MARKUS PÄÄKKÖNEN

Simplified Treatment of

Childhood

Acute Bone and Joint Infections

ACADEMIC DISSERTATION

University of Helsinki

Helsinki 2011

MARKUS PÄÄKKÖNEN

Simplified Treatment of Childhood

Acute Bone and Joint Infections

Hospital for Children and Adolescents

University of Helsinki

Finland

ACADEMIC DISSERTATION

To be presented with the permission of the Medical Faculty of the University

of Helsinki, for public examination in the Niilo Hallman lecture hall, Hospital

for Children and Adolescents, Helsinki University Central Hospital on April

15th, 2011, at 12 noon.

Helsinki 2011

2

Supervised by Professor Heikki Peltola, MD, DTM& H

Hospital for Children and Adolescents

Helsinki University Central Hospital

and

Docent Markku Kallio, MD

Hospital for Children and Adolescents

Helsinki University Central Hospital

Reviewed by Professor Olli Ruuskanen, MD

Department of Paediatrics

Turku University Hospital

and

Docent Jari Peltonen, MD

Hospital for Children and Adolescents

Helsinki University Central Hospital

Official opponent Docent Yrjänä Nietosvaara, MD

Hospital for Children and Adolescents

Helsinki University Central Hospital

ISBN 978-952-92-8338-5 (print)

ISBN 978-952-10-6730-3 (pdf)

http://ethesis.helsinki.fi

Helsinki 2011

Yliopistopaino

3

This book is dedicated to those countless children in non-privileged countries who

succumbed to or became disabled due to septic bone and joint infections during

the time this work was undertaken. Children, who could have been saved but were

not.

4

CONTENTS

ORIGINAL ARTICLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

ABBREVIATIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

ABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

REVIEW OF LITERATURE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

PURPOSE. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

PATIENTS AND METHODS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

RESULTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

CONCLUSIONS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

ACKNOWLEDGEMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

ORIGINAL PUBLICATIONS (reprints) . . . . . . . . . . . . . . . . . . . . . 82

5

ORIGINAL ARTICLES

The thesis is based upon the following original publications, referred to in the text

by roman numerals (I-V).

I Pääkkönen M, Kallio MJ, Kallio PE, Peltola H. Sensitivity of erythrocyte

sedimentation rate and C-reactive protein in childhood bone and joint infections.

Clin Orthop 2010; 468: 861-6.

II Pääkkönen M, Kallio MJ, Peltola H, Kallio PE. Pediatric septic hip with or

without arthrotomy: retrospective analysis of 62 consecutive nonneonatal culture-

positive cases. J Pediatr Orthop B 2010; 19: 264-9.

III Pääkkönen M, Kallio MJ, Kallio PE, Peltola H. Antimicrobial treatment for

childhood osteomyelitis complicated by adjacent septic arthritis. (Submitted)

IV Peltola H, Pääkkönen M, Kallio PE, Kallio MJ. Prospective, randomized trial

of 10 days versus 30 days of antimicrobial treatment, including a short-term

course of parenteral therapy, for childhood septic arthritis. Clin Infect Dis 2009;

48: 1201-10.

V Peltola H, Pääkkönen M, Kallio PE, Kallio MJ. Short- versus long-term

antimicrobial treatment for acute hematogenous osteomyelitis of childhood:

prospective, randomized trial on 131 culture-positive cases. Pediatr Infect Dis J

2010; 29: 1123-8.

Original communications are reproduced with permission of the following

journals: Clinical Orthopaedics and Related Research, Journal of Pediatric

Orthopaedics B, Clinical Infectious Diseases and Pediatric Infectious Disease

Journal.

6

ABBREVIATIONS

ca-MRSA community acquired methicillin-resistant S. aureus

CI confidence interval

CRP C-reactive protein

CT computerized tomography

ESR erythrocyte sedimentation rate

Hib Haemophilus influenzae type b

IQR interquartile range

iv intravenous

mg/L milligrams per liter

/mm3 per cubic millimeter

mm/h millimeters per hour

MRI magnetic resonance imaging

OM osteomyelitis

OM+SA osteomyelitis with adjacent septic arthritis

po peroral

qid quater in die, four times a day

SA septic arthritis

S. aureus Staphylococcus aureus

SD standard deviation

SEM standard error of mean

S. pneumoniae Streptococcus pneumoniae

S. pyogenes Streptococcus pyogenes

US ultrasound

WBC white blood cell count

7

ABSTRACT

Acute childhood osteomyelitis (OM), septic arthritis (SA), and their combination

osteomyelitis with adjacent septic arthritis (OM+SA), are treated with long

courses of antimicrobials and immediate surgery.

We conducted a prospective multi-center randomized trial among

Finnish children at age 3 months to 15 years in 1983-2005. According to the two-

by-two factorial study design, children with OM or OM+SA received 20 or 30

days of antimicrobials, whereas those with SA were treated for 10 or 30 days. In

addition, the whole series was randomized to be treated with clindamycin or a

first-generation cephalosporin. Cases were included only if the causative agent

was isolated. The treatment was instituted intravenously, but only for the first 2-4

days. Percutaneous aspiration was done to obtain a representative sample for

bacteriology, but all other surgical intervention was kept at a minimum.

A total of 265 patients fulfilled our strict inclusion criteria and were

analyzed; 106 children had OM, 134 SA, and 25 OM+SA. In the OM group, one

child in the long and one child in the short-term treatment group developed

sequelae. One child with SA twice developed a late re-infection of the same joint,

but the causative agents differed.

Regarding surgery, diagnostic arthrocentesis or corticotomy was the

only surgical procedure performed in most cases. Routine arthrotomy was not

required even in hip arthritis. Serum C-reactive protein (CRP) proved to be a

reliable laboratory index in the diagnosis and monitoring of osteoarticular

infections. The recovery rate was similar regardless of whether clindamycin or a

first-generation cephalosporin was used.

We conclude that a course of 20 days of these well-absorbing

antimicrobials is sufficient for OM or OM+SA, and 10 days for SA in most cases

beyond the neonatal age. A short intravenous phase of only 2-5 days often

suffices. CRP gives valuable information in monitoring the course of illness.

Besides diagnostic aspiration, surgery should be reserved for selected cases.

8

INTRODUCTION

Acute osteomyelitis (OM) is an acute infection of the bone or bone marrow (Harik

and Smeltzer 2010). Septic arthritis (SA) is a purulent infection of the joint. The

combination of these two entities is called osteomyelitis with septic arthritis

(OM+SA). The disease is defined to be acute if the symptoms of OM, SA or OM

+ SA have lasted for less than two weeks. If the history is longer than two weeks

the disease is defined to be subacute or – in case the duration is several months –

chronic (Krogstad 2004, Harik and Smeltzer 2010).

Acute bone and joint infections were once common killer diseases

(Smith 1874, Toziano et al. 1993). Under normal circumstances bone and joint

resist infections fairly well, but predisposing illnesses or environmental factors –

often associated with poverty- greatly increase the infectious disease burden

(Khan and Bhutta 2010). Thus today bone and joint infections are rare in high

income countries with generally healthy children and efficient health care

(Geirsson et al. 2008). This is not the case in most other parts of the word, as OM,

SA and OM+SA are rampant in many developing countries due to many

underlying illnesses and malnourishment affecting children (Tekou 2000, Ntsiba

2006).

We launched a prospective study in Finland in the early 1980s in

hope that the treatment for these potentially devastating infections could be

simplified from the traditional months-long medication and aggressive surgery.

The goal was to shorten the duration of antimicrobials, and hence, to reduce the

costs of treatment. Furthermore, unnecessary surgery was to be avoided. The

rationale behind our aim stemmed from realizing that unfortunately medical

resources are scarcest in settings where the need is greatest (Goudge 2009).

9

REVIEW OF LITERATURE

1. Definition of terms

“Osteomyelitis” (OM) and “septic arthritis” (SA) are taken here as acute,

hematogenous diseases of children at age 3 months to 15 years (if not otherwise

mentioned). “Osteomyelitis with adjacent septic arthritis” (OM+SA) is a

combination of these two entities. Neonatal infections, OM and SA among adults,

infections caused by Mycobacterium tuberculosis, and chronic OM are not

primarily discussed in this context.

2. History

OM and SA were likely rare diseases among ancient populations, and affected

mainly adult population (Vuorinen 2002). In the late 19th century acute

osteomyelitis was reported to be a common illness among adults in Finland

(Spoof 1885, Roos 1894). In modern days, OM and SA primarily affect children,

and no explanation exists for this difference (Vuorinen 2002). Effective

treatments for osteoarticular infections followed the invention of modern

anesthetics – primarily ether and chloroform – in the 1840s, with the support of

antiseptics during the latter part of the 19th century; now the surgeon could drain

the infectious focus of a bone or joint (Tröhler 1993). Regarding surgery one must

also mention the work of the great anatomist Andreas Vesalius, who already in

1543 published the exact anatomy of the musculoskeletal system in his book “De

humani corporis fabrica libri septem”. His work had an immense impact on the

development of orthopedic surgery in the subsequent centuries (Forsius 1996).

The pathogenesis of osteoarticular infections became better

understood after the work of such pioneers of modern bacteriology as Louis

Pasteur (1822-1895) and Robert Koch (1843-1910) (Sigerist 1931). Pasteur was

among the first to isolate S. pneumoniae, a common causative agent in these

infections. Even a more common organism is Staphyloccus aureus whose

discovery is attributed to Sir Alexander Ogston (1844-1929) from Scotland. He

10

viewed pus under a microscope and took advantage of Joseph Lister’s (1827-

1912) earlier work on antiseptics (Smith 1965). These pioneers paved way to

more accurate diagnostics and understanding of these diseases, but still, OM and

SA continued to be mortal diseases (Smith 1874, Mynter 1889).

A dramatic change occurred once sulphonamides and penicillin were

discovered by teams led by Gerhard Domagk (1895-1964) (Silverman 1944) and

Alexander Fleming (1881-1955), respectively (Kranz 1974, Fleming et al. 1970).

Finally, clinicians had effective medications to combat these severe infections

(Tiitinen 1944, Sulamaa 1945). Mortality and morbidity were significantly

reduced (Koistinen 1956, Torppi and Uurasmaa 1962). One important new 20th

century surgical treatment was the introduction of cancellous bone chip grafts for

the treatment of chronic OM (Mowlem 1944).

3. Pathogenesis and epidemiology

3.1. Pathogenesis

Bacteria may reach a bone or a joint by three routes: direct inoculation following a

trauma or a surgical intervention, local invasion (from for example cellulitis), or

hematogenically. The spread via blood is the most common route in children

(Krogstad 2004, Conrad 2010).

There are essentially two distinct types of bone: compact dense bone

and soft cancellous bone. The distinguishing feature in the growing skeleton of a

child is the presence of epiphyseal growth plates. Immediately adjacent to the

epiphysis is the “resting cell zone”, which supplies the developing cartilage cells.

Next, in the “zone of proliferating cartilage” bone length is created by active

growth. In the “maturation zone” the chondrocytes differentiate and calcification

occurs. Similarly, a vertebral body grows in height from the superior and inferior

endplates of the vertebrae, these are comparable to the growth plates of long

bones (Doskocil et al. 1993, Krogstad 2004, Conrad 2010).

OM usually begins in the metaphysis of a long tubular bone as

shown in Figure 1. As there is no direct blood flow from the metaphysis to the

11

epiphysis, the proliferative epifyseal chondrocytes receive nutrition mainly from

the epiphyseal arteries. All parts of the growth area, epiphysis, growth plate,

metaphysis and perichondrium, have their own distinct blood supply (Krogstad

2004, Conrad 2010). Hence, the growth plate functions as a barrier against

infection. On the other hand, joints where the metaphysis is intracapsular such as

hip, shoulder and ankle are thought to be especially at risk for infection (Jackson

et al. 1992, Perlman et al. 2000). The synovial membrane of the joint has

abundant blood flow, and large numbers of bacteria may enter the synovial

membrane during bacteremia (Krogstad 2004). Most of these episodes are

transient, but occasionally and largely for unknown reasons, bacteria stick onto a

particular site and launch an infection which, in turn, triggers massive

inflammatory reaction in the host. S. aureus is able to attach to the type I collagen

fibrils and establishes microcolonies surrounded by a glycocalyx (Zavin et al.

1993, Krogstad 2004).

Burn injuries may also cause hematogenous septic seeding leading

to OM (Fodor et al. 2004) or SA (Smoot et al. 1993).

Figure 1. Slow blood flow in the sinusoidal arteries of the metaphyseal-

epiphyseal junction predisposes children to hematogenous osteomyelitis. The

infection is initiated by bacteremic seeding.

12

3.2. Causative agents

3.2.1. Staphylococcus aureus

Gram-positive S. aureus - literally “golden cluster seed” due to the carotenoid

pigment staphyloxanthin (Marshall and Wilmoth 1981) - is the most common

causative organism both in OM and SA (Krogstad 2004). Besides osteoarticular

infections it causes a wide range of diseases such as skin infections (Frei et al.

2010), pneumonia (Li et al. 2010), meningitis (Aguilar et al. 2010), endocarditis

(Slabbekoorn et al. 2010) and wound infections (Suljagic et al. 2010). S. aureus is

part of the normal flora of skin and is also involved in impetigo, pimples, boils

and abscesses. Depending on the strain, S.aureus is capable of secreting several

toxins causing toxic shock syndrome (Kare and Dang 2008). S. aureus has

proven its capability to develop antimicrobial resistance. After penicillin

resistance had developed already in the 1940s (Plough 1945) methicillin has been

used as an anti-staphylococcal agent in many countries. The evolution and spread

of methicillin-resistant (ca-MRSA) and vancomycin-resistant Staphyloccus aureus

in the past decade has caused grave concern (Rehms and Tice 2010).

3.2.2. Streptococcus pyogenes

This spherical Gram-positive bacterium grows in long chains and displays

streptococcal group A antigen in the cell wall. S. pyogenes typically produces

large zones of beta-hemolysis, hence the name group A beta-hemolytic

streptococcus. It is common in upper respiratory tract infections, especially in

acute tonsillitis (Altamimi et al. 2010). S. pyogenes is part of the normal flora of

the skin (Brown et al. 1975) and is a common causative agent of erysipelas

(Berhard 2008). S. pyogenes remains sensitive to penicillin (Tempera et al. 2010).

