|Year : 2017 | Volume
| Issue : 2 | Page : 43-50
Early Surgical Intervention in Children with a Suspected Diagnosis of Acute Septic Arthritis or Osteomyelitis: Is it Justified?
Vasudev Children’s Orthopaedics Centre, Bellary, Karnataka, India
|Date of Web Publication||16-Jul-2018|
‘Vasudev’, Opp. Shanti Sishu Vihar School, Talur Road, Bellary - 583 103, Karnataka
Source of Support: None, Conflict of Interest: None
Context: Early surgical intervention is the key for good outcome in children with acute haematogenous osteomyelitis (AHO) or septic arthritis. Often there is an impasse to observe or intervene early when the diagnosis is inconclusive due to blunted clinical findings and equivocal investigations. Aim: This study is aimed at justifying early surgical intervention in these doubtful/suspected cases. Settings and Design: This study was a retrospective review of AHO/septic arthritis treated by the author between August 2010 and January 2015. A new scoring system, haematogenous osteomyelitis and septic arthritis (HOMSA) score was developed to aid in classifying and decision-making. With a maximum score of 8, a score >6 in the absence of infection elsewhere or a score <6 with radiological evidence makes the diagnosis of AHO/septic arthritis definite. A score ≤6 without radiological evidence makes the diagnosis suspected/doubtful. Outcome was measured by a new discrete criterion for the upper and lower limbs. Materials and Methods: The protocol was early surgical intervention in both the groups. In septic arthritis, open arthrotomy along with joint lavage and debridement was performed. In AHO, bone decompression with abscess drainage was performed. Initial parenteral administration of antibiotics was followed by oral antibiotic administration. Necessary adjuvant treatment was given. Results: Thirty-four children with 40 sites of infection were identified. Among them, 50% were neonates. Only 4/40 children were treated conservatively. Following surgery, outcome was excellent-to-good in 92.8% of the children with doubtful/suspected diagnosis and 96.6% with definite infection. One child in each group who were treated surgically, and two children in the group with definite infection treated non-operatively had fair-to-poor outcomes. Conclusion: Early surgical intervention is justified even in children with a doubtful/suspected diagnosis of AHO or Septic Arthritis. The new scoring system, HOMSA Score, is a better tool to diagnose Acute septic arthritis or osteomyelitis, even with limited resources.
Keywords: Acute haematogenous osteomyelitis, early surgery, infants and children, neonates, new outcome, measure, new scoring system, septic arthritis
|How to cite this article:|
Harish P. Early Surgical Intervention in Children with a Suspected Diagnosis of Acute Septic Arthritis or Osteomyelitis: Is it Justified?. Paediatr Orthop Relat Sci 2017;3:43-50
|How to cite this URL:|
Harish P. Early Surgical Intervention in Children with a Suspected Diagnosis of Acute Septic Arthritis or Osteomyelitis: Is it Justified?. Paediatr Orthop Relat Sci [serial online] 2017 [cited 2019 Feb 19];3:43-50. Available from: http://www.pors.co.in/text.asp?2017/3/2/43/236713
| Introduction|| |
Accurate diagnosis, appropriate prompt surgical intervention, and antibiotics have been established to form the cornerstone in the treatment of paediatric acute haematogenous osteomyelitis (AHO) and septic arthritis.,, However, decision-making to intervene surgically or treat conservatively in borderline situations is difficult. This can due to the blunting of clinical symptoms or inconclusive investigations and the lack of a single definitive tool in diagnosing bone and joint infection. These often are encountered in neonates, infants and young children, especially when deep joints such as the hip are involved. There is greater urgency in diagnosing and treating septic arthritis than AHO. It does not mean that AHO has to be treated less aggressively. Literature obliquely implies that, when in doubt, one should not hesitate to drain the joint or decompress the bone.,,,,,,, This has been referred to as presumptive surgery, because the risk of delaying the surgery in a child with infection outweigh the risks of surgery in a child with no infection. This study was intended at justifying early surgical intervention in these doubtful cases of septic arthritis/AHO.
| Materials and Methods|| |
The study was a retrospective analysis of outcome after early surgical intervention in patients who are a doubtful case of acute septic arthritis or AHO. The outcomes were compared with those who had definitive diagnosis.
