Paediatric musculoskeletal infections with Panton-Valentine leucocidin

DOI: https://doi.org/10.4414/smw.2018.14669

Introduction

Panton-Valentine leucocidin (PVL) is a cytotoxin produced by any strain of Staphylococcus aureus that causes leucocyte destruction and important tissue necrosis. This pathogen is currently responsible for extensive necrotic skin lesions, potentially fatal necrotising pneumonia and severe musculoskeletal infections. PVL may be produced by various S. aureus strains, both methicillin-sensitive S. aureus (MSSA) and community-acquired methicillin-resistant S. aureus (CA-MRSA) [1]. In a European study, the prevalence of PVL-positive community-acquired S. aureus amounted to 18.6% and 7.8% of the isolates were MRSA [2]. The correlation between production of PVL and the severity of infection is controversial, since the latter seems most likely to depend on the patient’s age. A recent meta-analysis in adults illustrated the lack of coherence between infection by PVL-positive strains and the outcome of pneumonia, musculoskeletal disease or bacteraemia [3]. Nevertheless, the paediatric literature reports severe local necrosis with a pronounced inflammatory response induced by PVL, with consequent short- and long-term complications. As a matter of fact, several studies highlight the peculiarity of PVL-positive musculoskeletal infections in children and demonstrate that affected patients presented more elevated inflammatory parameters and a higher complication rate requiring repeated surgical interventions, intensive care unit admission and a consequently long hospital stay [4–9]. At our two university medical centres, we identified the first case of PVL-positive musculoskeletal infection only in 2008. The purpose of this article was to present a series of nine cases of paediatric musculoskeletal infections due to PVL-producing S. Aureus, highlighting their clinical aspects, the results of laboratory studies and the outcome of therapy (both surgical procedures and length of antibiotic treatment), as well as the complications noted.

Patients and methods

After approval from the respective Children’s Hospital Ethics Review Committee, we retrospectively collected the medical charts of all children aged from 0–16 years old admitted in our two university hospitals (CHUV Lausanne and HUG Geneva) between January 2008 and December 2016 for PVL-positive musculoskeletal infections. The criteria established by Morrey were used to estimate children’s risk of having a bone or joint infection [10]. Diagnoses of osteoarticular infections were confirmed using imaging techniques (plain radiography, ultrasonography, ^99mTC bone scan, computed tomography, magnetic resonance imaging). Children diagnosed as having musculoskeletal infections were further categorised according to the following primary diagnoses: acute haematogenous osteomyelitis, septic arthritis, and skin and soft tissue infections. Demographics such as age, gender and place of origin were recorded for each patient as well as clinical parameters such as onset of symptoms before admission, fever at the time of admission, the fever duration under treatment, and the time elapsed between the onset of symptoms and diagnosis. Biological work-up included white blood cell (WBC) count, left-band shift percentage, platelet count, C-reactive protein (CRP) level, and erythrocyte sedimentation rate (ESR).

Microbiological methods

Blood cultures were typically used to isolate the microorganisms responsible for musculoskeletal infections. The study’s blood culture media used was an automated blood culture system (BD BACTEC FX). Joint fluid, pus from soft tissue abscesses or bone aspirate samples were sent to the laboratory for Gram staining, cell count, and immediate inoculation onto Columbia blood agar (incubated under anaerobic conditions), CDC anaerobe 5% sheep blood agar (incubated under anaerobic conditions), chocolate agar (incubated in a CO₂-enriched atmosphere), and brain-heart broth. These media were incubated for 10 days. The presence of the PVL gene in S. aureus strains was assessed with a rapid real time polymerase chain-reaction (PCR) assay in one centre until 2014. The remaining two cases in 2016 were analysed as described previously [11].

Treatment

If a suspect fluid was recovered from joint-fluid aspiration, joint lavage was performed via arthrotomy, or arthroscopy. Surgical exploration for osteomyelitis generally included taking subperiosteal or bone aspirate specimens, and aspiration-curettage of any abscesses. For skin and soft tissue infections, any abscesses were drained. Empirical antibiotic therapy started with an intravenous antibiotic chosen by the attending physician responsible for each patient, after having obtained samples. The antibiotic regimen was thereafter adjusted according to antibiotic sensitivity. The duration of antibiotic treatment and mode of administration were then determined with regard to each patient’s age, bacterial characteristics and, above all, their response to the treatment. Collected data about treatment and follow-up included duration of intravenous and oral antibiotic treatment, number of surgical procedures and length of hospital stay. Long-term outcome was evaluated through clinical assessment and radiological follow-up.

