Skip to main content

Toxic synovitis

When a child develops hip pain or limping, it draws the immediate apprehension of parent and physician alike, despite the fact that the vast majority of these children will improve with little or no treatment. The concern is with the differential diagnosis because the symptoms of benign disorders of the hip may mimic serious disease and distinguishing between them can be quite difficult and often time dependent.

Toxic synovitis (TS), also known as acute transient synovitis of the hip, is a condition in which there is nonspecific inflammation and hypertrophy of the synovium with an effusion in the hip joint. It is the most common cause of acute hip pain in children aged 3–10 years, and up to 3% of children have an episode at some time during their life. It is extremely uncommon in adults [9]. Although the diagnosis is usually one of exclusion, the etiology is unclear; parenthetically, many of these children have had an upper respiratory illness shortly before the onset of hip pain. Nearly all children recover within 2 weeks and without long-lasting effects. Four to 17% of children may have a recurrent episode. Children with recurrent episodes usually have a benign course, although, in about 10%, recurrent TS may be the presenting feature of a chronic inflammatory condition [10]. Treatment for TS focuses primarily on rest and anti-inflammatory medications, which shorten the course of the disorder [2]. While the treatment is rather straightforward, arriving at the diagnosis may be difficult.

The differential diagnosis for a limping child with coxalgia is long, yet the immediate concern is ruling out septic arthritis of the hip, which is one of the few true orthopedic emergencies. With septic arthritis, destruction of the articular cartilage begins quickly and is secondary to proteolytic enzymes released from synovial cells. Interleukin-1 triggers the release of proteases from chondrocytes and synoviocytes in response to polymorphonuclear leukocytes and bacteria. Impairment of intracapsular vascular supply also plays a role in the articular destruction via elevation of the intracapsular pressure which leads to thrombosis and progressive displacement of femoral head from the acetabulum [7]. A delay in treatment can have devastating consequences, with near total destruction of the hip for which there are no good reconstructive options. Despite this knowledge, poor outcomes occur in 27% of all patients with septic arthritis and nearly 40% when the hip is involved [8]. Treatment for septic arthritis requires surgical drainage and appropriate treatment with antibiotics for 3 to 6 weeks.

Early in the disease course, the two can often mimic each other and both present in similar populations, often with similar histories. Differential between septic arthritis and TS is challenging as both may present in young children with pain, limp, and limited motion of the hip. The child may be carried in or limps and tends to sit or lie with the leg flexed and in external rotation; in this position, the hip capsule is more relaxed and thus able to accommodate the increased volume of the joint fluid which leads to distension and pain.

The differentiation between these two disorders is extremely important and can be difficult, although there may be some clinical clues that may help with the differential. Although there is certainly overlap in the symptoms of the two disorders, the severity and acuity of these symptoms are usually greater in patients with septic arthritis of the hip. TS may be undulating in its course, with periods of relative comfort intermixed with discomfort. Hip motion is limited in TS, but usually you can move the hip a little, and often the child will be able to bear weight, albeit with a limb in TS. Septic arthritis will not get better; usually, any hip motion is extremely painful and the child will not usually bear weight at all.

In an effort make this diagnosis more accurate, Kocher, et al. retrospectively reviewed the cases of children who were evaluated at a major tertiary-care children’s hospital because of an acutely irritable hip. Four independent multivariate clinical predictors were identified to differentiate between septic arthritis and transient synovitis: history of fever, non-weight-bearing, erythrocyte sedimentation rate of at least 40 mm/h, and serum white blood cell (WBC) count of more than 12,000 cells per cubic millimeter. The predicted probability of septic arthritis was summarized as less than 0.2% for zero predictors, 3.0% for one predictor, 40.0% for two predictors, 93.1% for three predictors, and 99.6% for four predictors [3]. Focusing on these guidelines has allowed general orthopedists and residents in training to zero in on important data points which guide an appropriate but not foolproof level of suspicion. For instance, when this paradigm was applied prospectively by the same group, the predicted probability decreased by about 10% and when evaluated by different investigators the probability decreased to 59% [4, 6].