3.2.3. Streptococcus pneumoniae

S. pneumoniae a.k.a. pneumococcus is a Gram-positive member of the genus

Streptococcus. Apart from osteoarticular infections and pneumonia (Resti et al.

13

2010) this organism causes a wide variety of respiratory infections such as acute

sinusitis and otitis media. Severe pneumococcal infections include meningitis,

peritonitis (Rueda et al. 2010) and pedicarditis (Wei et al. 2010). While previously

sensitive to penicillin now an emergence of penicillin resistant strains has been

seen as well (Tempera et al. 2010). Penicillin resistance has still fortunately

remained low in Europe (Cooke et al. 2010).

3.2.4. Haemophilus influenzae type b (Hib).

Haemophilus influenzae is a gram-negative rod that used to be a devastating

pathogen causing pneumonia, meningitis (Smythe 1948) and epiglottitis (Margolis

et al. 1972). Thanks to large-scale vaccinations, the Hib etiology of osteoarticular

infections has been eliminated from Finland (Peltola et al. 1998). The same

sequence is likely to repeat once S. pneumoniae vaccination is adopted to the

routine childhood immunization program. Hib is mostly still sensitive to beta-

lactam antibiotics and common antimicrobial used is ampicillin (iv) or amoxicillin

(po). Recent study put the prevalence of beta-lactamase positive strains at 16%

(Perez-Trallero et al. 2010).

3.2.5. Kingella kingae

Kingella kingae, a Gram-negative coccobacillus, is another important pathogen in

SA in some parts of the world (de Groot et al. 1988, Lacour et al. 1991). In Israel,

it can even surpass S. aureus in frequency (Yagupsky et al. 1995). This pathogen

is thought to be underdiagnosed in SA since itʼs culturing is difficult (Host et al.

2000). K. kingae is susceptible to beta-lactams, erythromycin and ciprofloxacin

(Yagupsky et al. 2001).

3.2.6. Others

SA may also be caused by Neisseria meningitidis and Neisseria gonorrhoeae, this

etiology being more characteristic of newborns (Deshpande et al. 1990) and of

14

sexually active adolescents (Krogstad 2004). Salmonella spp. is rather commonly

encountered in the low-income countries (Smith et al. 2002, Al Arfaj 2008,

Scheider et al. 2009).

3.2.7. Regional differences

S. aureus is overwhelmingly most common causative organism in OM all over the

world (Craigen et al. 1992, Jenzri et al. 2008, Kharbanda and Dhir 1991, Riise et

al. 2008.). The most common causative organisms in SA by region are shown in

Table 1.

3.3. Epidemiology

3.3.1. Worldwide distribution

Osteoarticular infections are rather common in the developing world

(Onyemelukwe and Sturruck 1979, Molyneux and French 1982, Smith et al.

2002), but much less frequent in industrialized countries (Grimprel and Cohen

2007). Little information is available of the precise incidence of OM or SA in

Africa since most reports deal only with cases presented at single institutions

without a precisely defined source population (Lavy et al. 2007). Most likely these

infections are much more common in low income countries and in the tropics. A

recent study estimated the prevalence a tropical area to be 2-5 times that of

Europe (Labbe et al. 2010). In a Nepalese study 10% of childrensʼ hospital

admissions were sequelae of musculoskeletal sepsis (Spiegel et al. 2010). In

Finland, the annual incidence of OM or SA at age 0-14 years is less than five per

100.000 (Peltola and Vahvanen 1983, Kunnamo et al. 1986). Articles giving a

current estimate of the incidence of OM or SA that has been published in recent

years are shown in Table 2.

15

3.3.2. Incidence in boys vs girls

OM and SA universally tend to affect males more than females. The reason for

this unequal distribution among genders is not known. The ratio is approximately

2.0 : 1.2 (Krogstad 2004). Interestingly, as seen in Table 3, the ratio is fairly

similar in all over the world.

3.3.3. Concomitant OM +SA

High proportion of children with OM or SA have concomitant bone or joint

involvement. One third (Perlman et al. 2000) or even half the cases (Chen et al.

2010) may have OM+SA.

3.3.4. Predisposing factors

Factors generally contributing to child health also influence the prevalence of OM

and SA. Poverty, lack of purchasing power, household food insecurity and low

education lead to malnourishment and high infectious disease burden (Khan and

Bhutta 2010). Immunodeficiencies predispose to OM or SA. A special

characteristic among immunodeficient patients is the presence of low virulence

infections (Bloom et al. 2008). Blunt trauma preceding osteoarticular infections

has been described in up to 63% of the cases (Labbe et al. 2010). Burns may

precipitate OM. Surgery of a bone or joint always carries a risk of postoperative

infection.

3.3.5. Recent changes

Iatrogenic SA, due to increasing use of arthroscopy, seems to have slightly

increased in frequency in the affluent countries (Geirsson et al. 2008). Most

troubling development in the rise of community acquired methicillin resistant

Staphylococcus aureus. Current rate of ca-MRSA varies around the world, with a

16

rate of 24% published in Hawaii (Erdem et al. 2010), or even 74% in Taiwan

(Fang et al. 2004).

Table 1.

Most common causative organisms in septic arthritis by region

Country Most common bacteria

(%)

Reference

Finland S. aureus 74%

S. pyogenes 15%

Peltola et al. 1998

France S. aureus 44%

Kingella kingae 14%

Moumile et al. 2005

Israel S. aureus 35%

Streptococci 14%

Eder et al. 2005

Saudi Arabia S. aureus 39% Al Saadi et al. 2009

Taiwan S. aureus 43%

Streptococci 10%

Wang et al. 2003

United Kingdom S. aureus 41%

Streptococci 28%

Ryan et al. 1997

17

Table 2.

Incidence pediatric of OM or SA by region among (per year)

Country OM or SA Incidence/100 000 Reference

Australia SA 9 Morgan et al. 1996

Iceland SA 11 Geirsson et al. 2007

Finland SA 4 Peltola and Vahvanen 1983

Norway SA 4 Riise et al. 2008

Norway OM 13 Riise et al. 2008.

Norway OM 10 Dahl et al. 1998

Malawi SA 20 (children <5years) Lavy et al. 2005

Saudi Arabia SA 2 Al Arfaj 2008

United Kingdom OM 4 Craigen et al. 1992

Table 3.

Boys vs girls affected by OM or SA by country

Country Boys : Girls Note Reference

Finland 2.5 : 1 SA Peltola and Vahvanen 1983

India 2.7 : 1 OM Kharbanda and Dhir 1991

Italy 1.9 : 1 OM, SA Faesch et al. 2009

Malawi 2.0 : 1 SA Lavy and Thyoka 2005

Nigeria 3.0: 1 OM Tindimwebwa et al. 1983

Tunisia 1.4 : 1 Calcaneal OM Jenzri et al. 2008

Turkey 1.2 : 1 SA Caksen et al. 2000

Zambia 2.2 : 1 Salmonella SA Lavy et al. 1996

18

4. Diagnostics

4.1. Classic OM, SA and OM+SA

A child with a typical OM is clinically less ill than a child with SA. The patient

may have fever and refuse to walk. In a closer examination the anatomic region is

found to be warm, tender and hyperemic as seen in Figure 2. The most common

localization is around the metaphyseal area of a long bone, but may be anywhere

in the osseal structures. Even rare cases of primary patellar osteomyelitis have

been recorded (Waris 1938). Sometimes the child presents only with fever,

malaise and refusal to eat (Krogstad 2004).

Figure 2. Clinical presentation of calcaneal osteomyelitis can be deceptively

inconspicuous. Picture courtesy of Dr. Heikki Peltola.

19

In contrast, a patient with SA is usually more ill than a patient with

OM. Fever is often high, and the affected joint is swollen, warm and tender.

Motion causes considerable pain. The history reveals no prior trauma (Shaw et al.

1990). SA usually only affects one joint (Krogstad 2004), but septicemic seeding

may also initiate the infection in several joints simultaneously, an entity called

septic multiple joint involvement. The slogan “mono means gono” stems from

times when gonorrhea used to be a common etiology.

Once suspicion of an osteoarticular infection has arisen, the

diagnosis should be confirmed. Prompt corticotomy (drilling) in OM, aspiration of

pus in case of a periosteal abscess and an aspiration of the joint in SA are methods

to verify the clinical entity and to obtain a representative sample for bacteriology

(Tetzlaff 1978). Both Gram-staining and culture should be performed (Brannan

and Jerrard 2006). Traditionally, synovial fluid has been scrutinized by a

microscope, but for example, the leucosyte count overlaps with other types of

arthritis in such a manner (Kunnamo et al. 1987, Kortekangas et al. 1992) that its

usefulness is questionable. In contrast, the blood culture may be positive although

the local sample remains negative.

4.2. Differential diagnostics

Options to be considered in the differential diagnosis of OM or SA are listed in

Table 4. A bone or soft tissue tumor or dysplastic change within bone may show

symptoms suggesting OM (Sluga et al. 2002, Gomoll 2007) or SA (Jordanov et al.

2009). Ultrasound or magnetic resonance imaging (MRI) are valuable in this

differential diagnosis (Sluga et al. 2002). Computerized tomography (CT) is also

informative (Ma et al. 1997). A trauma, overuse osteopathy or a tress-fracture

may show symptoms compatible with OM (Roth et al. 2008). Pyomyositis, a

common disease in the tropics but rare in Nordic countries may also resemble OM

in such a way, that the differential diagnosis may be extremely difficult (Abdullah

et al. 2010). Child abuse may be a possible cause of musculoskeletal symptoms.

In SA, the most common likely confounding diagnoses are non-

bacterial arthritides such as reactive arthritis and juvenile idiopathic arthritis

20

(Manimegalai 2008, Solari 2008). In case of coxitis, transient synovitis of the hip

is the first entity to be distinguished from SA (Kocher et al. 1999). A number of

viruses may cause joint manifestations; among those are infections caused by

sindbis (Kurkela et al. 2008 ), chikungunya (Sebastian et al. 2009), rubella and

parvovirus (Franssila and Hedman 2006).

Systemic diseases such as sickle cell anemia for example may also

present musculoskeletal symptoms that may resemble OM or SA (Balogun et al.

2010). Conditions such as growing pains rarely resemble OM in children, as the

child moves normally and the symptoms appear periodically (Asadi-Pooya and

Bordbar 2007).

Table 4.

Diseases to be considered in the differential diagnosis OM or SA.

OSTEOMYELITIS SEPTIC ARTHRITIS

Trauma Aseptic arthritides

Stress fracture Transient synovitis of the hip

Pyomyositis Legg-Calve-Perthes disease

Cellulitis Slipped capital femoral epiphysis

Benign/malignant tumours Viral arthritis

Acute leukemia Juvenile rheumatoid arthritis

Child abuse Reactive arthritis

Scurvy Sickle cell anemia

4.3. Laboratory measurements in the diagnostics of osteoarticular infections

The erythrocyte sedimentation rate (ESR) is the traditional laboratory marker used

in the diagnostics of osteoarticular infections (Lorrot 2007). ESR increases in 1-3

21

days, but normalization takes several weeks even in case of an uneventful

recovery (Peltola and Räsänen 1982, Peltola and Vahvanen 1984, Unkila-Kallio

1994 [1], Unkila-Kallio 1994 [2], Kallio et al. 1997, Lorrot 2007).

Finnish researchers were probably the first to utilize serum C-

reactive protein (CRP) in the diagnostics and follow-up of osteoarticular

infections (Peltola and Räsänen 1982, Peltola and Vahvanen 1984; Unkila-Kallio

et al. 1994 [1], Unkila-Kallio et al. 1994 [2], Kallio et al. 1997).

White blood cell count (WBC) is too unspecific a marker to be of

value in a clinical setting, especially since it may be normal in up to 80% of cases

(Lorrot 2007). Recently, serum procalcitonin concentrations have been used in the

differentiation of OM and SA (Butbul-Aviel et al. 2005, Hugle et al. 2008), but no

study shows it to serve the clinician better than CRP.

4.4. Radiological findings

Plain radiograph has been used in OM almost since the times of Wilhelm Konrad

Röntgen (Smith 1896), because when positive, it is a strong argument for the

diagnosis. The problem is that significant bone matrix destruction is required for a

positive x-ray with leading to sensitivity as low as 15% in an early x-ray (Malcius

et al. 2009). In a Finnish study (Peltola et al. 1997), lytic x-ray changes were

found in only 19% of the cases at the beginning of treatment, but in 68% of the

cases at day 29 from the onset of treatment. On the other hand, plain radiograph

can be used to rule out other possible causes of joint pain or limp, such as

traumatic/stress fractures or slipped upper femoral epiphysis (Reed 2009).

Scintigraphy a.k.a bone scan and more recently magnetic resonance

imaging (MRI) have paved their ways to be used in diagnostics as seen in Figure

3, albeit they too, might be negative in an early stage of the disease (Hughes and

Aronson 1994, Jenzri et al. 2008). Especially during the first 3 days a bone scan

may yield a false negative result (Conrad 2010), but still changes to positive much

earlier than the traditional radiograph (Azoulay et al. 2007).

Apart from differential diagnosis a plain radiograph is of little

relevance in recognizing SA, because if anything it usually only identifies a

22

swollen joint space. Ultrasound is slightly better for detecting swelling and

facilitating aspiration (Lavy et al. 2003). Good accuracy has been demonstrated

with MRI which also adds to the difficult differential diagnosis between OM and

SA (Hammond and Macnicol 2001). Fat-enhancement or using gadolium contrast

does not improve accuracy of MRI (Averill et al. 2009, Kan et al. 2010).

Computer tomography (CT) has few indications, but it may be performed when a

sequestrum in chronic osteomyelitis is suspected, or when MRI is not possible.

In the routine follow-up a plain x-ray is often sufficient as it reveals

a pathological fracture or chronic processes sequestrum. In a recent study routine

repeat MRI did not offer additional benefit (Courtney et al. 2010). However, if the

recovery is slow or absent, a repeat MRI can affect treatment in 10 % of the cases

(Courtney et al. 2010).

Figure 3. Previously healthy 15-year-old boy suffered a mild distorsion injury of

the hip while playing ice hockey. In the following days the patient developed

fever and pain in the left sacro-iliacal region. US suggested sacroilitis. MRI

showed an abscess of 23 millimeters adjacent to the sacro-iliacal joint. S. aureus

grew from the blood culture.