Results

The study enrolled nine children admitted for PVL-positive musculoskeletal infection (seven males, two females) during the 9 years considered, with ages ranging from 2 to 15 years old (median 12 years) (table 1). A retrospective analysis of all paediatric bacteriological samples positive for S. aureus available in one centre yielded five PVL-positive cases out of 571 samples, resulting in a prevalence of 0.9% in our observation period from 2008 to 2016. Vaccination status was recorded for seven of the nine patients, all of them being vaccinated according to national guidelines. The children studied had haematogenous acute osteomyelitis in four cases, septic arthritis in two and skin and soft tissue infections in the remaining three. The mean admission temperature was 38.7°C (range 37.3–41.0°C), whereas three patients (33%) had a temperature beneath 38°C on admission.

The median WBC count was 12,700/mm³ (range 6500–22,600) and was considered elevated in five patients (cut-off: 12,000/mm³). CRP was available for eight patients and was elevated in six, with a median value of 171 mg/l. The ESR was available for seven patients and was elevated in all of them with the mean value of 62 mm/h. Finally, the mean platelet count was 241,000/mm³.

The pathogen was identified from blood cultures in six patients and in operative samples cultures in seven cases, with five patients having both positive blood cultures and positive operative samples. Bacteriological investigations demonstrated eight cases of MSSA and one case of CA-MRSA.

All patients underwent surgical procedures (open surgical drainage/lavage) and the mean number of interventions per patient was three (range one to eight). The site of infection was accessed by bone drilling in cases of osteomyelitis, arthrotomy for arthritis and local incision for skin and soft tissue infection. Surgical treatment was complemented by antibiotherapy, and the antibiotics used are listed in table 2. The duration of intravenous treatment amounted in average to 16 days (range 0–62 days); all but one patient were subsequently treated with oral antibiotics (mean duration 25 days). Intravenous antibiotherapy was stopped for the remaining case after 54 days without switching to oral antibiotics.

Patients sustained various complications (fig. 1) in the form of infection recurrence (five cases), growth arrest with limb length discrepancy (two cases) and pathological fracture (one case) from underlying oncological disease; one patient developed psychological troubles (suicidality) in the course of treatment.

Figure 1 Clinical course of patient 6.

Note the rapid evolution from a normal X-ray on day 1 (a) to persisting defects after 5 years (d), and between magnetic resonance imaging on day 3 (b) and at 10 months (c).

Finally we noted that patients with PVL-positive osteoarticular infections had a higher temperature on admission, and WBC, CRP and ESR were more elevated compared with patients with skin and soft tissue infections.

The following risk factors were identified among the patients with osteoarticular infections: one had a history of acute lymphoid leukaemia and a medical port in situ. Two patients had a history of previous trauma, one had furunculosis at the site of infection and one reported a scratch injury from intense itching. One patient performed contact sport. Among the patients with skin and soft tissue infections, only one case of furunculosis was described.

Discussion

Nine cases of musculoskeletal infections due to PVL-producing S. aureus were identified during this 9-year study. Six were osteoarticular and three affected skin and soft tissue. Our case series illustrates that the prevalence of PVL-positive musculoskeletal infections is still low in western Switzerland. This study is consistent with the results of investigations conducted by the Paediatric Infectious Disease Group Switzerland, which found only 1.6% of PVL producers in a cohort of 572 MSSA strains [12]. Our observation is in line with other studies reporting a low prevalence of PVL-positive strains [13] and on the predominance of MSSA-PVL in continental Europe [7, 9, 14, 15]. Elsewhere, it is recognised that the proportion of PVL-producing strains is consistently higher in MRSA (74–100%) than MSSA (9–46%) [1], and that the rate of MRSA differs between locations (USA 56–68% MRSA-positive vs Greece 26%). Thus, the overall reported proportion of PVL-producing S. aureus depends on the prevalence of MRSA (USA 78% vs Greece 37%) [1].