Imaging may assist in the diagnosis. Radiographs may reveal subtle signs early in the disease process such as capsular distention, joint space widening, diminution of the definition of soft tissue planes around the hip joint, or slight demineralization of the bone of the proximal femur, particularly in the metaphyseal region. None of these finding are particularly helpful at differentiating TS from septic arthritis. The primary role of plain radiographs is ruling out other disorders and insuring that the metaphyseal bones of the femur and acetabulum do not have signs of other processes such as osteomyelitis which can present as either septic arthritis or reactive aseptic synovitis.

Technetium bone scan may show decreased uptake (“cold”) early in the disease process and increased uptake (“hot”) later due to hyperemic response. Gallium- and indium-labeled leukocyte scans may be helpful in the diagnosis of atypical cases but are difficult to perform and take 48 to 72 h to perform [1].

Radiologists usually become involved in the care of a child with suspected TS when the emergency physician or orthopedist requests an urgent hip ultrasound. Ultrasound is quick, painless, and imparts no ionizing radiation and can detect an effusion in nearly all of the established cases. The specificity of a hip joint effusion on ultrasound is low and unfortunately is not specific for septic arthritis [12], but the absence of an effusion makes septic arthritis unlikely except very early in the disease process. Ultrasound is also useful for guiding needle aspiration. If an effusion is seen, the child is typically sedated and a needle placed into the effusion under ultrasound guidance and fluid sent for cell count and culture–Gram stain.

Other imaging modalities have been proposed, primarily contrast-enhanced magnetic resonance imagine (MRI), to quickly and noninvasively differentiate between septic arthritis and TS. High signal on T2-weighted images in the femoral head or acetabulum is more common in septic arthritis than in TS. After injection of intravenous gadolinium, the femoral head usually has decreased signal intensity in the perfusion phase early after the injection, with increased signal intensity enhancement on delayed fat-suppressed T1-weighted images. It can be difficult in some settings to obtain an emergent MRI and immediate reading, however, and some patients in this age group will require sedation especially if the hip is painful [5, 11]. MRI is a very reliable method to assess the periarticular bone for osteomyelitis.

The treatment of septic arthritis is a true emergency that often involves the emergency department as well as radiology and orthopedics. Because these children can present at inopportune times such as in the early morning hours, it is important to have an established algorithm for treatment. Any evaluation of a painful hip must consider the possibility of septic arthritis and, if it is suspected, it must be treated emergently. Aspiration should be performed for all hips for which there is a significant concern of septic arthritis. This is one of the few situations where the “conservative approach” is the more aggressive diagnostic and surgical approach. It is far better to inadvertently drain a symptomatic hip than miss a septic arthritis. Aspiration of the hip can be performed by different methods and different staff but should have some method to easily guide the needle to the septic joint and document that the aspiration needle was accurately positioned. Aspiration of the proximal femur should be considered at the time of hip joint aspiration if concomitant osteomyelitis is suspected.

In our institution, the limping toddler with hip or knee pain is evaluated by carefully obtaining a history and physical examination with emphasis on hip range of motion. Routine blood work (complete blood count with differential, sedimentation rate and C-reactive protein, and other inflammatory parameters) is then performed. If the child “could” have septic arthritis as a cause of their symptoms and laboratory values, we routinely have radiology perform an ultrasound and aspiration of the hip to confirm or dissuade that diagnosis. If the patient has a positive hip aspiration, as defined by an increased WBC count, then this fluid is examined via gram stain and culture; clinicians are advised to also obtain blood cultures which can increase the yield of bacterial identification. Surgical irrigation and debridement of the joint is then carried out urgently. If the aspiration is negative yet the patient seems “too sick” for TS, then we will usually expand our study for other inflammatory conditions and obtain an MRI with contrast to look for other causes of hip pain. Should the location of the pain origin be in doubt, a bone scan can be done to detect potential sources of pain which can lead to further testing if indicated.