23

5. Treatment

5.1. Antimicrobial therapy, and other medications

Antimicrobial treatment for osteoarticular infections should be started without

delay after the relevant samples have been obtained. Because no sufficiently-

powered study comparing different antimicrobials or examining dissimilar

durations of treatments has been available, clinicians have been forced to base

their choice on the expertsʼ recommendations and own experience. Sometimes

the initial treatment requires an adjustment according to the causative agent

(Krogstad 2004).

The primary antimicrobial should cover the most common

organisms. In high-income countries these are essentially gram-positive bacteria

(Krogstad 2004, Shaw and Kasser 1990, Smith et al. 1995) since Hib has been

eliminated by vaccination (Peltola et al. 1997). The agent must have good joint

and bone penetration (Nelson et al. 1978). Clindamycin and first-generation

cephalosporins fulfill these criteria. Large doses were recommended decades ago

and probably with good reasoning as these antimicrobials are time dependent and

thus their efficacy depends on the time their tissue concentration exceeds the

minimal inhibitory concentration (Nelson et al. 1978). For clindamycin, 40 mg per

kilogram per 24 hours should be administered divided in 4 daily doses (qid), and

for the first-generation cephalosporins no less than 150 (-200) mg per kilogram

qid. Because beta-lactams and lincosamides (clindamycin) are time-dependent

antimicrobials, it is probably important that administration occurs four, not three,

times, per 24 hours. Antimicrobial recommendations vary greatly in different

regions, as seen in Table 5.

24

Table 5.

Recommended empiric antimicrobial treatments vary greatly between

countries or even between hospitals. A list of some recommended primary

antimicrobial agents in osteoarticular infections by country.

Country Antimicrobial

recommendation

Reference

Chile Cloxacillin + Ceftriaxone (age<5)

Cloxacillin+Aminoglycoside (age>5) Prado et al. 2008

France Beta –lactam Cohen and Grimprel 2007

Malawi Chloramphenicol Peek et al. 2006

United Kingdom Flucloxacillin or Cefuroxime Jagodzinski et al. 2009

United States Beta-lactam + Vancomycin Saphyakhajon et al. 2008

The optimal duration of antimicrobials is under continuous

discussion. Many authorities opine that the initial intravenous phase should last

for at least a week (Krogstad 2004, Glorion et al. 1993, Ross et al. 2003). Then, a

shift to oral administration may be possible, circumstances permitting (Feigin et

al. 1975, Nelson et al. 1978, Tetzlaff et al. 1978, Kolyvas et al. 1980, Toziano et

al 1993). The total course, both for OM and SA, is globally recommended to last

several weeks as shown in Table 6 (Mollam and Biggott 1977, Syrogiannopoulos

and Nelson 1988, Syrogiannopoulos and Nelson 1988, Shaw et al. 1990, Smith an

Piercy 1995, Dahl et al. 1998, Christiansen et al. 1999, Bonhoeffer et al. 2001,

Lew and Waldvogel 2004, Malcius 2005, Krilov and McCracken 2006), even

though the risks involved with prolonged treatments have been documented

(Ceroni et al. 2003). Retrospective studies have shown early transition from

intravenous to oral therapy to be safe even in OM (Zaoutis et al. 2009).

One study on SA examined the value of adjuvant dexamethazone

(Odio 2001). It slightly quickened healing but did not influence long-term

outcomes. Non-steroidal anti-inflammatory drugs to relieve pain and fever should

be given at the discretion of the attending clinician (Autrec-Leca 2003).

25

Table 6.

Recommended length of total antimicrobial treatment of pediatric OMSA by

country

Country Treatment length Reference

Australia 3 weeks Vinod et al. 2002

Chile 4 - 6 weeks Prado et al. 2010

France 5 - 6 weeks Abuamara et al. 2004

India 3 - 4 weeks Shetty and Gedalia 2004

Iran 5 weeks Shetty and Gedalia 2004

Nigeria 2 - 4 weeks Eyichukwu et al. 2010

Taiwan 5 weeks Kao et al. 2003

United States 4 weeks Ballock et al. 2010

5.2. Surgical treatment of OM

Traditional surgical treatment of OM comprises drainage of the subperiosteal

abscess by drilling of the bone (trepanation). If adjacent joint is involved,

capsulotomy or continuous lavage is often performed (Tetzlaff et al. 1978,

Kolyvas et al. 1980, Karwowska et al. 1998, Vinod et al. 2002, Gutierrez 2005).

Since no high-quality studies have been available, such active surgical

intervention is largely empirical, or based on retrospective analyses of variable

patient series (Kenney 1944).

26

5.3. Surgical treatment of SA

In SA, the traditional surgical treatment includes prompt surgical drainage of the

joint (Welkon 1986, Parsch 1997, Gordon 2005, Nunn 2007). This has been

deemed especially important for the hip and shoulder joints because of the tight

surrounding capsule (Krogstad 2004). Although no study has documented the

need for routine arthrotomy, the rationale is to drain pus, to decrease intra-

articular pressure, and thus, to decrease the risk of avascular necrosis of the

femoral head, or prevent the pressure induced dislocation of the hip joint (Hunka

1982, Vidigal 1997). In recent years, researchers have also challenged this view

(Lavy et al. 2006). Interestingly, SA in an adult patient is more often treated with

arthrotomy/arthroscopy and lavage if the attending physician is a surgeon; if not,

the patient is more often treated with daily arthrocentesis (Manadan and Block

2004). The results have not differed.

Some clinicians have moved to repeated (ultrasound guided)

aspirations (Griffet and Hayek 1996, Lavy et al. 2003, Manadan and Block 2004,

Smith et al. 2002, Givon 2004), or conservative treatment which consists only of

antimicrobials and possibly immobilization of the affected joint (Lipczyk 2001).

The results have equaled to those deriving from the traditional approach.

5.4. Treatment of OM+SA

Although there is no consensus on the treatment modalities for OM+SA (Tetzlaff

et al. 1978, Kolyvas et al. 1980, Karwowska et al. 1998, Carek et al. 2001,

Dieckmann et al. 2008, Weichert et al. 2008), the approach is traditionally more

aggressive than for plain OM or SA (Hartwig 2006, Shetty and Kumar 2007)

because OM+SA is considered to have a poorer outcome (Gafur et al. 2008, Jenzri

et al. 2008). Since the differential diagnosis between OM and OM+SA may be

difficult (Hammond and Macnicol 2001), especially at an early stage of disease,

the clinician often has to treat the patient without precise knowledge of the extent

of the infection.

27

5.5. Monitoring of the treatment

Measuring the effect of antimicrobials and/or surgical treatments warrants

repeated clinical examination, combined with a follow-up of the parameters of

infection/inflammation. ESR and WBC are the traditional markers used (Gold

1991, Del Beccaro et al. 1992). Being an excellent tool for monitoring response

to a chosen treatment, serum CRP has challenged both these laboratory indices

(Peltola and Räsänen 1982, Peltola and Vahvanen 1984, Unkila-Kallio et al. 1994,

Andreola et al. 2007). A special advantage of CRP is its rapid reaction to changes

in the clinical situation compared to ESR (Unkila-Kallio 1994, Andreola 2007).

Furthermore, a precise CRP value is currently obtainable quickly (in 10 minutes)

from a whole-blood sample taken with a finger prick. CRP costs 1.40-6.00 € and

is thus much cheaper than procalcitonin measurement that costs 20.00-30.00 €

(Tykslab 2010).

5.6. Prevention of osteoarticular infections

Before the era of conjugate vaccines, Hib used to be the second most common

causative agent of SA in Finland. Large-scale vaccinations wiped out this

etiology, and as a consequence, the initial antimicrobial spectrum could be

narrowed (Peltola et al. 1998). Conjugate vaccines against S. pneumoniae can be

expected to decrease the incidence of osteoarticular infections caused by

pneumococci (McClure et al. 2006), although its role is smaller than what Hib

used to be. So far, there is no effective vaccine against S. aureus (Stranger-Jones

et al. 2006, Clarce et al. 2006).

6. Outcomes

6.1. Natural course of OM and SA

Although large series of various osteoarticular infections have been published

since or even before the beginning of last century (Smith 1874, Homans 1912),

28

their natural course is not well characterized. A study of 110 patients in India

showed that sequelae such as chronic osteomyelitis and sequestrum are fairly

common in OM if antimicrobial treatment has been delayed or not instituted

(Kharbanda and Dhir 1991). In a Finnish study early treatment in under 72 hours

did not correlate with a more favorable outcome compared to children presenting

within a week (Lindell and Parkkulainen 1960).

By experience we have learned that sequelae may develop slowly

(Spindler et al. 1998, Nunn and Rollison 2007), and therefore a long follow-up for

at least a year is needed. If a new episode develops later, it is likely that it is not a

true recrudescence of the same disease but a late re-infection (Uckay et al. 2006)

The risk of recrudescence in OM, SA, or OM+SA is not well

known. A large retrospective study on 168 cases of osteoarticular infections in the

United States from the 1974-83 (Syrogiannopoulos and Nelson 1988) found four

(3.8%) recurrences of OM, and other sources (Cole et al. 1982, Kharbanda and

Dhir 1991, Tindimwebwa et al 1983) suggest an incidence of somewhere around.

However, if a new episode develops later, it is not-at-all clear that the case has

recrudesced. A Swiss report (Uckay et al. 2006) demonstrates that it may only be

a late re-infection in the same anatomical site caused by a different agent, perhaps

a similar phenomenon as observed with endocarditis. Whatever the sequence, a

recrudescence or a late reinfection tend to occur within months, or at least, in the

few years after the primary affection.

6.2. Long-term sequelae

The feared complications of OM include pathological fractures, abscesses,

sequestrum and periosteal necrosis. Acetabulum dysplasia is a severe sequela that

develops after an enlarged femoral head – caput magna – is formed because of

overgrowth of cartilage after SA. Pathological fractures most commonly develop

due to chronic osteomyelitis (Akinyoola et al. 2008), which is unfortunately

common in less priviledged countries (Agaja and Ayorinde 2008) and may

require extensive surgical procedures as seen in Figure 4. Sequestrum is a

segment of bone that has been avascularised by necrosis. Involucrum is the

29

formation of new bone around a sequestrum during the healing process (Jones et

al. 2009).

Growth disturbance is a devastating sequela that may lead to severe

functional disability (Suger et al. 2000, Nunn and Rollison 2007). For example, a

partial closure of the growth plate of a single femoral condyle leads to angular

varus or valgus deformity (Langenskiöld 1984). These deformities may be

asymptomatic even for decades before arthrosis develops. Compartment

syndrome secondary to pediatric OM has also been described (Mulcaney et al.

2009).

A series of 110 patients from India showed that sequelae such as

chronic OM and sequestrum are fairly common if the antimicrobial treatment has

been delayed or not instituted (Kharbanda and Dhir 1991). The sequelae may

develop slowly (Spindler et al. 1998, Nunn and Rollison 2007). Therefore, a long

follow-up for at least a year post-hospitalization is well founded.

Complications of SA include the avascular necrosis of the

subchondral bone or joint cartilage destruction leading to arthrosis (Hunka et al.

1982, Vidigal Junior EC et al. 1997). Even death may occur. The length of

antimicrobial administration needed to avoid recrudescences or sequlae – if such

exists – is not known (Krogstad 2004). Reported global complication rates in OM

and SA are given in Table 7ab. Articles often use different criteria for reporting

complications making definitive comparisons between different regions and

patient populations difficult.

30

Figure 4. Hematogenous osteomyelitis is the most common cause of chronic

osteomyelitis, a devastating and sometimes lifelong disease that is unfortunately

common in impoverished countries (Agaja and Ayorinde 2008). A Kenyan boy

required extensive surgical debridement because of a chronic infection focus of

the right femur. Courtesy of Dr. Jukka Tiittanen.

31

Table 7a

Outcome of OM by country.

Country Sequelae/Cases (%) Reference

Australia 1/75 (1) Cole et al. 1982

India 32/110 (29) Kharbanda and Dhir 1991

Nigeria 3/28 (11) Tindimwebwa et al. 1983

Russia 6/348 (2) Rusak et al. 1991

Tunisia 7/26 (27) Jenzri et al. 2008

United States 1/77 (1) OʼBrien et al. 1982

Table 7b

Outcome of SA by country

Country Sequelae/Cases (%) Reference

Australia 1/75 (1) Cole et al. 1982

India 32/110 (29) Kharbanda and Dhir 1991

Nigeria 3/28 (11) Tindimwebwa et al. 1983

Russia 6/348 (2) Rusak et al. 1991

Tunisia 7/26 (27) Jenzri et al. 2008

United States 1/77 (1) OʼBrien et al. 1982

32

THE PURPOSE OF THE STUDY

1. To determine if the duration of antimicrobial treatment for OM/OM+SA

could be shortened to 20 days (V).

2. To determine if the duration of antimicrobial treatment for SA could be

shortened to 10 days (IV).

3. To compare clindamycin with first generation cephalosporin in the

treatment for OM, SA or OMSA (IV, V).

4. To assess the need for surgery in OM, SA or OM+ SA (II, III, IV, V).

5. To evaluate the usefulness of serum CRP determinations in the diagnostics

of and monitoring the course of acute osteoarticular infections of childhood

(I, II, III, IV, V).

6. To estimate to which extent concurrent adjacent SA complicates the

treatment of OM (III).

33

PATIENTS AND METHODS

1. Study design

The study was a randomized controlled, multicentre, open labeled, parallel group,

non-inferiority trial carried out in 7 referral hospitals (Aurora Hospital, Etelä-

Saimaa Central Hospital, Helsinki University Central Hospital, Jorvi Hospital,

Kuopio University Hospital, Päijät-Häme Central Hospital, Seinäjoki Central

Hospital) in Finland in 1983-2005. The study protocol was approved by each

institution’s ethical committee, and the child was included only if consent was

given by the legal guardian. The trial was designed, conducted, and analyzed

independently of any medical companies or manufacturers. Only previously

healthy children were included.