One must nevertheless keep in mind that our result may be biased by two facts, as previously reported [9]. Firstly, we know neither the number of PVL-positive cases that are treated in primary institutions nor the proportion of S. aureus strains referred for PVL testing. We can thus expect that the number of musculoskeletal infections due to PVL-positive S. aureus is currently probably underestimated. Secondly, the rate of MRSA in the paediatric population is very low in western Switzerland. Since the prevalence of PVL-producing strains depends on the prevalence of MRSA, this very low prevalence of MRSA might play a role in the distribution of PVL in our region. The majority of European reports, mainly from France and the UK, report on MSSA-PVL. Hence, our results confirm both the low general prevalence of PVL in the paediatric population and the predominance of MSSA-PVL in Europe [7, 9, 14, 15].

PVL is secreted by any S. aureus strains as two components that assemble to a pore-forming complex on polymorphonuclear leucocyte membranes [16]. Depending on the concentration, PVL induces either apoptosis or cytolysis with release of inflammatory factors and subsequent tissue inflammation [16]. Generally, destruction of leucocytes leads to the characteristic discordance of a low/normal leucocyte count and high levels of inflammatory parameters (e.g., CRP) [17]. In our series, this discrepancy of low WBC and high CRP level was detected in only two cases on admission (table 1). Interestingly, the discordance arose from osteoarticular infections only, since skin and soft tissue infections presented with a normal white blood cell count in all three patients and merely increased CRP levels. The low number of two out of nine patients with the “typical” initial laboratory findings illustrates the risk of misinterpreting the infection at a first glance.

To our knowledge, no risk factors for PVL-positive infections have been defined so far. However, it is well known that sharing of personal items, particularly in sports clubs and poor hygiene practices, including insufficient disinfection and protection of skin lesions, increase the rate of PVL-related infections [17]. Among other factors that increase the suspicion are non-white ethnicity, preceding injury, contact sport, recent travel and influenza or other respiratory infection [9, 11]. Our results correspond with these, apart from influenza status, which was not tested in our patients given the lack of relevant symptoms of pneumonia. The risk factors listed here overlap with the risk factors for transmission of MSSA and MRSA. Therefore, no PVL-specific risk factors were identified. Controversy persists as to whether nasal carriage of PVL is systematically linked to infection [17, 18]. Since nasopharyngeal swabs was tested in only one of our patients, we cannot comment on this potential coherence.

The rate of complications is recognised to be high with PVL-producing S. aureus infections. In our series, we recorded orthopaedic and psychological sequels. Two of them merit particular attention: a pathological fracture, which occurred in the context of an underlying oncological disease following local PVL infection; and suicidality in a patient with a particularly long evolution, requiring 8 weeks of hospitalisation followed by 30 weeks of peroral antibiotic treatment. Interestingly, one of the antimicrobial agents used in the latter case was levofloxacin, a substance known to potentially favour psychological deterioration (swissmedic database). Hence, these observations highlight multiple facets of PVL-positive infections: a rapidly evolving disease with the need for urgent invasive treatment and the potential for multiple long-term sequels. However, neither of our cases suffered from pulmonary involvement or thrombosis [3].

Despite the low prevalence of PVL in our area, PVL represents an important component in the cascade of paediatric musculoskeletal infections. The observed rapid deterioration, the high recurrence rate and the long list of complications underline the need for early and targeted treatment. Generally, a combination of an anti-staphylococcal regimen (amoxicillin / clavulanic acid or flucloxacillin) and an anti-toxic agent blocking bacterial protein-synthesis of PVL (clindamycin, linezolid, or gentamicin) is recommended [8, 14, 19]. Antibiotic therapy is essential, yet not exclusive. Inhibition of PVL production requires an anti-staphylococcal agent at a drug concentration that ranges far above the minimal inhibitory concentration at the site of infection [19]. Insufficient dosage of β-lactams and, to a lesser extent, vancomycin have been shown to even enhance PVL secretion, thus aggravating the symptoms [19–21]. Antibiotic treatment is even more challenging in the case of CA-MRSA, where the use of β-lactams is limited. Consequently, standard care limited to antibiotic therapy (even in the case of approved regimens) results in failure of treatment in 85% of cases [19], especially as antibiotic distribution is reduced in necrotic tissue. Therefore, surgery plays a major role in achieving a positive outcome. Early and complete removal of PVL suppuration from the body and debridement of necrotic tissues with surgery is a mandatory goal in treatment [19]. Surgical resection of affected areas is mandatory to stop the circle of encapsulation of bacteria and their toxin. According to the literature, 71% of patients with bone and joint infections require more than one procedure, with a median of three interventions per patient [19], which is consistent with our results. In soft tissue infections, outcomes are particularly dependent on surgical drainage and remain favourable, even in the event of inappropriate antibiotic treatment [19].