To establish the diagnosis efficiently and to have an objective criterion for defining a case as definite or suspicious, a new scoring system termed the haematogenous osteomyelitis and septic arthritis (HOMSA) score was developed. The HOMSA score, as shown in [Figure 1], has clinical, laboratory and radiological parameters. With a score of one for each of the four parameters in both clinical and lab, a total score of eight is possible. Radiological parameters were not given a score because any evidence of infection on radiological assessments would make the diagnosis definitive, even with a low score. With a maximum score of 8, score >6 in the absence of infection elsewhere or a score <6 with radiological evidence makes the diagnosis of AHO/septic arthritis definite. A score ≤6 without radiological evidence makes the diagnosis suspected/doubtful [Figure 2]. The possible situations are shown in [Figure 3].
Infant or child with a clinical suspicion of septic arthritis/AHO was managed as per the protocol in [Figure 4]. Whether it was a doubtful or a definitive case, the method of surgical intervention was the same. For a septic joint, the surgical intervention would involve arthrotomy, joint lavage and, if necessary, debridement. AHO was addressed with the drainage of the abscess if any along with a decompression of the bone. Care was taken to minimise periosteal stripping. Vacuum drain tube was left in for 5–7 days, until there was no drainage for 48 consecutive hours. The empirical antibiotics used were intravenous amoxycillin and clavulanic acid combination, as well as amikacin. If necessary, on the basis of the culture and sensitivity, antibiotics were changed. Intravenous antibiotics were given for 10–14 days or, if needed, longer. Cell counts, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were repeated every fifth day. The decision to change to oral antibiotics was based on clinical improvement and the normalisation of CRP, as well as the fall in ESR and the availability of effective oral antibiotic and compliance. Oral antibiotics were continued for at least 4–6 weeks. If it was not possible to switch to oral antibiotics, then parenteral antibiotics were continued for 4 weeks. Adjuvant treatment such as pavlik harness, spica cast or splinting were given as per the need.
|Figure 4: Protocol for the management of acute haematogenous osteomyelitis and septic arthritis|
Click here to view
To assess the outcome, two similar criteria, one each for the upper and lower limbs, were developed. Each of these had five variables. These criteria were a combination of the range of movement, deformity, function and also the possible need for further treatment based on the sequelae if any. The outcome was graded as excellent, good, fair or poor, as shown in [Figure 5] and [Figure 6].
| Results|| |
All children treated for musculoskeletal infections between August 2010 and January 2015 were reviewed. Of the forty-one children who were identified, thirty-four children diagnosed with acute septic arthritis or osteomyelitis and followed up till complete resolution were included in the study. Their age ranged from 7 days to 13 years (mean age 2.25 years). Seven children with soft-tissue infection of the extremities were excluded. Fifty percent (17 babies) of those included were neonates, while the other 17 children were aged between 1.5 and 13 years [Table 1].
According to the HOMSA score, 14 children with 15 sites were categorised as doubtful/suspected, and 20 children with 25 sites were categorised as definitive infections [Table 2]. Two children with a score >6 were still considered to have a doubtful/suspected infection of the joint, because they had an additional focus of soft-tissue infection without any radiological evidence. The mean duration of symptoms at the time of presentation in the group with definitive diagnosis was 6.65 days (range 3–14 days), and in the group with suspected diagnosis, the mean duration was 4.78 days (range 2–14 days). Infections in 23/25 Definitive sites and 13/15 doubtful sites were treated by appropriate surgery. All the suspected patients were confirmed to have infection based on the intra-op findings of the presence of purulent/sero-purulent collection in a joint with suspected septic arthritis and an abscess in the juxta-periosteal/sub-periosteal/intra-osseous sites in patients with suspected osteomyelitis. Thirty-six percent (5/14) of the patients in this group had positive cultures. Fifty-five percent (11/20) of the patients in the definitive group had positive cultures. Of the total 16 positive cultures, seven were gram-positive cocci methicillin sensitive staphylococcus aureus (MSSA), five were methicillin resistant staphylococcus aureus (MRSA) and four were gram-negative rods [Table 3].
The average follow-up after surgery was close to 3 years (2.9 years) in either group [Table 4]. 94.7% of the children had good-to-excellent outcome after surgical treatment. It was marginally better in the patients with definitive infection (96.6% vs. 92.8%) compared to the group of children with suspected infection [Table 5]. While both the children with suspected infection had excellent-to-good outcome after non-operative treatment, one baby with definitive infection in both her hips had fair outcome after being managed conservatively. Overall, three children with definitive diagnosis had a fair-to-poor outcome. A child with radius and ulna osteomyelitis, treated with abscess drainage and decompression, developed a wrist deformity secondary to radius shaft non-union. Another child who had bilateral septic arthritis, but presented late, was managed conservatively. This child developed residual acetabular dysplasia bilateral, requiring bilateral Pemberton at 5 years of age. This child was categorised as fair outcome. The third child with multi-focal osteomyelitis and septic arthritis developed genu valgum due to the lateral physeal bar of the distal femur and also a bone defect of the ulna, both requiring surgery. This child was assessed to have poor outcome. However, there was no statistically significant difference in the outcomes. None of the patients had surgical or anaesthesia-related complications.