Once a child with OM (painful and swollen limb without trauma,

restriction of motion, often a tender and warm area combined with fever), acute

SA (painful and swollen joint without trauma, restriction of motion, tenderness

and warmth combined with fever) (Shaw and Kasser 1990) or OM+SA (combined

symptoms of OM and SA) was suspected in a 3-month to 15-year-old child, the

clinician obtained a randomized study number by telephone from a computer-

generated list held at Children’s Hospital, Helsinki, 24 hours a day. The number

was marked in the hospital chart, and it dictated the length of the treatment, 20 or

30 days in OM or OM + SA and 10 or 30 days in SA. The antibiotic, clindamycin

or a 1st generation cephalosporin was chosen according to the childʼs birthdays

(even or odd) as seen in Figure 5.

34

Figure 5. Study protocol. Adjuvant ampi-/amoxicillin was also given to children

under 5 years.

35

2. Patients

Patients from age 3 months to 15 years with culture positive OM, SA, or

combined OM+SA were enrolled in the study. 131 children with OM or OM+SA

were randomized between 20 days (67 cases) or 30 days (64 cases) of

antimicrobials. 130 children with SA were randomized to receive the 10-day-

treatment (63 cases) or the 30-day-treatment (67 cases). A quasirandomization

was done between clindamycin and cephalosporin treatments. As four post-

enrollment cases were included in the subanalysis of the surgical treatment and

laboratory monitoring, the final series comprised 265 patients. The patient

characteristics are described in Table 8. Proportion of cases from each

participating hospital is described in Figure 6. Number of cases in proportion to

the pediatric source population of each participating hospital is described in

Figure 7. Average enrollment of enrolled patients in proportion to source

population was 2.4/100 000/year among 0-15 year olds. Localization of OM or

SA in proportion to the whole series is described in Figure 8a. Figure 8b depicts

the rate of adjacent joint involvement in OM by localization. 8c shows the

proportion of OM cases by localization in proportion to all OM cases and SA by

localization in proportion to all SA cases.

Figure 6. Patients (%) from participating hospitals.

36

Table 8. Patient characteristics on admission (N=265)

Gender, Males/Females 161/104 Age, years, median 8.4 History, days (median, IQR) 3 (2-4) Diagnosis Septic arthritis Osteomyelitis Osteomyelitis with adjacent septic arthritis

134 106 25

Agent cultured from Joint and blood Bone and blood Joint only Bone only Blood only*

55 21 75 33 81

Causative agent Staphylococcus aureus Haemophilus influenzae type b S. pyogenes S. pneumoniae Other

199 26 25 12 3

*provided that the patient had an indisputable evidence of an osteoarticular

affection by US, radiograph and/or MRI, or local pus was obtained.

Figure 7. Number of enrolled patients during the study period from each hospital

in proportion to the pediatric source population of the hospital in question

(/100000).

37

Figure 8a. Localization of osteoarticular infections in this series of 265 cases in

all. Skeleton used as a model was originally drawn by the great anatomist,

physician and genius Andreas Vesalius (1514-1564).

38

Figure 8b. Proportion of OM cases complicated by SA by localization. Special

predisposition of the proximal tibia is surprising.

39

Figure 8c. The anatomical site of OM/OM+SA (N=131) and SA (N=130), shown

as percentage of their total numbers, respectively.

40

3. Treatment

3.1. Antimicrobial treatment

Since a gram-positive agent was the likely pathogen (Krogstad 2004, Glorion et

al. 1993, Nelson et al. 1978, Syrogiannopoulos and Nelson 1988, Shaw and

Kasser 1990, Smith and Piercy 1995, Peltola and Vahvanen 1983, Christiansen et

al. 1999), the children were given clindamycin (Feigin et al. 1975, Nicholas et al.

1975, Kaplan et al. 1982) 40 mg per kilogram per 24 hours divided qid (Kolyvas

et al. 1980), or a first-generation cephalosporin (cephradine/cephalothin i.v.,

cephradine/cehalexin/cephadroxil orally) 150 mg per kilogram per 24 hours

divided qid (Kolyvas et al. 1980). In the first years of the study Hib was still

common, the children with SA at age 0-4 years received initially additional

ampicillin (Kolyvas et al. 1980) intravenously, followed by amoxicillin orally,

each 200 mg per kilogram per 24 hours qid until the etiology was identified. The

course was completed with one antimicrobial only.

Being the only first-generation cephalosporin available for

parenteral and oral use, cephradine (Kolyvas et al. 1980, Zaki et al. 1974, Brosof

et al. 1979, Leigh et al. 1989) was our first choice, but its later withdrawal from

Scandinavia forced us to change to cephalothin (iv), and oral cephalexin (Tetzlaff

et al. 1978) or cephadroxil; the doses were the same as for cephradine. This switch

was not deemed critical for the study, because all these cephalosporins are very

similar (Lambert and OʼGrady 1992).

Antimicrobial was instituted intravenously for 2-4 days, and

completed orally with the same high doses. Serum or joint fluid concentrations

were not assayed. Nor was adjuvant dexamethasone (Odio et al. 2003) used.

Instead, non-steroidal anti-inflammatory drugs were given routinely.

3.2. Agent identification

To identify the causative agent, drilling or trepanation for OM, and a needle

aspiration for SA was recommended. Gram-staining and bacterial culture of the

41

synovial fluid and/or bone sample were done. Blood culture was to be taken from

all patients. To avoid debate whether the diagnosis was correct only bacterial

positive patients were included in the final analysis.

3.3. The role of surgery

Beyond initial drilling, no surgery was recommended for OM, unless considered

essential by an experienced pediatric or orthopedic surgeon. Surgery was kept at

minimum for SA as well, and repeated aspirations, or arthrotomy and lavage were

not recommended even in shoulder or hip arthritis, unless the clinical response

was unsatisfactory, or the responsible orthopedic surgeon deemed it mandatory.

3.4. Monitoring the course of disease

Preset laboratory and radiographic investigations comprised plain radiograph on

admission, and on days 10 and 19; basic blood analysis at presentation, and on

day 5 and 10. CRP (Peltola and Räsänen 1982, Peltola et al. 1984) and ESR were

measured sequentially during the entire follow-up. At the last visit, the child was

examined thoroughly, and radiography, ESR and CRP were checked. Values

exceeding 20 mg/L for CRP or 20 mm/h for ESR were deemed increased. CT and

MRI were done on demand. The findings were recorded in special forms from

which the data were computerized and analyzed in Helsinki. A researchers’

meeting was held once a year.

3.5. Dealing with special problems, control visits

Medication was discontinued if most (though not all) symptoms and signs had

subsided and CRP declined to less than 20 mg/L; current ESR value did not affect

this decision. If CRP remained elevated or the signs were still prominent,

medication was continued until two normal CRP values (extra checks) were

obtained. In case of likely allergy, medication was switched from 1st generation

cephalosporin to clindamycin or vice versa.

42

Because of the potential late problems of osteoarticular infections

(Howard et al. 1976, Spindler et al. 1998, Nunn et al. 2007, Tice et al. 2003)

control visits were scheduled at 2 weeks, 3 months, and at least 1 year from

discharge. A study group member executed all investigations, paying special

attention to potential sequelae. Plain radiograph, and the ESR and CRP

determinations were included in the check-ups.

4. Outcome measures and statistical analysis

The primary endpoint was full clinical recovery without reinstitution of treatment

for an osteoarticular infection within the follow-up of 12 months. Secondary

endpoints comprised all potential sequelae.

Regarding the comparative treatment trials (IV, V), the sample size

calculation took into consideration the 95% confidence interval (CI95%) for the

difference in success rates using the normal approximation to the binomial

distribution. The non-inferiority test was based on the lower bound of the

confidence intervals being within pre-specified non-inferiority margin of 15% and

the upper bounds containing 0%. Assuming a 90% efficacy in both groups, a

statistical power of 80%, and a one-sided significance level of 0.025, 63 patients

were needed in the groups to test the null hypothesis (difference of at least 15% in

treatment). All P-values were two-sided and not adjusted for multiple analysis.

After the conclusion of the original trial the gathered data was also

to be used in retrospective analyses.

RESULTS

1. Shortened duration of antimicrobial treatment for OM

The number of cases affecting bone was 131 (V). Bacteria grew from bone and

blood in 35 cases, from bone alone in 44 cases, and from blood alone in 52 cases

as depicted in Figure 9. S. aureus, all methicillin-susceptible strains, was the

43

overwhelmingly most common agent being responsible for 89% (117/131) of

cases. Causative organisms are presented in Figure 10.

Figure 9. Isolation of causative organisms from bone or blood in OM.

Figure 10. Causative organisms in OM.

All age groups were affected as the median age was 9.0 years in the

short-term and 9.3 years in the long-term treatment group (V). Medical attention

was sought within 3 days and 7 days in 47% (N=61) and 84% (N=110),

respectively. An adjacent joint (OM+SA) was involved in 13 and 11 cases in the

short and long treatment groups (III, V). OM+SA induced significantly greater

changes in ESR and CRP than OM (p<0.05) (III, IV).

44

Antimicrobials were administered intravenously for 3.7 (median)

and 4.1 days in the short-term or long-term treatment group, respectively. The

duration of entire antimicrobial medication in the former group was 20.0 days

(interquartile range 0 day, 90% range 10-21 days), the shortest course (of

cephalosporin) being no more than 9 days (V). In the long-term treatment group,

the children were administered antimicrobials for 30 days (III, IV, V). The

extreme prolongations in medication were 108 days in the short-term, and 91 and

80 days in the long-term treatment group (V).

The children in both groups recovered fast without difference

between groups in the follow-up indices. Importantly, no laboratory or clinical

marker in the 20-day group deviated from the other group after the antimicrobial

had been discontinued (IV, V).

At the 1-year checkup, 15 children in the short-term, and 18 in the

long-term treatment group did not show up since they considered themselves

already fully recovered. There were 98 attendees at the final control.

Complications are described in Results 7: Problem cases. One child in the long

and one child in the short treatment group had major sequelae (V).

No case relapsed, nor was corrective surgery needed (II, V).

2. Shortened duration of antimicrobial treatment for SA

This series comprised 130 SA patients who represented all age groups, with some

preponderance among under 2-year-olds. The mean age was 6.5 years. Medical

attention was sought within six days in 93% of the patients. No clear association

prevailed between the length of history and the presenting status.

The lower extremities were usually affected, with hip, knee and

ankle comprising 85% of the cases (III, IV, Figure 8a). The joints were affected

rather similarly in the short-term and long-term treatment groups. Synovial fluid

and blood grew bacteria in 32%, synovial fluid alone in 46% and blood only in

22% (IV) as seen in Figure 11.

45

Figure 11. Isolation of causative organisms from joint or blood in SA.

The majority of cases were caused by S. aureus, Hib, S. pyogenes and S.

pneumoniae (I, II, IV). Proportion of causative organisms is given in Figure 12.

63 children were in the short-term and 67 children in the long-term treatment

group. Antibiotics were given intravenously, on average, for 3 days (I, II, IV).

Figure 12. Causative organisms in SA.

There was no difference between the treatment groups in any of the

follow-up indexes. In the short-term treatment group, the median of the entire

46

antibiotic course was 10 days, whereas in the long-term treatment course the

median was 30 days (II, IV). Percutaneous aspiration was performed in 110

cases. Arthrotomy was carried out in 15 cases, occasionally with drilling of

adjacent bone to exclude OM (II, IV, V).

One case in the long treatment group relapsed. One child with

previous Erb-palsy showed extension deficit of the affected elbow. Two children

had asymptomatic radiographic abnormalities [Results 7. Problem cases].

3. Clindamycin and first generation cephalosporin in the treatment of OM,

OM+SA or SA

Of the 265 enrolled children, 96 were excluded from this analysis, because of a

full course of ampi-/amoxicillin, or another agent had been used (data to be

published). A short initial course of ampi-/amoxicillin was given to 6

clindamycin and 8 cephalosporin recipients; they were analyzed. The final series

consisted of 169 patients treated primarily with clindamycin 1st generation

cephalosporins (data to be published).

S. aureus, invariably susceptible to methicillin, was overwhelmingly

the most common agent in both groups (I, II, III, IV, V). Both groups were

treated intravenously for a median of three days (data to be published). The total

median lengths of the antimicrobial course were 23 and 24 days for the

clindamycin and the cephalosporin groups, respectively. Despite high doses, both

antimicrobials were tolerated without major problems (III, IV, V). Loose stools

were reported in 1% of the clindamycin and 7% of the cephalosporin recipients.

Two clindamycin recipients developed rash (data to be published).

In general, the patients responded well to treatment, no matter which

antimicrobial was used (IV, V). The peak CRP appeared one day sooner in the

clindamycin group, but the difference was not significant (p=0.13). Otherwise,

CRP and ESR normalized to the level of 20 mg/L and 20 mm/h (I) on days 9 and

29 in both groups, respectively (to be published). Discontinuation of the

antimicrobials did not lead to a re-increase of any of these parameters (IV, V, data

47

to be published). At two weeks after discharge 36% of the children had

symptoms, with no difference between groups.

The children were followed up for one year (II, III, IV, V). As

characteristic of osteoarticular infections, the symptoms and signs subsided

gradually. Complications are described in Results 7. As mentioned above, one

SA patient initially treated with cephradine experienced two late re-infections

(IV,). The antibiotics used in cases that developed complications are described in

Results 7.

4. Need for surgery in OM or SA

4.1. Need for surgery in OM

Among the 131 OM patients, corticotomy (drilling or bone biopsy) was performed

in 62 cases, combined with arthrocentesis in 14 cases. Percutaneous bone

aspiration was performed in 26 cases. Seeking for arthritis, 8 children underwent

joint aspiration and a subperiosteal abscess was drained in 4 cases. The length of

medication, or the choise of antimicrobial treatment did not affect the need for

surgical intervention, no matter which treatment group was in question (III, V).

Thirty-one OM patients (24%) did not undergo procedures. Proportions of

surgical interventions among OM patients are described in Figure 13.

Figure 13. Surgical procedures among OM patients

48

4.2. Need for surgery in SA

Because the hip joint is perhaps deemed to be the most perilous entity of all SAs

(Vidigal et al. 1997), the 62 children with septic coxitis were analyzed separately

(II). The median age was 7.2 years, and medical attention was sought within 7

days in 97% of the cases. In 21/62 (34%) cases bacteria grew from the synovial

fluid only, from blood only in 17/62 (27%) cases and from both joint and blood in

24/62 (39%) cases. The most common causative agent was S. aureus in 71%.

Multiple joint involvement was seen with 2 children, both having the shoulder

joint affected. An adjacent bone was involved in 10 cases (16%) (II).