Conclusion

Paediatric musculoskeletal infections by Panton-Valentine leucocidin represent a rare, but highly dangerous entity. They are characterised by rapid local extension and aggravation despite antibiotic treatment. A discordance of low/normal leucocyte count and high inflammatory parameters (CRP) may be present at the time of diagnosis. Suspected cases require rapid identification of the toxin and early, comprehensive imaging. Successful treatment relies on surgical interventions with two-agent antimicrobial therapy, and a multidisciplinary approach by health professionals of paediatrics, paediatric surgery, radiology and infectious diseases.

Acknowledgements

We kindly thank Guy Prod’Hom, Katia Jaton-Ogay and Jacques Schrenzel for their support with microbiological assays.

References

1 Ritz N , Curtis N . The role of Panton-Valentine leukocidin in Staphylococcus aureus musculoskeletal infections in children. Pediatr Infect Dis J. 2012;31(5):514–8. doi:.https://doi.org/10.1097/INF.0b013e31824f18cb  

2 Gijón M , Bellusci M , Petraitiene B , Noguera-Julian A , Zilinskaite V , Sanchez Moreno P , et al. Factors associated with severity in invasive community-acquired Staphylococcus aureus infections in children: a prospective European multicentre study. Clin Microbiol Infect. 2016;22(7):643.e1–6. doi:.https://doi.org/10.1016/j.cmi.2016.04.004  

3 Shallcross LJ , Fragaszy E , Johnson AM , Hayward AC . The role of the Panton-Valentine leucocidin toxin in staphylococcal disease: a systematic review and meta-analysis. Lancet Infect Dis. 2013;13(1):43–54. doi:.https://doi.org/10.1016/S1473-3099(12)70238-4  

4 Martínez-Aguilar G , Avalos-Mishaan A , Hulten K , Hammerman W , Mason EO, Jr , Kaplan SL . Community-acquired, methicillin-resistant and methicillin-susceptible Staphylococcus aureus musculoskeletal infections in children. Pediatr Infect Dis J. 2004;23(8):701–6. doi:.https://doi.org/10.1097/01.inf.0000133044.79130.2a  

5 Bocchini CE , Hulten KG , Mason EO, Jr , Gonzalez BE , Hammerman WA , Kaplan SL . Panton-Valentine leukocidin genes are associated with enhanced inflammatory response and local disease in acute hematogenous Staphylococcus aureus osteomyelitis in children. Pediatrics. 2006;117(2):433–40. doi:.https://doi.org/10.1542/peds.2005-0566  

6 Sdougkos G , Chini V , Papanastasiou DA , Christodoulou G , Tagaris G , Dimitracopoulos G , et al. Methicillin-resistant Staphylococcus aureus producing Panton-Valentine leukocidin as a cause of acute osteomyelitis in children. Clin Microbiol Infect. 2007;13(6):651–4. doi:.https://doi.org/10.1111/j.1469-0691.2007.01713.x  

7 Dohin B , Gillet Y , Kohler R , Lina G , Vandenesch F , Vanhems P , et al. Pediatric bone and joint infections caused by Panton-Valentine leukocidin-positive Staphylococcus aureus. Pediatr Infect Dis J. 2007;26(11):1042–8. doi:.https://doi.org/10.1097/INF.0b013e318133a85e  

8 Gillet Y , Dohin B , Dumitrescu O , Lina G , Vandenesch F , Etienne J , et al. Infections ostéoarticulaires à staphylocoques dorés sécréteurs de la leucocidine de Panton-Valentine [Osteoarticular infections with staphylococcus aureus secreting Panton-Valentine leucocidin]. Arch Pediatr. 2007;14(Suppl 2):S102–7. Article in French. doi:.https://doi.org/10.1016/S0929-693X(07)80043-1  

9 Cunnington A , Brick T , Cooper M , Danin J , Hunt D , Jeanes A , et al. Severe invasive Panton-Valentine Leucocidin positive Staphylococcus aureus infections in children in London, UK. J Infect. 2009;59(1):28–36. doi:.https://doi.org/10.1016/j.jinf.2009.05.003  

