multifocal osteomyelitis: a hitherto undescribed manifestation of autoinflammation. Rheumatol Int. 2012;32(6):1827–9.

124. Colina M, Pizzirani C, Khodeir M, et al. Dysregulation of P2X7 receptor-inflammasome axis in SAPHO syndrome: successful treatment with anakinra. Rheumatology. 2010;49: 1416–8.