The mean initial CRP was 91 mg/L, ESR rate 57 mm/h and

leukocyte count 13,400 /mm³. The duration of total antimicrobial treatment

depended on the randomization to 10 or 30 days (IV, V). The mean CRP peak

level was slightly higher among operated (162 mg/L) vs. non-operated patients

(134 mg/L) (II). In both groups CRP peaked on the second day of treatment (I, II).

One patient escaped all procedures under anesthesia, whereas 61/62

(98%) children underwent one or more joint aspirations as shown in Figure 14.

The majority of the patients underwent 1 aspiration only (II). Arthrotomy was

carried out in 12 cases.

95% of the coxitis patients who showed-up in a control visit of at

least 12 months post-hospitalization were either asymptomatic or were found to

be asymptomatic in an additional follow-up control arranged later on (II, III, IV).

Two patients had radiographic abnormalities [Results 7.].

In reviewing the series of SA (all localizations, N=130) we found

that 85% of the patients were treated with aspiration alone. We also performed a

subanalysis of shoulder arthritis –another entity previously thought to be

especially perilous (data to be published). The series was small, but anyway 8/9

cases had been treated with a mere diagnostic aspiration after which antimicrobial

was promptly administered.

49

Figure 14. A child with a hip arthritis undergoes a diagnostic aspiration from the

inguinal angle under fluoroscopic control. Courtesy of Dr.Pentti Kallio.

5. Usefulness of serum CRP in the diagnostics and monitoring of the course

of OM, OM+SA or SA

The entire series of our 265 patients are summarized in Table 8+. ESR on

admission was elevated in 94 % of the cases (I). Although the peak was reached

approximately as soon as with CRP, on day 2 (I, II, IV), ESR normalized

significantly slower (p<0.05) (I). The mean ESR values (mm/h ± SEM) on days 1,

3, 5, 7, 10, 14, 19 and 29 were 51 ± 2, 67 ± 2, 66 ± 2, 62 ± 2, 52 ± 2, 45 ± 3, 30 ±

2 and 20 ± 1 as seen in Figure 15.

The initial CRP value was elevated in 95% of cases (I), mean being

87 ±4 mg/L. CRP peaked on day 2, but normalized rapidly in 10 days (I); the

mean values (mg/L ± SEM,) on days 2, 3, 5, 7, 9, 10, 12, 14, 17, 19, 23, 26, and

29 were 107 ± 5, 100 ± 5, 58 ± 3, 35 ± 3, 23 ± 3, 21 ± 2, 20 ± 3, 15 ± 1, 16 ± 1 ,

50

12 ± 1, 13 ± 2, 15 ± 4 and 12 ± 1 , respectively. CRP was higher in OM+SA than

in SA, whereas the lowest values were found in OM (I). Patients undergoing

extensive surgery had on average higher CRP values than others (II).

Figure 15. Mean CRP (mg/L) on days (D) from admission.

The mean WBC on admission was 12,600 (± 400) per mm³, and on

days 5, 10, 19 and 29 it was 7,900 ± 200, 8,100 ± 200, 6,900 ± 200, 6,900 ± 200,

respectively. WBC on admission exceeded >15,000 per mm3 in 25% (65/265) of

cases. The only difference between the disease manifestations was that the mean

WBC was lower in OM than SA or OM+SA.

The duration of symptoms before hospitalization was reflected in the

CRP and the ESR but not the WBC values. Had the history been 3 days or less,

ESR, CRP and WBC on admission for the whole series was 46 ± 2 mm/h, 85 ± 5

mg/L, and 13,700 ± 600 per mm3, respectively. If the history was more than three

days, initial ESR, CRP and WBC 58 ± 3 mm/h, was 90 ± 6 mg/L, and 11,000 ±

51

500 per mm3. In this group ESR peaked on day four (± 0.5), CRP on day two (±

0.3).

6. Treatment of OM + SA in respect to the treatment of OM

Of the 131 patients with OM, we excluded cases where there was no logical

potential to clinically, bacteriologically or radiologically separate OM from

OM+SA. This left us 117 cases. In those, an adjacent joint was found affected in

25 (21%) cases. The antimicrobials were administered according to the protocol

for 20 or 30 days (III, IV), and a quasi-randomization was done between

clindamycin and a 1st generation cephalosporin (V).

Overall, the mean duration of antimicrobials was 26 days

(±SEM=1.3), the median being 21 (IQR 20-30) days (I). Both the intravenous and

the total course of antimicrobials lasted about a week longer in OM+SA than OM

group, the mean total duration in OM+SA being 32 days (p<0.05) and 25 days in

OM (III). Compared with plain OM, OM+SA induced greater initial CRP, ESR

and WBC response (I), and normalization took longer than in plain OM (II).

Two patients in this series - one with OM and one with OM+SA -

showed major sequelae when controlled at 1-year or later (V) [see below 7.2, 7.3].

7. Problem cases

7.1. SA, boy, 10-years, late reinfection

Staphylococcal tibiotalar arthritis of a 10-year-old boy was treated with

cephradine for 28 days. Seventeen months later the same joint was affected, and S.

aureus was isolated. Cephradine was reinstituted, but due to suboptimal response,

the course was completed with clindamycin orally for 30 days. Again, the

recovery seemed uneventful until eight months later the same ankle was once

more affected, this time by coagulase-negative staphylococci. Clindamycin was

administered for 30 days. Since 1990 the child has remained well. No surgery

besides aspiration was ever carried out (IV).

52

7.2. OM+SA, boy, 12-years, symptomatic arthrosis of the tibiotalar joint

In the long treatment group a 12-year-old boy with ankle OM+SA was treated

with iv clindamycin for 3 days and oral penicillin for 77 days. This remarkable

prolongation of medication occurred because of slow clinical recovery. Only

diagnostic arthrocentesis was carried out.

At the 1-year-control the patient told he felt pain during exercise

(V). Radiography identified joint destruction in the tibiotalar joint.

7.3. OM, boy, 1-years, 8 degree varus deformity

1.6-year-old boy with S.aureus tibial osteomyelitis was assigned to the short-term

treatment group of 20 days. Antimicrobial administration was prolonged to 108

days due to slow recovery. The treatment was started with cephalosporin, but was

changed to clindamycin after 10 days. Surgical abscess drainage was carried out

on day 26 of treatment.

At the final control the patient presented with 8 degree varus

deformity of the affected limb.

7.4. SA, boy, 11-years, extension deficit of the elbow joint

An 11-year-old boy with perinatal Erb’s palsy and elbow arthritis. He was left

with a mild extension deficit of the affected elbow joint (considered to be related

with previous palsy) (IV). The boy had received clindamycin for 30 days.

7.5. SA, boy, 2-years, radiographic abnormality

A mild coxa magna was found in a 2-year-old boy who had been treated with

aspiration and lavage for hip arthritis. Antibiotic therapy was started with

cephradine but changed to ampi-/amoxicillin after 2 days. Total course was 11

days. The patient was asymptomatic.

53

7.6. SA, boy, 5-years, radiographic abnormality

A slightly narrow hip joint space in a 5-year-old boy who had undergone repeated

aspiration for hip arthritis. The patient was asymptomatic. He had been treated

with clindamycin for 30 days.

7.7. Minor sequelae

A 14-year-old boy with a previous hip arthritis showed one centimeter limb

shortening at the one-year control. The defect normalized in the following 12

months (II). Fibular osteomyelitis with adjacent tibiotalar arthritis of an 11-year-

old boy recovered uneventfully. A possibility of a developing postural flat foot

was suspected in the 1-year-control. An additional control was arranged at 2

years, but no postural flat foot was detected and the function of the ankle joint was

normal.

7.8. Adherence to the protocol

Deviations from the preassigned protocol occurred (II, III, IV, V). Most common

was a prolongation of the total duration of the antimicrobial course. The

intraquartile range of treatment in the short treatment groups was 10-15 days for

SA (IV), but most often this prolongation occurred with OM+SA patients (III),

who received antimicrobials for approximately a week longer than OM patients

on average (mean 32 days in OM+SA vs 25 days in OM). CRP normalization was

also the slowest in OM+SA (I).

54

DISCUSSION

1. Shortened duration of antimicrobial treatment for OM

The traditional view is that acute pediatric OM requires a long treatment course of

antimicrobials (Mollam and Biggott 1977, Syrogiannopoulos and Nelson 1988,

Syrogiannopoulos and Nelson 1988, Shaw et al. 1990, Smith an Piercy 1995, Dahl

et al. 1998, Christiansen et al. 1999, Bonhoeffer et al. 2001, Lew and Waldvogel

2004, Malcius 2005, Krilov and McCracken 2006). Our study (V) showed that

OM is often curable without a notable risk of recrudescence or sequlae with

antimicrobials for only 20 days. The intravenous phase can last only a few days (if

needed at all) and the course can be completed orally (IV, V). No serum assays

are needed, but large doses of well-absorbing agent such as clindamycin or 1st

generation cephalosporin and qid dosing are probably mandatory.

With this approach we did not have any recrudescences among our

patients, even though short courses were used (V). Sequelae were found in 2

cases, one in both treatment groups. This does not in any way suggest that the

short course would have been inferior.

For the first time it was shown that antimicrobials fulfilling the

above mentioned criteria for a total of 20 days in OM is not inferior to 30 days,

provided the clinical response is good and CRP normalizes soon.

Our data on shorter and cheaper treatment courses should be good

preliminary news especially in the resource-poor settings, where most cases of

osteomyelitis occur.

2. Shortened duration of antimicrobial treatment for SA

SA is also a disease in which tradition mandates a lengthy course of

antimicrobials (Ross et al. 2006, Vinod et al. 2002). Our study challenges this

view too (II, III, IV).

Only a 10-day course sufficed in most cases of pediatric SA (IV).

The rate of a recrudescence in SA is not known. Interestingly, among 168 cases in

55

the USA (Syrogiannopoulos and Nelson 1988) there were no recrudessences. In

our series we had only one patient experiencing two late re-infection among 130

cases (IV). These interesting episodes were not true recrudessences but rather late

re-infections. Thus it seems SA has a lower tendency to reoccur than OM. Two

patients had asymptomatic radiographic abnormalities, but the overall clinical

relevance of these findings remained unclear as no symptoms developed during

the follow-up period or came to attention at a later stage. Caput magna may

develop into acetabulum dysplasia or joint degeneration may develop even

decades later, and thus cannot be totally excluded.

Again it was shown that the antimicrobial course can often – though

indisputably not always – be considerably shortened from that which is routine in

most institutions dealing with childhood SA.

3. Clindamycin and 1st generation cephalosporin in the treatment of OM,

OM+SA or SA

Both clindamycin and 1st generation cephalosporins proved in our hands effective

and safe agents. They have a good activity against the most common bacteria

causing OM or SA (S. aureus and streptococci). They are also well absorbed and

the tissue penetration is so good that oral administration is possible (IV, V). No

serum assaying is needed, but good adherence to treatment is necessary. Oral

administration is much easier to the patient and personnel, and lowers the costs.

This is especially important in low income settings as oral agents are considerably

cheaper than parenteral antimicrobials.

No major difference was observed between clindamycin and the

cephalosporins we used. Most children responded well to either antimicrobial, and

this was soon reflected in the declining CRP and ESR values. Interestingly, CRP

began to normalize one day sooner in the clindamycin group, although the

difference was not significant. The difference may perhaps suggest a slightly

greater activity of clindamycin. Thus clindamycin successfully defends its

position as a relevant alternative to the newer, costly antimicrobials. Diarrhea was

not especially common, although we used large doses (IV, V). Instead, two

56

patients developed rash, but it remained unclear whether the association was

causal. If it was, a low two-percent incidence seems acceptable in light of the

aforementioned advantages of clindamycin.

One patient developed two late infections as described in Results

7.1. This may not have been an antimicrobial agent failure. First of all, the

causative agents differed and secondly, the intervals between episodes were very

long, 17 and eight months (IV). One can also question whether a prolonged

medication would have prevented this late re-infection from occurring (Uckay et

al. 2006).

Two patients treated with clindamycin monotherapy developed

radiographic abnormalities as mentioned earlier, but the clinical significance was

not confirmed during the study period. One patient with Erb palsy was treated

with clindamycin and developed an extension deficit of the elbow. The extension

deficit remained but was suspected to be related to the previous palsy.

Clindamycin and first-generation cephalosporins are effective and

safe agents for the treatment of osteoarticular infections of childhood if local

antimicrobial resistance permits their use. We proved that most childhood OM,

SA and OM+SA cases can still be safely treated with a first generation

cephalosporin or clindamycin.

4. Need for surgery in OM or SA

4.1. Need for surgery in OM

Minimal surgery in OM (percutaneous aspiration or drilling) aiming primarily to

obtain a representative sample for bacteriology, sufficed in most cases of OM. No

less than 24% of our patients escaped all procedures – and recovered

uneventfully. Primary routine surgery besides a diagnostic sample did not seem to

offer additional benefit to the patient.

An old study from 1944 (Kenney 1944) showed that when acute

osteomyelitis was operated early, mortality increased but late sequelae decreased.

In our material most early presenting cases do not require extensive surgical

57

procedures. This said, our protocol did not give strict criteria to suggest when to

operate, because it left the ultimate decision to the clinician in charge. Even

keeping this shortcoming in mind, we opine that most early presenting cases do

not require routine extensive surgery. Modern imaging (US, MRI) help todayʼs

collegues to detect better abscesses or other processes that may warrant drainage.

In our series a subperiosteal abscess was drained in 4 cases. We believe that

extensive surgery should be reserved to cases showing no or little improvement

during the first days of treatment, or cases of chronic infection.

4.2. Need for surgery in SA

Our analysis on pediatric septic hip showed that among our 62 cases arthrotomy

could be avoided with most patients (II, IV). The series comprised of consecutive

patients who were carefully followed-up to reveal any possible sequelae. We think

the finding is important and should raise some rethinking.

Actually, as similar results have been published previously by other

authors (Lavy et al. 2003, Manadan and Block 2004) one may wonder why

aggressive routine arthrotomy is still favored by most orthopedic surgeons.