Although the above approach is extensive and not needed in every patient, failing to vigorously evaluate a presumptive diagnosis of TS can lead to unacceptable patient morbidity. Clearly, a quick and accurate method for identifying TS, and in turn ruling out septic arthritis, would decrease apprehension of parent and physician alike.


  1. 1.

    Jaramillo D, Treves ST, Kasser JR, et al. Osteomyelitis and septic arthritis in children: appropriate use of imaging to guide treatment. Am J Roentgenol 1995; 165: 399–403.

    CAS  Article  Google Scholar 

  2. 2.

    Kermond S, Fink M, Graham K, Carlin JB, Barnett P. A randomized clinical trial: should the child with transient synovitis of the hip be treated with nonsteroidal anti-inflammatory drugs? Ann Emerg Med 2002; 40(3): 294–299, Sep.

    Article  Google Scholar 

  3. 3.

    Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm. J Bone Jnt Surg Am 1999; 81(12): 1662–1670, Dec.

    CAS  Article  Google Scholar 

  4. 4.

    Kocher MS, Mandiga R, Zurakowski D, Barnewolt C, Kasser JR. Validation of a clinical prediction rule for the differentiation between septic arthritis and transient synovitis of the hip in children. J Bone Jnt Surg Am 2004; 86-A(8): 1629–1635, Aug.

    Article  Google Scholar 

  5. 5.

    Kwack KS, Cho JH, Lee JH, Cho JH, Oh KK, Kim SY. Septic arthritis versus transient synovitis of the hip: gadolinium-enhanced MRI finding of decreased perfusion at the femoral epiphysis. AJR Am J Roentgenol 2007; 189(2): 437–445, Aug.

    Article  Google Scholar 

  6. 6.

    Luhmann SJ, Jones A, Schootman M, Gordon JE, Schoenecker PL, Luhmann JD. Differentiation between septic arthritis and transient synovitis of the hip in children with clinical prediction algorithms. J Bone Jnt Surg Am 2004; 86-A(5): 956–962, May.

    Article  Google Scholar 

  7. 7.

    McCarthy JJ, Dormans J, Kozin SH, Pizzutillo PD. Musculoskeletal infections in children: basic treatment principles and recent advancements. Instr Course Lect 2005; 54: 515–528.

    PubMed  Google Scholar 

  8. 8.

    Nelson JD. Skeletal infections in children. Adv Pediatr Infect Dis 1991; 6: 59–78.

    CAS  PubMed  Google Scholar 

  9. 9.

    Quintos-Macasa AM, Serebro L, Menon Y. Transient synovitis of the hip in an adult. South Med J. 2006; 99(2): 184–185.

    Article  Google Scholar 

  10. 10.

    Uziel Y, Butbul-Aviel Y, Barash J, Padeh S, Mukamel M, Gorodnitski N, et al. Recurrent transient synovitis of the hip in childhood. Long term outcome among 39 patients. J Rheumatol 2006; 33(4): 810–811, Apr.

    PubMed  Google Scholar 

  11. 11.

    Yang WJ, Im SA, Lim GY, Chun HJ, Jung NY, Sung MS, et al. MR imaging of transient synovitis: differentiation from septic arthritis. Pediatr Radiol 2006; 36(11): 1154–1158, Epub 2006 Sep 20, Nov.

    Article  Google Scholar 

  12. 12.

    Zamzam MM. The role of ultrasound in differentiating septic arthritis from transient synovitis of the hip in children. J Pediatr Orthop B 2006; 15(6): 418–422, Nov.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to James J. McCarthy.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

McCarthy, J.J., Noonan, K.J. Toxic synovitis. Skeletal Radiol 37, 963–965 (2008).

Download citation