Septic Arthritis in Paediatric Patients: A 10-year Experience at a Single Institution

Abstract

Background

We evaluated the clinical presentation and microbiological profile of a cohort of paediatric patients with septic arthritis at a tertiary institution in Singapore.

Methods

After obtaining institutional board approval, records of all patients below 18 years presenting with septic arthritis between 2010 and 2019 were reviewed. Patient demographic and medical data were analysed.

Results

Of 24 patients with 26 infected joints with a mean age of 7.1 years, 50.0% had pre-existing atopic dermatitis. The most common site infected was the hip (n = 11, 42.3%). The most common pathogen isolated from tissue cultures was methicillin-sensitive-Staphylococcus aureus (MSSA) (n = 9, 37.5%). Twenty-three (95.8%) of the patients underwent surgical drainage.

Conclusions

The skin of patients with atopic dermatitis has been shown to be more frequently colonised with Staphylococcus aureus compared to healthy individuals. The prevalence of atopic dermatitis in our cohort was higher compared to the reported national average of 20.8%. MSSA was the most commonly reported pathogen, and the hip joint most commonly affected. Less than half of the cohort had positive tissue or blood cultures. In paediatric patients with known atopic dermatitis who present with a fever, a painful joint and limited range of motion, septic arthritis should be considered and early drainage and antibiotics instituted.

Introduction

Septic arthritis in the child is a surgical emergency that requires timely diagnosis and prompt intervention in order to avoid its devastating complications. The incidence of septic arthritis in a paediatric population ranges from 0.001% to 0.02% with a higher incidence reported in developing nations. ¹ Given its variable clinical presentation and diagnostic challenges, a high index of suspicion and clinical acuity is imperative.

The aim of this study is to describe the clinical and microbiological profiles of a cohort of paediatric patients admitted to a tertiary care institution and identify associated risk factors for developing septic arthritis.

Methods

This retrospective, single institution study is institutional review board approved (CIRB 2018/2226). From a database search, all patients below 18 years of age who presented between January 2010 and June 2019 with a diagnosis of septic arthritis were included. Patients with aseptic arthritis were excluded.

Patient demographic data were recorded, as were clinical data including time to presentation, positive clinical findings at admission, joint affected, preceding trauma and the presence of existing atopic dermatitis. In addition, surgical drainage requirement, duration of hospital stay and any complications were recorded. Laboratory markers such as C-reactive protein (CRP), white blood cell (WBC) count and microbiology results were also recorded.

Results

Patient Demographics and Site of Infection

24 patients with a total of 26 infected joints met the inclusion criteria for study. There were 12 (50.0%) females and 12 (50.0%) males ranging from 0.33 to 15 years of age. The mean age was 7.1 years (standard deviation, SD = 4.6). The most common site infected was the hip (n = 11, 42.3%), followed by the knee (n = 5, 19.2%), the metacarpophalangeal joint (n = 3, 11.5%), the ankle (n = 2, 7.7%) and the shoulder (n = 2, 7.7%). The remaining sites, specifically the toe interphalangeal (IP) joint, elbow and sacroiliac (SI) joint were all reported once (n = 1, 3.8%). One of the patients with septic arthritis of the left hip also presented with infected knees bilaterally. The patient with the left toe IP joint infection also had concomitant osteomyelitis.

Prevalence of Eczema

Twelve (50.0%) of the patients had atopic dermatitis. The diagnosis of atopic dermatitis had been made prior to presentation by a primary healthcare provider.

Presentation and Initial Blood Markers

The most common presentation was fever, pain and reduced range of motion (ROM) at the affected site (n = 11, 45.8%), followed by fever with pain (n = 4, 16.7%), pain with swelling (n = 3, 12.5%), fever with reduced ROM (n = 3, 12.5%), pain with reduced ROM (n = 2, 8.33%) and pain associated with a limp (n = 1, 4.17%). The mean time to presentation to the ED was 6.1 ± 5.5 days, range 1–21 days). Eight (33.3%) patients experienced preceding trauma to the affected site.

The mean initial CRP level was 78.7 ± 62.5 MG/L (range 0.6–212.3), and the mean initial WBC count was 13.0 ± 6.5 ×10⁹/L (range 7.8–31.4).

Microbiology Investigations

The most common pathogen isolated from tissue cultures was methicillin-sensitive-staphylococcus aureus (MSSA) (n = 9, 37.5%), followed by streptococcus pyogenes Group A (n = 1, 4.17%) and streptococcus pneumoniae (n = 1, 4.17%). No pathogens were isolated from the remaining patients (n = 13, 54.2%).