Perhaps a historical aspect may offer one explanation. Before the antimicrobial

era SA was a killer disease and, if the child survived, caused major disability for

life (Smith 2007). Thus, understandably, decompressive surgery for especially the

hip and the shoulder joint were recommended by experienced clinicians (Welkon

et al 1986, Parsch and Savvidis 1997, Gordon et al. 2005) because of the fear of

avascular necrosis of the femoral head or pressure induced dislocation of the

femoral head leading to acetabular dysplasia. Accepting the logic of the fear of

pressure–induced avascular necrosis, it might be a surprise that evidence proving

this sequence to be true is lacking. Extremely high, position dependent intra-

articular hydrostatic pressures have been measured in experimental animals

(Levick 1979) and children (Kallio and Ryöppy 1985), but in a prospective

follow-up on children with hip joint effusion, hyperpressure did not result in

avascular necrosis in any patient (Kallio et al. 1985).

58

Pressure conditions in entirely encapsulated joints are much

dependent on the positioning of the extremity. Position-induced hyperpressure can

be produced experimentally, but it decreases rapidly with time (Kallio and

Ryöppy 1985). Thus, combining all information deriving from animal

experiments and from prospective clinical studies on children to our experience of

62 consecutive cases, one has good grounds to cast doubts to this old hypothesis

of pressure-induced avascular necrosis.

Thus, the value of decompressive arthrotomy can rightly be disputed

(Lavy et al. 2003, Manadan and Block 2004, Griffet and Hayek 1996). In fact,

both CRP and ESR normalize considerably slower if surgery/arthrotomy has been

performed (Peltola et al. 1984, Kallio et al. 1997, II). This finding agrees with the

observation on trauma surgery in which inflammatory parameters have been

monitored sequentially (Kallio et al. 1990).

Routine arthrotomy in the septic hip and also in shoulder (results to

be published) arthritis should generally be avoided. Instead, percutaneous

aspiration for diagnostic purposes should be done. Sometimes, in non-responding

cases, arthrotomy should be carried out.

As we have shown that routine arthrotomy offers no benefit even

these perilous hip and shoulder joints, we can safely assume that routine

arthrotomy can be avoided in the majority of cases in other localizations as well.

This hypothesis was confirmed by our large series of cases (IV, V) – including

among others ankle, knee and elbow arthritides– mostly treated successfully with

aspiration and antimicrobials.

5. Usefulness of serum CRP in the diagnostics and monitoring of the course

of OM, OM+SA or SA

ESR and CRP and WBC behaved similarly in OM, SA, and OM+SA: a fast ascent

with highly increased values usually found on admission, a peak on the 2nd or 3rd

day of treatment, and then a fast descend. The speedy normalization differed

significantly in favor of CRP (I).

59

ESR should be measured on arrival, because in rare cases, the first

CRP value was unincreased being <20 mg/L (I). We assume that the

inflammatory process is sometimes too weak to trigger hepatocytes to produce

CRP in large quantities, and in this respect, ESR is useful (I). Measuring WBC

has little value (I).

Sequential CRP determinations were of great value in the

monitoring of the course of disease. Between OM, SA and OM+SA, there were

however, differences as OM induced a low, SA greater, and OM+SA the greatest

CRP response. Although this was the case, we could not always distinguish

OM+SA from SA or OM with CRP alone (Unkila-Kallio et al. 1994).

To maximize the informative value of CRP, sequential

determinations are needed, and the test results should arrive the same day (we

receive them in 2-3 hours) (I). One should be acquainted with the dynamics of this

valuable yardstick: even when an osteoarticular infection heals uneventfully, CRP

usually increases for 1-2 days after the institution of an antimicrobial, but then

descends soon reaching 20 mg/L in 7-10 days. If CRP continues to rise, or

remains high on the 4th day, a complication is likely (Roine et al. 1995).

Today CRP can be measured fast and reliably from a whole-blood

sample which has been taken from a finger or heel prick. The costs are also

tolerable.

6. Treatment of OM + SA vs treatment of OM

Although children with OM+SA were treated along the lines of OM with

antimicrobial for 20 or 30 days there were no recrudescences among our 25

OM+SA patients. Sequelae developed in two cases. Our treatment was

significantly shorter than still recommended by many (Karwowska et al. 1998,

Dieckmann et al. 2008). In fact, antimicrobials for no more than 20 days generally

did as well as 30 days (III, V). This was often the case OM+SA. The intravenous

phase was short, 4 days in OM and 6 days in OM + SA, on average. As might be

expected, OM+SA patients were treated somewhat longer – on average about a

week – than those with OM. Since the discontinuation of antimicrobials was

60

bound to lowering of CRP this might reflect the slower decline of inflammatory

parameters we have already reported (I). It may even be that cases might have

recovered even with a shorted medication, but we have no data to evaluate this

assumption.

No difference in the clinical effectiveness of clindamycin vs. a 1st

generation cephalosporin was observed. However, we intentionally used

exceptionally large doses, both intravenously and orally, as recommended long

ago (Nelson et al. 1982), and a qid regimen was used. We believe that both of

these aspects are important.

7. Limitations of the study

7.1. General

Since our study did not include neonates or immunodeficient patients (I, II, III,

IV, V) we have no comment on how treatment should be modified in these cases.

As no prospective trials on these populations have been published, the clinicians

have to tailor treatments according to the present recommendations.

All our S. aureus strains were susceptible to methicillin (III, IV, V).

Therefore we have little to say how to cope with resistant staphylococci, except

that those strains have still mostly remained susceptible to clindamycin (Rehms

and Tice 2010). The emergence of ca-MRSA no doubt poses a true challenge to

the treatment.

Not a single case caused by Salmonella spp or K. kingae were found

in this series whose enrollment lasted more than two decades. Salmonella is a

rather common pathogen in the developing world (Lavy et al. 1996), and it may

require longer treatments than we used. Regarding K. kingae, we made sure with

the bacteriologists that there were no methodological problems in the detection of

this pathogen, and this was not the case. Evidently, K. kingae is rare in Finland (as

it is in many other countries as well), at least so far. Interestingly, since 1995, K.

kingae has been isolated with an increasing frequency in pediatric OM in Iceland

61

(Masson et al. 2007). It will be interesting to see whether the same will also

happen here, in the eastern part of Scandinavia.

Our enrollment rate was less than the presumed incidence, as we

enrolled 2.4 patients/ 100 000/ year from the pediatric population. OM is rare in

Finland; the annual incidence at age 0-14 years is only 4.5 per 100.000 for OM

and essentially the same for SA (Peltola and Vahvanen 1983). The incidence may

have reduced somewhat because of the decline in Hib arthritis (Peltola et al.

1998). Still, we cannot call our study population based. Many reasons probably

contributed to the low enrollment rate. As neonatals, immunodeficient patients

and patients with underlying illnesses were excluded this also reduced the

enrollment rate. Cases where the causative organism was not identified were also

excluded. Hib vaccination also reduced the incidence of Hib arthritis during the

last years of the study (Peltola 1998). Despite the low rate we are happy to note

that the enrollment was fairly even in most of the participating hospitals.

Our study was collected in a well developed high income country

and thus cannot be directly applied in a low income setting, where health facilities

are not always available. Our patients presented early – usually in 2-4 days from

the onset of symptoms – so we cannot comment whether surgery or longer

treatment would have been needed had the history been much longer.

7.2. Randomized study on treatment length and the choice of antimicrobial

Collecting 131 culture-positive cases took long, because our strict enrollment

criteria our enrollment rate was fairly low. We felt it necessary to limit our study

to cases where the diagnosis was well confirmed and this slowed the enrollment

considerably. However, yearly researchers’ meetings kept the rules the same. We

are happy to note that the enrollment rate was fairly similar in most of the

attending hospitals.

A special concern was the withdrawal of the primarily chosen

cephalosporin (cephradine) from the market. Fortunately pharmacologically very

similarly performing cephalotin (iv) and cephalexin (po) were available.

62

Regarding the protocol, plain X-rays were taken routinely during the

controls but MRI was done only on demand. This practice is consistent with a

recent study (Courtney et al. 2010). We also had dropouts from our carefully

planned follow-up protocol (II, III, IV, V). This said, one should realize that

patient care in Finland is virtually free of cost, and children recovering from such

a grave disease as OM, OM+SA or SA would almost certainly have sought further

treatment, should problems have arisen. Also, there are only a limited number of

centers in Finland which treat sequelae from pediatric osteoarticular infections.

We find it unlikely that a child with severe sequelae would have been lost from

our attention during the study years and thus are confident with our results.

We do realize that unfortunately some sequelae, such as coxa

magna, narrow joint space or varus deformity, may remain asymptomatic and

only show symptoms such as arthrosis only after decades. Thus practically no

study on pediatric OM or SA can claim a 100% certainty that no undetected

sequelae occurred. We also had to exclude patients from further analysis, because

antimicrobials were not invariably used as recommended. Not all children

received the recommended agents, and neither was the treatment always exactly

20 or 30 days in OM and OM+SA or 10 or 30 days in SA. Only 169 of the 265

patients were included in the analysis. These deviations are regrettable, but we

believe that the message of the study was not blurred even by this shortcoming.

7.3. Analyses on surgery

The study protocol did not randomize the children in the arthrotomy and non-

arthrotomy groups. Even so, for example our 62 patients with hip arthritis were a

fairly homogenous group (II). Because of the rarity of septic hip or even septic

arthritis in general a randomized trial on surgery of OM or SA may well prove

impossible to arrange. It was not feasible in our circumstances. It may prove

impossible to follow strict uniform criteria to randomize children to different

treatment protocols. It probably would have proved impossible to force individual

surgeons in different settings to adhere to such protocols, because they were

responsible for the children and, of course, wanted to treat their patients according

63

to their best knowledge. Even realizing these facts, we have already proven that

even a loose protocol of suggesting to avoid surgery may still reduce it by about

80% (II). We find this to be a convincing achievement in its own right.

Since medical advice was sought most often in 2-4 days, we cannot

comment whether surgery would have been of greater value had the history been

significantly longer. We have found that children presenting early – say under 5

days from the onset of disease – tend to respond faster to treatment (data to be

published). Even so, even late presenting cases in our series tended to recover

fairly well.

7.4. Problem cases among patients

There were patients referenced in Results 7. who did not recover optimally. How

did these cases affect the interpretation of the results?

We find it important to note that most of these patients had been

treated with a long course of antimicrobials. In fact, they often underwent

significantly longer treatments than average. Patients presented at Results 7.1.,

7.2, 7.3, underwent treatments of 28, 80 and 108 days, respectively. We may

claim that our protocol was able to identify special cases from the sluggish CRP

descent or slow clinical recovery early on, and a prolonged antimicrobial regimen

was assigned. Even a prolonged antibiotic regimen did not save all the patients

from unfortunate sequelae. These problem cases are just one more challenge for

the clinicians to face.

8. From complex treatment to simplified treatment

The incidence of OM, SA and OM + SA is decreasing in many - if not all

(Masson et al. 2007) - industrialized countries, patients present early and are

treated with effective antimicrobials. Our results bring solid data to support the

view that most of these cases can be treated with a simplified treatment course.

The simplified treatment may proceed for example in the following

manner: On admission diagnostic joint/bone aspiration or drilling is done and

64

blood culture is taken (IV, V). No other surgical procedures are done routinely

(II). Intravenous antimicrobial is promptly instituted and is switched to oral

administration after a few days. The patient is followed up with sequential clinical

investigations and CRP monitoring (I, II, IV, V). Both ESR and CRP are

measured on arrival (I) but only CRP needs to be measured sequentially (I). Fever

and CRP should usually be descending by the second or third day of treatment

(II), and intravenous administration can be changed to oral, with large doses qid.

The antimicrobial is discontinued after 20 days in OM or OM+SA (V) or after 10

days in SA (IV).

Only if the clinical course is complicated by abnormal healing or

underlying illnesses prevent normal recovery, the clinician will either resort to

invasive surgical procedures, change the antimicrobial, or prolong the

administration. Fortunately, according to our data this is seldom required in high

income countries (II, III, IV, V).

Even though we can now move to simple, safe, efficient, evidence

based (I, II, III, IV, V) treatment protocol of childhood septic osteoarticular

infections in high income setting, the real work is still to be done. Osteoarticular

infections are rampant in low income countries where the true need for

inexpensive treatments is most dire. There is an urgent need to test our approach

in a low income setting where a true need for cheap and effective solutions exists.

65

CONCLUSIONS

Most cases of childhood septic bone and joint infections can be treated in a short,

simple and inexpensive manner. For OM and OM+SA, a course of 20 days

usually suffices, whereas for SA, a 10-day-course heals most cases. After a short

(2-5 days) initial intravenous phase the administration can be changed to oral, but

large doses and a qid regimen are likely pivotal. Clindamycin and 1st generation

cephalosporins are both time proven and equally effective antimicrobials suitable

to be used as first-line treatment, local resistance permitting.

Diagnostic drilling or aspiration discloses the etiology; otherwise,

more extensive surgery may be reserved exclusively for complicated cases.

Routine arthrotomy should generally be avoided even in hip and shoulder arthritis.

Even in most cases of OM + SA extensive surgery is not needed. The prognosis of

OM+SA is comparable to that of plain OM.

CRP is an inexpensive aid in diagnostics of OM and SA, and is an

excellent tool in monitoring treatment and recognizing cases demanding special

attention.

ACKNOWLEDGMENTS

The parents of the 265 children enrolled in this study are to be thanked for

allowing their children to participate and for their patience during the one year

follow-up controls. It makes one humble to think of the trust these parents put in

their physicians.

The OM-SA Study Group has performed an amazing feat in

gathering patients for a quarter of a century for this study. No words are enough to

thank you all. Kari Aalto, Aurora Hospital, and Helsinki University Central

Hospital, Hospital for Children and Adolescents, Helsinki. Ilkka Anttolainen,

Päijät-Häme Central Hospital, Lahti. Marja Heikkinen, Kuopio University

Hospital, Kuopio. Anita Hiippala, Etelä-Saimaa Central Hospital, Lappeenranta .

66

Ulla Kaski, Seinäjoki Central Hospital, Seinäjoki. Niilo Kojo, Etelä-Saimaa

Central Hospital, Lappeenranta. Pentti Lautala, Päijät-Häme Central Hospital,

Lahti. Juhani Merikanto, Helsinki University Central Hospital, Hospital for

Children and Adolescents, Helsinki. Pekka Ojajärvi, Jorvi Hospital, Espoo. Eeva

Salo, Aurora Hospital, and Helsinki University Central Hospital, Hospital for

Children and Adolescents, Helsinki.