10 Morrey BF , Bianco AJ, Jr , Rhodes KH . Septic arthritis in children. Orthop Clin North Am. 1975;6(4):923–34.  

11 Jaton L , Pillonel T , Jaton K , Dory E , Prod’hom G , Blanc DS , et al. Common skin infection due to Panton-Valentine leucocidin-producing Staphylococcus aureus strains in asylum seekers from Eritrea: a genome-based investigation of a suspected outbreak. Clin Microbiol Infect. 2016;22(8):739.e5–8. doi:.https://doi.org/10.1016/j.cmi.2016.05.026  

12 Mégevand C , Gervaix A , Heininger U , Berger C , Aebi C , Vaudaux B , et al.; Paediatric Infectious Disease Group Switzerland Staphylococcus aureus Study Group. Molecular epidemiology of the nasal colonization by methicillin-susceptible Staphylococcus aureus in Swiss children. Clin Microbiol Infect. 2010;16(9):1414–20. doi:.https://doi.org/10.1111/j.1469-0691.2009.03090.x  

13 Petrovic-Jeremic L , Kuljic-Kapulica N , Ristanovic E , Josic D , Lepsanovic Z . Prevalence of Panton-Valentine leukocidin genes in community-associated methicillin-resistant Staphylococcus aureus in the District of Pomoravlje. Vojnosanit Pregl. 2016;73(3):256–60. doi:.https://doi.org/10.2298/VSP140715003P  

14 Gillet-Vittori L , Afanetti M , Dupont A , Gondon E , Dupont D . Infections sévères à Staphylococcus aureus sécréteurs de la leucocidine de Panton–Valentine chez l’enfant : un large spectre de presentations cliniques [Life-threatening Panton-Valentine leukocidin-associated staphylococcal infections in children. A broad spectrum of clinical presentations]. Arch Pediatr. 2014;21(11):1220–5. Article in French. doi:.https://doi.org/10.1016/j.arcped.2014.08.016  

15 Shallcross LJ , Williams K , Hopkins S , Aldridge RW , Johnson AM , Hayward AC . Panton-Valentine leukocidin associated staphylococcal disease: a cross-sectional study at a London hospital, England. Clin Microbiol Infect. 2010;16(11):1644–8. doi:.https://doi.org/10.1111/j.1469-0691.2010.03153.x  

16 Boyle-Vavra S , Daum RS . Community-acquired methicillin-resistant Staphylococcus aureus: the role of Panton-Valentine leukocidin. Lab Invest. 2007;87(1):3–9. doi:.https://doi.org/10.1038/labinvest.3700501  

17 Couvé-Deacon E , Tristan A , Pestourie N , Faure C , Doffoel-Hantz V , Garnier F , et al. Outbreak of Panton-Valentine Leukocidin-Associated Methicillin-Susceptible Staphylococcus aureus Infection in a Rugby Team, France, 2010-2011. Emerg Infect Dis. 2016;22(1):96–9. doi:.https://doi.org/10.3201/eid2201.150597  

18 Boubaker K , Diebold P , Blanc DS , Vandenesch F , Praz G , Dupuis G , et al. Panton-valentine leukocidin and staphyloccoccal skin infections in schoolchildren. Emerg Infect Dis. 2004;10(1):121–4. doi:.https://doi.org/10.3201/eid1001.030144  

19 Gillet Y , Dumitrescu O , Tristan A , Dauwalder O , Javouhey E , Floret D , et al. Pragmatic management of Panton-Valentine leukocidin-associated staphylococcal diseases. Int J Antimicrob Agents. 2011;38(6):457–64. doi:.https://doi.org/10.1016/j.ijantimicag.2011.05.003  

20 Dumitrescu O , Boisset S , Badiou C , Bes M , Benito Y , Reverdy ME , et al. Effect of antibiotics on Staphylococcus aureus producing Panton-Valentine leukocidin. Antimicrob Agents Chemother. 2007;51(4):1515–9. doi:.https://doi.org/10.1128/AAC.01201-06  

21 Dumitrescu O , Badiou C , Bes M , Reverdy ME , Vandenesch F , Etienne J , et al. Effect of antibiotics, alone and in combination, on Panton-Valentine leukocidin production by a Staphylococcus aureus reference strain. Clin Microbiol Infect. 2008;14(4):384–8. doi:.https://doi.org/10.1111/j.1469-0691.2007.01947.x