GraphPad Prism version 7 software (GraphPad Software, CA, USA) was used for analysis. The average age and time of follow-up were calculated by measuring the mean. Fisher’s exact test was used to assess the statistical significance of the difference of various parameters. The parameters assessed were the difference in culture-positive cases in both the groups, final outcome between the two groups and sub-group analysis in the suspected group where the surgical and conservative treatments were compared.
| Discussion|| |
The study hinges on the principle of early surgical intervention in septic arthritis or osteomyelitis to prevent further damage to the concerned joint or bone. Herein, the term early implies even before there are positive investigations confirming AHO/septic arthritis. These have been termed by some as presumed AHO/septic arthritis.
The classical clinical features for septic arthritis or osteomyelitis are high fever, white blood cells (WBC) count and septicaemia. In spite of few favourable clinical findings, many a times, the diagnosis is inconclusive because of the lack of convincing findings on investigation. These are frequently blunted due to multiple factors., Temperature more than 38°C occurs only in 36–74% of the patients.,, The irrational use of antibiotics for variable periods before presentation adds to the blunting of symptoms. The white cell count, differential counts, ESR, ultrasound and X-ray may not aid in the diagnosis, especially in very early stages, and more so in infants and neonates. Kocher’s criteria, which predicts the probability of infection to differentiate septic arthritis from transient synovitis, is often found to be wanting in these borderline cases because of the limited variables. The high rates of probability of septic arthritis in the initial study by Kocher et al. were found to be low in subsequent studies., CRP is very sensitive with a peak on day 2 or day 3; it aids in the early identification of infection. Caird et al. later on added the fifth variable CRP to Kocher’s criteria and found the probability to be 98% when all five were present. Given the combination of a high incidence of hip septic arthritis (40%) and a great proportion of neonates (50%), Kocher’s criteria was not applicable to majority to them in this study. Early bone scan or magnetic resonance imaging (MRI), blood and tissue culture and supportive lab parameters − ESR and CRP − have been suggested as the diagnostic procedures of choice by Bonheffer et al. MRI is very sensitive, specific and has a high positive predictive value (85%) in picking up bone and joint infection even in very early stages. However, given the background of deprived infrastructure and cost constraints in small towns and rural places, it becomes prudent to go by the clinical evaluation, simple laboratory and radiological investigations. Peltola and Vahvanen found the diagnosis of osteomyelitis to be definite when four of the five variables (pus aspirated from the bone, positive bone/blood culture, classical symptoms, limited movements and typical radiological changes) were positive. Morrey and Peterson classified osteomyelitis as definite, probable or likely based primarily on isolating the organism from the affected site or blood. Both these diagnostic criteria are applicable only after intervention and not on presentation; hence, they are not useful for early decision-making. Moreover, cultures were negative in 40–70% of the cases., This study had a similar finding of 53% negative cultures (47% culture positive).
A need for an objective tool that is comprehensive, to be applicable in any situation, was felt. The aforementioned three diagnostic criteria were used as a framework to design the new scoring system, HOMSA score. This was developed by including more synonymous symptoms and clinical findings, the age-specific normal values of white cell counts, ESR and CRP. The radiological findings of infection on X-rays or ultrasound would make the diagnosis more definitive and, hence, were added to the scoring system. Of the total score of 8, clinical score and lab score have 4 each. A watershed line of score 6 was fixed to make a diagnosis as definitive or suspected. This was again extrapolated from the aforementioned scores, wherein the presence of more than 75% of the criteria would make the diagnosis definite. Applying HOMSA score to all the cases and classifying them into suspected and definite made further evaluation objective. The sub-group analysis of group with suspected infection showed that 71% (10/14) of the children had clinical scores 3 or 4, whereas 57% (8/14) of the children had lab scores of 3 or 4. While this difference is not statistically significant, it emphasises the importance of clinical examination in diagnosing acute joint and bone infection. When it is a neonate or infant, a high index of suspicion is needed. They are often brought for the paucity of movements (synonymous with the inability to bear weight or use the limbs in older children). Although painful movements could be elicited, fever response was seldom seen in them. The blunting of clinical findings, which can possibly be due to inadequate immune response, has been observed in this sub-group, as only 50% (4 out of the 8 neonates) had a clinical score of 3. It could not be ascertained as to how many of these were transient synovitis. Literature suggests that cultures were negative in 40–70% of the cases., A similar rate of 64% negative cultures (36% positive cultures) were seen in children with suspected infection, while the cultures were negative even in 45% of the children in the definitive group. This difference was not significant (P = 0.3151). It can be implied that positive culture from the site of infection was not essential to make a diagnosis of infection.