I am indebted to my co-authors, who have helped me in every

possible way in analyzing the results and in the writing of the articles. I warmly

thank Dr. Heikki Peltola for his guidance and always excellent comments. It was

Your enthusiasm, dedication and unique creativity that convinced me that

research work can really be thrilling and amazingly fun. I want to thank Dr.

Markku Kallio who first introduced me the field of septic osteoarticular infections

and taught me how to use Statview. Many excellent ideas for these articles were

invented in our meetings in my small student flat while having tea and blueberry

pie. The assistance and guidance I have received from Dr. Pentti Kallio has been

irreplaceable. I have been profoundly influenced by your ideas and innovations in

the field of pediatric orthopedic surgery. I also want to mention the important

work on OM-SA by Dr. Leila Unkila-Kallio, which has both inspired and greatly

facilitated my own research work.

I would like to thank my colleagues in Kuusankoski District

Hospital and Mikkeli Central Hospital for your support during the past years.

I would like to dedicate this book to my family, my parents Taisto

and Marja-Leena, everything I accomplish I owe it all to your support,

understanding and encouragement. To Rauha and Hilja thank you kindly for

participating in our poster. Of course to Jenni, the love of my life.

And finally to the Almighty God in heaven, who makes all things

possible.

REFERENCES

Abdullah ZS, Khan MU, Kodali SK, Javaid A. Pyomyositis mimicking osteomyelitis detected by SPET/CT.

Hell J Nucl Med 2010; 13: 277-9.

67

Abuamara S, Louis JS, Quyard MF, Barbier-Frebourg N, Lechevallier J. Osteoarticular infection in children:

evaluation of a diagnostic and management protocol. Rev Chir Orthop Reparatrice Appar Mot 2004; 90: 703-

13.

Agaja SB, Ayorinde RO. Chronic osteomyelitis in Ilorin, Nigeria. S Afr J Surg 2008; 46: 116-8.

Aguilar J, Urday-Cornejo V, Donabedian S, Perri M, Tibbetts R, Zervos M. Staphylococcus aureus

meningitis: case series and literature review. Medicine (Baltimore) 2010; 89: 117-25.

Akinyoola AL, Orimolade EA, Yusuf MB. Pathologic fractures of long bones in Nigerian children. J Child

Orthop 2008; 2: 475-9.

Al Arfaj AS. A prospective study of the incidence and characteristics of septic arthritis in a teaching hospital

in Riyadh, Saudi Arabia. Clin Rheumatol 2008; 27: 1403-10.

Al Saadi MM, Al Zamil FA, Bokhary NA, Al Shamsan LA, Al Alola SA, Al Eissa YS. Acute septic arthritis

in children. Pediatr Int 2009; 51: 377-80.

Altamimi S, Khalil A, Khalaiwi KA, Milner R, Pusic M, Al Othman MA. Short versus standard duration of

antibiotic therapy for acute streptococcal pharyngitis in children. Sao Paulo Med J 2010; 128: 48.

Andreola B, Bressan S, Calleqaro S, Liverani A, Plebani M, Da Dalt L. Procalcitonin and C-reactive protein

as diagnostic markers of severe bacterial infections in febrile infants and children in the emergency

department. Pediatr Infect Dis J 2007; 26: 672-7.

Asadi-Pooya AA, Bordbar MR. Are laboratory tests necessary in making the diagnosis of limb pains typical

for growing pains in children? Pediatr Int 2007; 833-5.

Autret-Leca E. A general overview of the use of ibuprofein in paediatrics. Int J Clin Pract Suppl 2003; 135: 9-

12.

Averill LW, Hernandez A, Gonzalez L, Peňa AH, Jaramillo D. Diagnosis of osteomyelitis in children: utility

of fat-suppressed contrast-enhanced MRI. AJR Am J Roentgenol 2009: 193: 1232-8.

Azoulay R, Alison M, Sekkal A, Sebag G, Adamsbaum C. Imaging of child osteoarticular infections. Arch

Pediatr 2007; 14 Suppl 2: S113-21.

Ballock RT, Newton PO, Evans SJ, Estabrook M, Farnsworth CL, Bradley JS. A comparison of early versus

late conversion from intravenous to oral therapy in the treatment of septic arthritis. J Pediatr Orthop 2009; 29:

636-42.

68

Balogun RA, Obalum DC, Giwa SO, Adekoya-Cole TO, Ogo CN, Enweluzo GO. Spectrum of musculo-

skeletal disorders in sickle cell disease in Lagos, Nigeria. J Orthop Surg Res 2010;5:2.

Baskett A, Hoskinq J, Aickin R. Hip radiography for the investigation of nontraumatic, short duration hip

pain presenting to a childrensʼ emergency department. Pediatr Emerg Care 2009; 25: 78-82.

Berhard P. Management of common bacterial infections of the skin. Curr Opin Infect Dis 2008; 21: 122-8.

Bloom KA, Chung D, Cunningham-Rundles C. Osteoarticular infectious complications in patients with

primary immunodeficiencies. Curr Opin Rheumatol 2008; 4: 480-5.

Bonhoeffer J, Haeberie 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.

Brannan SR, Jerrard DA. Synovial fluid analysis. J Emerg Med 2006; 30: 331-9.

Brosof AB, Spitzer TQ. Cephradine for the treatment of bone infections due to Staphylococcus aureus. Curr

Ther Res 1979; 26: 317-21.

Brown J, Wannamaker LW, Ferrieri P. Enumeration of beta-hemolytic streptococci on normal skin by direct

agar contact. J Med Microbiol 1975; 8: 503-11.

Butbul-Aviel Y, Koren A, Halevy R, Sakran W. Procalcitonin as a diagnostic aid in osteomyelitis and septic

arhritis. Pediatr Emerg Care 2005; 21: 828-32.

Caksen H, Ozturk MK, Uzum K, Yuksel S, Ustunbas HB, Per H. Septic arthritis in childhood. Pediatr Int

2000; 42: 534-40.

Ceroni D, Requsci M, Pazos JM, Saunders CT, Kaelin A. Risks and complications of prolonged parenteral

antibiotic treatment in children with acute osteoarticular infections. Acta Orthop Belq 2003; 69:400-4.

Chen WL, Chang WN, Chen YS, Hsieh KS, Chen CK, Peng NJ, Wu KS, Cheng MF. Acute community-

acquired osteoarticular infections in children: high incidence of concomitant bone and joint involvement. J

Microbiol Immunol Infect 2010; 43: 332-8.

Christiansen P, Frederiksen B, Glazowski MJ, Scavenius M, Knudsen FU. Epidemiologic, bacteriologic, and

long-term follow-up data of children with acute haematogenous osteomyelitis and septic arthritis: a ten-year

review. J Pediatr Orthop 1999; 8: 302-5.

Clarce SR, Brummell KJ, Horsburgh MJ, McDowell PW, Mohamad SA, Stapleton MR, Acevedo J, Read RC,

Day NP, Peacock SJ, Mond JJ, Kokai-Kun JF, Foster SJ. Identification of in vivo-expressed antigens of

69

Staphylococcus aureus and their use in vaccinations for protection against nasal carriage. J Infect Dis 2006;

193: 1098-108.

Cohen R, Grimprel E. Pharmacokinetics and pharmacodynamics of antimicrobial therapy used in child

osteoarticular infections. Arch Pediatr 2007; 14 Suppl 2: S122-7.

Cole WG, Dalziel RE, Leitl S. Treatment of acute osteomyelitis in childhood. J Bone Joint Surg Br 1982; 64:

218-23.

Conrad DA. Acute hematogenous osteomyelitis. Pediatr Rev 2010; 31: 464-71.

Cooke B, Smith A, Diggle M, Lamb K, Robertson C, Inverarity D, Jefferies J, Edwards G, Mitchell T, Clarke

S, McMenamin J. Antibiotic resistance in invasive Streptococcus pneumonia isolates identified in Scotland

between 1999 and 2007. J Med Microbiol 2010; 59: 1212-8.

Courtney PM, Flynn JM, Jaramillo D, Horn BD, Calabro K, Spiegel DA. Clinical indications for repeat MRI

in children with acute hematogenous osteomyelitis. J Pediatr Orthop 2010; 30: 883-7.

Craigen MA, Watters J, Hackett JS. The changing epidemiology of osteomyelitis in children. J Bone Joint

Surg Br 1992; 74: 541-5.

Dahl LB, Hoyland AL, Dramsdahl H, Kaaresen PI. Acute osteomyelitis in children: a population based

retrospective study 1965 to 1994. Scand J Infect Dis 1998; 30: 573-79.

Del Beccaro MA, Champoux AN, Bockers T, Mendelman PM. Septic arthritis versus transient synovitis of

the hip: the value of screening laboratory tests. Ann Emerg Med 1992; 21: 1418-22.

de Groot R, Glover D, Clausen C, Smith AL, Wilson CB. Bone and joint infections caused by Kingella

kingae: Six cases and review of literature. Rev Infect Dis 1988; 10: 998-1004.

Deshpande PG, Wagle S, Mehta SD, Bharucha BA, Irani SF. Neonatal osteomyelitis and septic arthritis.

Indian Pediatr 1990; 27: 453-7.

Dieckmann R, Hardes J, Ahrens H, Flieger S, Gosheger G, Göte C, Rödl R. Treatment of acute and chronic

osteomyelitis in children. Z Orthop Unfall 2008; 146: 375-80.

Doskocil M, Valouch P, Pazderka V. On vertebral body growth. Funct Dev Morphol 1993; 3: 149-55.

Eder L, Zisman D, Rozenbaum M, Rosner I. Clinical features and aetiology of septic arthritis in northern

Israel. Rheumatology 2005; 44: 1559-63.

Erdem G, Salazar R, Kimata C, Simasthien T, Len KA, Bergert L, Melish M. Staphylococcus aureus

osteomyelitis in Hawaii. Clin Pediatr (Phila) 2010; 49: 477-84.

70

Eyichykwu GO, Onyemaechi NO, Onyegbule EC. Outcome of management on non-gonococcal septic

arthritis at National Orthopaedic Hostpital, Enugu, Nigeria. Niger J Med 2010; 19: 69-76.

Faesch S, Cojocaru B, Hennequin C, Pannier S, Glorion C, Lacour B, Cheron G. Can procalcitonin

measurement help the diagnosis of osteomyelitis and septic arthritis? A prospective trial. Ital J Pediatr 2009;

35: 33.

Fang YH, Hsueh PR, Hu JJ, Lee PI, Chen JM, Lee CY, Huang LM. Community-acquired methicillin-

resistant Staphylococcus aureus in children in northern Taiwan. J Microbiol Immunol Infect 2004; 37: 29-34.

Feigin RD, Pickering LK, Anderson D, Keeney RE, Schackleford PG. Clindamycin treatment of

osteomyelitis and septic arthritis in children. Pediatrics 1975; 55: 213-23.

Fleming PC, Huda SS, Bobechko WP. Cephaloridine and the penicillins in the treatment of staphylococcal

osteomyelitis and arthritis. Postgrad Med J 1970; Oct Suppl 89-93.

Fodor L, Peled IJ, Lerner A, Ullmann Y. Acute osteomyelitis after a minor, 2nd degrees burn of the finger.

Burns 2004; 30: 185-7.

Forlin E, Milani C. Sequelae of septic arthritis of the hip in children, a new classification and a review of 41

hips. J Pediatr Orthop 2008; 28: 524-8.

Forsius A. Andreas Vesalius (1514-1564) – anatomian uudistaja. Suomen Lääkärilehti 1996; 17: 1954.

Franssila R, Hedman K. Infection and musculoskeletal conditions: Viral causes of arthritis. Best Pract Res

Clin Rhematol 2006; 20: 1139-57.

Frey CR, Miller ML, Lewis JS 2nd, Lawson KA, Peddaiahgari R, Talbert RL. Retrospective cohort study of

hospitalized adults treated with vancomycin or clindamycin for methicillin resistant skin infections. Clin Ther

2010; 32: 2024-9.

Gafur OA, Copley LA, Hollmig ST, Browne RH, Thorton LA, Crawford SE. The impact of the current

epidemiology of pediatric musculoskeletal infections on evaluation and treatment guidelines. J Pediatr Orthop

2008; 28: 777-85.

Geirsson AJ, Statkevicius S, Vikingsson A. Septic arthritis in Iceland 1990-2002: increasing incidence due to

iatrogenic infections. Ann Rheum Dis 2008; 67: 638-43.

Givon U, Liberman B, Schindler A, Blankstein A, Ganel A. Treatment of septic arthritis of the hip joint by

repeated ultrasound-guided aspirations. J Pediatr Orthop 2004; 24: 266-70.

71

Glorion Ch, Palomo J, Bronfen C, Touzet Ph, Padovani JP, Rigault P. Les arthrites aiguës infectieuses du

genou de l’enfant. Pronostic et discussion thérapeutique à propos de 51 cas ayanat un recul moyen de 5 ans.

Revue Chir Orthop 1993; 79: 650-60.

Gold R. Diagnosis of osteomyelitis. Pediatr Rev 1991; 12: 292-7.

Gomoll AH. Childhood leukemia presenting as sternal osteomyelitis. Am J Orthop 2007; 36: E148-50.

Gordon JE, Wolff A, Luhman SJ, Ortman JR, Dobbs MB, Shoenecker PL. Primary and delayed closure after

open irrigation and debriment of septic arthritis in children. Journal of Pediatric Orthopaedics 2005; 14: 101-

4.

Goudge J, Gilson L, Russell S, Gumede T, Mills A. Affordability, availability and acceptability barriers to

health care for the chronically ill: Longitudinal case studies for South Africa. BMC Health Serv Res 2009; 9:

75.

Griffet J, El Hayek T. Percutaneus drainage of septic hip arthritis in children. Rev Chir Orthop Reparathice

Appar Mot 1996; 82: 251-4.

Grimprel E, Cohen R. Epidemiology and physiopathology of osteoarticular infections in children (newborns

excempt). Arch Pediatr 2007; Suppl 2: S81-5.

Gutierrez K. Bone and joint infections in children. Pediatr Clin North Am 2005; 52: 779-94.

Hammond PJ, Macnicol MF. Osteomyelitis of the pelvis and proximal femur: diagnostic difficulties. J Pediatr

Orthop 2001; 10 B: 113-9.

Harik NS, Smeltzer MS. Management of acute hematogenous osteomyelitis in children. Expert Rev Anti

Infect Ther. 2010; 8: 175-81.