Drainage and Duration of Hospital Stay

Twenty-three (95.8%) of the patients underwent an arthromy and washout. In the remaining patient, although the magnetic resonance imaging (MRI) of the pelvis showed synovial thickening, periacetabular enhancement and marrow oedema within the left sacroiliac joint, the patient responded well to intravenous high dose antibiotics with downtrending inflammatory markers noted 2 days after admission. After consultation with the infectious disease specialist, the decision was made to continue with antibiotic therapy and to hold off any surgical drainage. She continued to improve clinically with further downtrending of her inflammatory markers till her discharge 8 days after she first presented. The duration of hospital stay across the patients ranged from 3 to 28 days, with a mean of 10.3 ± 6.5 days (Table 1).

Table 1.

Patient Characteristics and Clinical Outcomes.

Mean age (years)	7.1 ± 4.6
Gender (n, %)	Male: 12 (50)
	Female: 12 (50)
Site affected (n, %)	Metacarpophalangeal joint: 3 (12)
	Elbow: 1 (4)
	Shoulder: 2 (8)
	Hips: 11 (42)
	Sacroiliac joint: 1 (4)
	Knees: 5 (19)
	Ankle: 2 (8)
	Interphalangeal joint (toe): 1 (4)
Presentation (n, %)	Fever and pain: 4 (17)
	Fever and reduced range of motion: 3 (13)
	Pain and reduced range of motion: 2 (8)
	Fever, pain and reduced range of motion: 11 (46)
	Pain and swelling: 3 (13)
	Pain and limp: 1 (4)
Mean time to presentation (days)	6.1 ± 5.3
Preceding trauma (n, %)	Yes: 7 (29)
	No: 17 (71)
Initial mean of C-reactive protein level (MM/L)	78.7 ± 62.5
Initial mean of total white blood cell count (×109/L)	13.0 ± 6.5
Pathogen isolated (n, %)	MSSA: 9 (38)
	Streptococcus pyogenes: 1 (4)
	Streptococcus pneumonia: 1 (4)
	Negative: 13 (54)
Drainage (n, %)	Yes: 23 (96)
	No: 1 (4)
Mean duration of hospital stay (days)	10.3 ± 6.5
Atopic dermatitis (n, %)	Yes: 12 (50)
	No: 12 (50)

Mean values are presented as the mean ± standard deviation.

Discussion

Atopic dermatitis refers to a chronic inflammatory condition that affects approximately 20.8% of school-going children in Singapore. ² As a result of chronic inflammation, the integrity of the epidermis may be reduced, ³ allowing pathogens to more easily penetrate the skin barrier, which may lead to infection, including septic arthritis. There is a paucity of literature on the association between atopic dermatitis and septic arthritis in the paediatric population.

The skin of patients with atopic dermatitis is more frequently colonised with Staphylococcus aureus (SA) compared to healthy individuals. Higaki et al. ⁴ found that 85.7% of AD patients had cutaneous carriage of SA versus 25% of healthy controls. Saxena et al. ⁵ observed a 49% increase in hospital admissions for SA infections amongst paediatric patients over a 10-year period. In particular, they noted a 46.1% and 28.3% increase in admissions for osteomyelitis and septic arthritis, respectively.

In patients with atopic dermatitis, systems at risk for SA infection include the cardiovascular, pulmonary, ocular and musculoskeletal systems. A systemic review by Seranno et al. ⁶ found that atopic dermatitis was associated with increased odds of developing an extracutaneous infection including those involving the musculoskeletal system. The current literature on the association between septic arthritis and atopic dermatitis remains sparse, consisting largely of case reports. Patel et al. reported on a nine-year-old girl with poorly controlled and severe atopic dermatitis who presented with recurrent MRSA septic arthritis and osteomyelitis of the shoulder requiring repeated drainage. ⁶ Kitamura reported on a fifteen-year-old girl with chronic dermatitis who presented with septic arthritis of the hip. Cultures from her skin and joint aspirate were positive for SA. ⁷

Kitamura suggested that the mechanism for septic arthritis in a patient with AD was a result of skin barrier damage from scratching. ⁸ A reduction in the secretory component of immunoglobulin A (sIgA) in AD patients as observed by Imayama can lead to a hyperproliferation of SA on the skin which then enters the systemic circulation more easily as a result of the breach in the skin barrier. Haemtogenous spread of the bacterium can then lead to septic arthritis and osteomyelitis in these patients. ⁷

Our study noted the prevalence of atopic dermatitis in our cohort of patients with septic arthritis to be 50.0%, which is higher compared to the reported national average of 20.8%. ² Less than half of the patients in our cohort had positive tissue or blood cultures, which further highlights the importance of maintaining a high index of clinical suspicion when diagnosing septic arthritis. MSSA was the most commonly reported pathogen in both groups of patients, and the hip joint most commonly affected. In paediatric patients with known atopic dermatitis, who present with fever, elevated initial CRP and WBC and a painful joint with limited range of motion, septic arthritis should be considered and timely drainage together with the appropriate empirical antibiotics instituted.