There is a considerable disparity of protocol regarding the timing, extent and the necessity of surgery to treat AHO. The treatment of a definitive case of infection is well accepted to be early surgical intervention along with appropriate antibiotics. The decision needs to be made in doubtful cases whether to intervene early by appropriate surgery or wait till clinical, lab and radiological parameters are conclusive. The protocols established in the developed nations, which have better infrastructure and the economics of healthcare cannot be replicated in the rural and semi-urban places of underdeveloped/developing nations. They have to be adapted to the local needs, so as to deliver the right treatment. Cole and associates observed that majority of the children with AHO could be cured with a single course of antibiotics and immobilisation if treatment was initiated within a day or 2 of the onset of symptoms. In our setting, seldom a child is brought this early. Almost every child is brought later than this, when the condition has advanced, but before a stage where only clinical signs and symptoms are positive; however, X-rays are still normal, and the level of blood parameters can vary. At this stage, early surgical intervention is warranted even before a positive radiological evidence, so as to prevent long-term sequelae. Segbefia and Howard, in their perspective of acute septic arthritis and osteomyelitis among the African community, found early drainage and uncomplicated recovery to be the exception rather than the rule in their reported African series. They found that early drainage resulted in prompt recovery with no long-term morbidity.
A minimally invasive way of diagnosing and treating a joint infection is needle or arthroscopy-assisted aspiration, lavage, and the analysis of the aspirate. This again requires sedation or anaesthesia and ultrasound/C-arm guidance for the deep joints. If the condition is not resolved, then repeat aspiration and lavage or surgical intervention is warranted. These will add up to the cost and risk of repeat anaesthesia. This step-wise approach is difficult to implement in a situation of limited resources. For the same reason, it may not be possible to follow the clinical practice guidelines (CPG) as suggested by Kocher et al. It is prudent to hit the bull’s eye at once by adapting the CPG to the local facilities.Smith et al., in their experimental study on rabbit knees found the first measurable articular damage by 8 h after the inoculation of bacteria and predicted a possibility of permanent damage when delayed beyond 8 h. Once inflammatory cytokines are released into the joint, the presence of living bacteria is not necessary for cartilage destruction to continue. Nunn et al. have shown uncomplicated recovery in septic arthritis of the hip when intervened within 5 days by arthrotomy and appropriate antibiotics. Kuong et al. have found that intervention more than 4 days after onset had poor outcome. Welcon et al. found a 1.6 times higher probability of poor outcome when intervened <5 days. It is the experience of most of the orthopaedic surgeons to find an sequelae of septic arthritis or osteomyelitis when surgical intervention has been delayed. Hence, the protocol of early surgical intervention, a more definitive treatment even in a patient who is a doubtful case of septic arthritis or osteomyelitis, was adopted.
Given the average follow-up of close to 3 years, it is long enough for any manifestations of post-infection sequelae. An excellent-to-good outcome in 92.8% of the children who underwent early surgery for suspected infection proves the protocol to be appropriate. The only conflicting finding to this is the excellent-to-good outcome in two children with suspected infection who were managed conservatively. Because the number of children in this group is very small (n = 2), it can be considered to be by chance. The parents of one particular neonate classified as having definitive bilateral septic arthritis of the hip had not consented for surgery. They were managed conservatively. A poor result in these two hips treated emphasises the fact of early surgical intervention. A positive culture is a definitive evidence of infection. Considering the literature and a 55% culture positive in the definitive group and 36% in the group with suspected infection, unwarranted surgery would have occurred in 19% of the cases in the group with suspected infection. Going by the simple calculation, the other 81% of the children benefited from early surgery in this group with suspected infection. Intra-operative findings (as mentioned earlier), suggestive of septic arthritis or osteomyelitis in all these cases, further accentuate the decision in the favour of early surgical intervention even in doubtful cases. Until refuted by Level-1 evidence, this study justifies the protocol of early surgical intervention, as shown by the excellent-to-good results in 92.8% of the children.
One needs to weigh the risk of anaesthesia in a very sick neonate or infant, because the risk from surgery per se is very less given the small incision and limited dissection. The surgery can actually be accomplished much faster than aspiration and lavage.
This study being a retrospective analysis of a small cohort is a drawback to establish statistical significance. The HOMSA score needs to be validated statistically. This score can also form the basis for future randomised trials comparing non-operative or early surgical intervention in children with suspected AHO or septic arthritis.
| Conclusion|| |
Early surgical intervention is justified in patients who are a doubtful/suspected case of AHO or septic arthritis. The new scoring system, HOMSA score, is a better tool to diagnose septic arthritis or osteomyelitis, even with limited resources.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Kuong EE, To M, Yuen MH, Choi AK, Fong CM, Chow W. Pitfalls in diagnosing septic arthritis in Hong Kong children: Ten years experience. Hong Kong Med J 2012;18:482-7.
Stans AA. Musculoskeletal infection. In: Weinstein SL, Flynn JM, editors. Lovell and Winter’s Paediatric Orthopaedics. 7th
ed. Philadelphia: Lippincott Williams & Wilkins, Wolters Kluwer 2014. p. 369–425.
Herring JA, editor. Infections of the musculoskeletal system. Tachdjian’s Paediatric Orthopaedics. 4th
ed. Philadelphia: Saunders Elsevier 2008. p. 2089-155.
Bennet OM, Namnyak SS. Acute septic arthritis of the hip joint in infancy and childhood. Clin Orthop Relat Res 1992;281:123-32.
Chen CE, Ko JY, Li CC, Wang CJ. Acute septic arthritis of the hip in children. Arch Orthop Trauma Surg 2001;121:521.
Kocher MS, Mandiga R, Murphy JM, Goldmann D, Harper M, Sundel R et al.
A clinical practice guideline for treatment of septic arthritis in children. J Bone Joint Surg Am 2003;85:994-9.
Lyon RM, Evanich JD. Culture-negative septic arthritis in children. J Pediatr Orthop 1999;19:655.
Morrey BF, Bianco AJ Jr, Rhodes KH. Septic arthritis in children. Orthop Clin N Am 1975;6:923-34.
Cole WG, Dalziel RE, Leitl S. Treatment of acute osteomyelitis in childhood. Bone Joint J 1982;64:218-23.
Wang CL, Wang SM, Yand YJ, Tsai CH, Liu CC. Septic arthritis in children: Relationship of causative pathogens, complications and outcomes. J Microbial Immunol Infect 2003;36:41-6.
Scoor RJ, Christofersen MR, Roberson WW Jr, Davidson RS, Rankin L, Drummond DS. Acute osteomyelitis in children: A review of 116 cases. J Pediatr Orthop 1990;10:649-52.
Klein DM, Barbera C, Gray ST, Spero CR, Perrier G, Teicher JL. Sensitivity of objective parameters in the diagnosis of paediatric septic hips. Clin Orthop Relat Res 1997;338:153-9.
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 Joint Surg Am 1999;81:1662-70.
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 Joint Surg 2004;86:956-62.
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 Joint Surg 2004;86:1629-35.
Caird MS, Flynn JM, Leung YL, Millman JE, Joann GD, Dormans JP. Factors distinguishing septic arthritis from transient synovitis of the hip in children. J Bone Joint Surg Am 2006;88:1251-7.
Bonheffer J, Haeberle B, Schaad UB, Heininger U. Diagnosis of acute haematogenous osteomyelitis and septic arthritis: 20 years experience at the University Children’s Hospital Basel. Swiss Med Wkly 2001;131:575-81.
Peltola H, Vahvanen V, Aalto K. Fever, C-reactive protein, and erythrocyte sedimentation rate in monitoring recovery from septic arthritis: A preliminary study. J Pediatr Orthop 1984;4:170-4.
Morrey BF, Peterson HS. Hematrogenous pyogenic osteomyelitis in children. Orthop Clin North Am 1976;6:935-51.
Smith RL, Schurman DJ, Kajiyama G, Mell M, Gilkerson E. The effect of antibiotics on the destruction of cartilage in experimental infectious arthritis. J Bone Joint Surg Am 1987;69:1063-8.
Nunn TR, Cheung WY, Rollinson PD. A prospective study of pyogenic sepsis of the hip in childhood. J Bone Joint Surg Br 2007;89:100-6.
Welcon CJ, Long SS, Fisher MC, Alburger PD. Pyogenic arthritis in infants and children: A review of 95 cases. Pediatr Infect Dis 1986;5:669-76.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]