Hartwig NG. How to treat acute musculoskeletal infections in children. Adv Exp Med Biol 2006; 582: 191-

200.

Homans J. Osteomyelitis of the long bones: a study of ninety-four cases from the childrenʼs hospital, Boston.

Ann Surg 1912; 55: 375-87.

Host B, Schumacher H, Prag J, Arpi M. Isolation of Kingella kingae from synovial fluids using four

commercial blood culture bottles. Eur J Clin Microbiol Infect Dis 2000; 19: 608-11.

Howard JB, Highboten CL, Nelson JD. Residual effects of septic arthritis in infancy and childhood. JAMA

1976; 236: 932-5.

Hughes LO, Aronson J. Skeletal infections in children. Curr Opin Pediatr 1994; 6: 90-3.

72

Hugle T, Schuetz P, Mueller B, Laifer G, Tyndall A, Regenass S, Daikeler T. Serum procalcitonin for

discrimination between septic and non-septic arthritis. Clin Exp Rheumatol 2008; 26: 453-6.

Hunka L, Said SE, MacKenzie DA, Rogala EJ, Curess RL. Classification and surgical management of the

severe sequelae of septic hips in children. Clin Orthop Relat Res 1982; 171: 30-6.

Jaberi FM, Shahcheraghi GH, Ahadzadeh M. Short-term intravenous antibiotic treatment of acute

hematogenous bone and joint infection in children: a prospective randomized trial. J Pediatr Orthop 2002; 22:

317-20.

Jackson MA, Burry VF, Olson LC. Pyogenic arthritis associated with adjacent osteomyelitis: identification of

the sequel-prone child. Pediatr Infect Dis 1992; 11: 9-13.

Jagodzinski NA, Kanwar R, Graham K, Bache CE. Prospective evaluation of a shortened regimen of

treatment for acute osteomyelitis and septic arthritis in children. J Pediatr Orthop 2009; 29: 518-25.

Jenzri M, Safi H, Nessib MN, Jalel C, Smida M, Ammar C, Ben Ghachem M. Acute hematogenous

osteomyelitis of the neck of femur in children: 28 cases. Rev Chir Orthop Reparatrice Appar Mot 2008; 94:

434-42.

Jones HW, Harrison JW, Bates J, Evans GA, Lubega N. Radiologic classification of chronic hematogenous

osteomyelitis in children. J Pediatr Orthop 2009; 29: 822-7.

Jordanov MI, Block JJ, Gonzalez AL, Green NE. Transarticular spread of Ewing sarcoma mimicking septic

arthritis. Pediatr Radiol 2009; 39: 381-4.

Kallio P, Michelsson JE, Lalla M, Holm T. C-reactive protein in tibial fractures. Natural response to the

injury and operative treatment. J Bone Joint Surg Br 1990; 72: 615-7.

Kallio P, Ryöppy S. Hyperpressure in juvenile hip disease. Acta Orthop Scand 1985; 56: 211-4.

Kallio P, Ryöppy S, Jäppinen S, Siponmaa AK, Jääskeläinen J, Kunnamo I. Ultrasonography in hip disease in

children. Acta Orthop Scand 1985; 56: 367-71.

Kallio MJ, Unkila-Kallio L, Aalto K, Peltola H. Serum C-reactive protein, erythrocyte sedimentation rate and

white blood cell count in septic arthritis of children. Pediatr Infect Dis J 1997; 16: 411-3.

Kan JH, Young RS, Yu C, Hernanz-Schulman M. Clinical impact of gadolinium in the MRI diagnosis of

musculoskeletal infection in children. Pediatr Radiol 2010; 40: 1197-1205.

Kao HC, Huang YC, Chiu CH, Chang LY, Lee ZL, Chung PW, Kao FC, Lin TY. Acute hematogenous

osteomyelitis and septic arthritis in children. J Microbio Infect 2003; 36: 260-5.

73

Kaplan SL, Mason EO, Feigin RD. Clindamycin versus nafcillin or methicillin in the treatment of

Staphylococcus aureus osteomyelitis in children. South Med J 1982; 75: 138-42.

Kare M, Dang A. Staphylococcal toxic shock syndrome. J Assoc Physicians India 2008; 56: 192-4.

Kariminisab MH, Shayesteh Azar M, Sajjadi Saravi m. Surgical intervention for treatment of septic arthritis

in infancy and childhood; a retrospective study. Arch Iran Med 2009; 12: 409-11.

Karwowska A, Davies D, Jadavji T. Epidemiology and outcome of osteomyelitis in the era of sequential

intravenous-oral therapy. Pediatr Infect Dis J 1998; 17: 1021-6.

Kenney W. The prognosis in acute hematogenous osteomyelitis with or without chemotherapy. Surgery 1944;

16: 477-84.

Khan Y, Bhutta ZA. Nutritional deficiencies in the developing world: current status and opportunities for

intervention. Pediatr Clin North Am 2010; 57: 1409-41.

Kharbanda Y, Dhir RS. Natural course of hematogenous pyogenic osteomyelitis. J Postgrad Med 1991; 37:

69-75.

Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic arthritis of the hip in children: an

evidence based prediction algorithm. J Bone Joint Surg Am 1999; 81: 1662-70.

Koistinen P. Akuutista hematogeenistä osteomyeliitistä ennen penisilliinihoitoa ja sen jälkeen. Duodecim

1956; 72: 370.

Kolyvas E, Ahronheim G, Marks MI, Gledhill R, Owen H, Rosenthal L. Oral antibiotic therapy of skeletal

infections in children. Pediatrics 1980; 65: 867-71.

Kortekangas P, Aro HT, Tuominen J, Toivanen A. Synovial fluid leucosytos in bacterial arthritis vs. reactive

arthritis and rheumatoid arthritis in the adult knee. Scand J Rheumatol 1992; 21: 283-8.

Krantz JC. Historical Medical Classics Involving New Drugs. Williams & Wilkins Company, Baltimore.

Baltimore 1974

Krilov LR, McCracken GH. Pediatric infectious diseases. In: Cunha BA, ed.. Antibiotic essentials. 5th ed.

Physicians’ Press 2006: 293-313.

Krogstad P. Osteomyelitis and septic arthritis. In: Feigin RD, Cherry JD, Demmler GJ, Kaplan SL, eds.

Textbook of Pediatric Infectious Diseases. 5th ed. Philadelphia: Saunders, 2004:713-36.

Kunnamo I, Kallio P, Pelkonen P. Incidence of arthritis in urban Finnish children. Arthritis Rheum

74

1986; 29: 1232-8.

Kunnamo I, Kallio P, Pelkonen P, Hovi T. Clinical signs and laboratory tests in the differential diagnosis of

arthritis in children. Am J Dis Child 1987; 141: 34-40.

Kurkela S, Helve T, Vaheri A, Vapalahti O. Arthritis and arthralgia three years after Sindbis virus infection:

clinical follow-up of a cohort of 49 patients. Scand J Infect Dis 2008; 40: 167-73.

Labbe JL, Peres O, Leclair O, Goulon R, Scemama P, Jourdel F, Menager C, Duparc B, Lacassin F. Acute

osteomyelitis in children: the pathogenesis revisited? Orthop Traumatol Surg Res 2010; 96: 268-75.

Lacour M, Duarte M, Beutler A, Auckenthaler R, Suter S. Osteoarticular infections due to Kingella kingae in

children. Eur J Pediatr 1991; 150: 612-8.

Lambert HP, O’Grady FW. Cephalosporins. In: Antibiotic and Chemotherapy. Churchill Livingstone, 6th

edition 1992: 89-135.

Langenskiöld A. Growth disturbance after osteomyelitis of femoral condyles in infants. Acta Orthop Scand

1984; 55: 1-13.

Lavy CB. Septic arthritis in Western and sub-Saharan African children – a review. Int Orthop 2007; 2: 137-

44.

Lavy CB, Lavy VR, Anderson I. Salmonella septic arthritis of the shoulder in Zambian children. J R Coll

Surg Edinnb. 1996; 41: 197-9.

Lavy CB, Peek AC, Manjolo G. The incidence of septic arthritis in Malawian children. Int Orthop 2005; 3:

195-6.

Lavy CBD, Thyoka, M, Mannion S, Pitani A. Aspiration or lavage in septic arthritis? A long-term

prospective study on 150 patients. J Bone Joint Surg Br 2003 85-b: 109.

Lavy CB, Thyoka M, Pitani AD. Clinical features and microbiology in 204 cases of septic arthritis in

Malawian children. J Bone Joint Surg Br 2005; 87: 1545-8.

Leigh DA. Determination of serum and bone concentrations of cephradine and cefuroxime by HPLC in

patients undergoing hip and knee replacement surgery. J Antimicrob Chemother 1989; 23: 877-83.

Le Saux N, Howard A, Barrowman NJ, Gaboyry I, Sampson M, Moher D. Shorter courses of parenteral

antibiotic therapy do not appear to influence responce rates for children with acute hematogenous

osteomyelitis : a systematic review. BMC Infect Dis 2002; 2 : 16.

75

Levick JR. An investigation into the validity of subatmospheric pressure recordings from synovial fluid and

their dependence on joint angle. J Physiol 1979; 289: 55-67.

Lew DP, Waldvogel FA. Osteomyelitis. Lancet 2004; 364: 369-79.

Li HT, Zhang TT, Huang J, Zhou YQ, Zhu JX, Wu BQ. Factors associated with the onutcom of life-

threatening necrotizing pneumonia due to community-acquired Staphylococcus aureus in adult and

adolescent patients. Respiration 2010 [Epub ahead print]

Lindell L, Parkkulainen KV. Osteitis in infancy and early childhood with special reference to neonatal

osteitis. Ann Paediatr Fenn 1960; 6: 34-52.

Lipczyk Z, Niedzielski J, Midel A, Krol R. Management of early symptoms of acute septic hip in neonates

and infants – conservative of surgical? Chir Narzadow Ruchu Ortop Pol 2001; 66: 429-34.

Lorrot M, Fitoussi F, Faye A, Mariani P, Job-Deslandre GF, Bingen E, Bourrillon A. Laboratory studies in

pediatric bone and joint infection. Arch Pediatr 2007; 2: S86-90.

Ma LD, Frassica FJ, Bluemke DA, Fishman EK. CT and MRI evaluation of musculoskeletal infection. Crit

Rev Diagn Imaging 1997; 38: 535-68.

Malcius D, Jonkus M, Kuprionis G, Maleckas A, Monastyreckiene E, Uktveris R, Rinkevicius S, Barauskas

V. The accuracy of different imaging techniques in diagnosis of acute hematogenous osteomyelitis. Medicina

(Kaunas) 2009; 45: 624-31.

Malcius D, Trumpulyte G, Barauskas V, Kilda A. Two decades of acute haematogenous osteomyelitis in

chidren; are there any changes? Pediatr Surg Int 2005; 21: 356-9.

Manadan AM, Block JA. Daily needle aspiration versus surgical lavage for the treatment of bacterial septic

arthritis in adults. Am J Ther 2004; 11: 412-5.

Manimegalai M, Rajamurugan PS, Rajendran CP, Rukmangatharajan S, Rajeshwari S, Ravichandran R,

Vasanthy N. Childhood reactive arthritis. J Clin Rheumatol 2008; 14: 292-3.

Margoli CZ, Ingram DL, Meyer JH. Routine tracheotomy in Hemophilus influenzae type b epiglottitis. J

Pediatr 1972; 81: 1150-3.

Marshall JH, Wilmoth GJ. Pigments of Staphylococcus aureus, a series of triterpenoid carotenoids. J

Bacteriol 1981; 147: 900-13.

Másson A, Gudnason B, Gudmundur K, et al. Osteomyelitis and septic arthritis in Icelandic children during

1996-2005. 5th World Congress of the World Society for Pediatric Infectius Diseases – WSPID. Bangkok,

Nov 15-18, 2007.

76

McClure CA, Ford MW, Wilson JB, Aramini JJ. Pneumococcal conjugate vaccination in Canadian infants

and children younger than five years of age: Recommendations and expected benefits. Can J Infect Dis Med

Microbiol 2006; 17: 19-26.

Mollam RAB, Piggott J. Acute osteomyelitis in children. J Bone Joint Surg 1977; 59-B: 2-7.

Molyneux E, French G. Salmonella joint infection in Malawian children. J Infection 1982; 4:131-8.

Morgan DS, Fisher D, Merianos A, Currie BJ. An 18 year clinical review of septic arthritis from tropical

Australia. Epidemiol Infect 1996; 117: 423-8.

Mowlem R. Cancellous chip bone-grafts. Reports on 75 cases. Lancet 1944; 2: 746.

Moumile K, Merckx J, Glorion C, Pouliquen JC, Berche P, Ferroni A. Bacterial aetiology of acute

osteoarticular infections in children. Acta Paediatr 2005; 94: 419-22.

Mulcaney M, Thakur N, Tocci S, Eberson C. Compartment syndrome in a child secondary to acute

osteomyelitis of the ulna. Pediatr Infect Dis J 2009; 28: 258-9.

Mynter H. Acute infectious osteomyelitis and periostitis. Ann Surg 1889; 10: 87-98.

Nelson JD, Howard JB, Shelton S. Oral antibiotic therapy for skeletal infections in children. I.

Antibiotic concentrations in suppurative synovial fluid. J Pediatr 1978; 92: 131-4.

Nicholas P, Meyers BR, Lewy RN, Hirschman SZ. Concentration of clindamycin in human bone. Antimicrob

Agents Chemother 1975; 8: 220-1.

Ntsiba H, Makosso E, Ngandeu-Sinqwe M, Yala F. Septic arthritis in tropical environment. 176 cases report

in Brazzaville. Mali Med 2006; 21: 49-53.

Nunn TR, Cheung WY, Rollinson PD. A prospective study of pyogenic sepsis of the hip in childhood. J Bone

Joint Surg 2007; 89: 100-6.

Nunn T, Rollison P. Haematogenous pyogenic bone and joint sepsis—reducing avoidable morbidity. S Afr

Med J 2007; 97: 456-60.

OʼBrien T, McManus F, MacAuley PH, Ennis JT. Acute haematogenous osteomyelitis. J Bone Joint Surg Br

1982; 64: 450-3.

Odio CM, Ramírez T, Arias G, Abdelnour A, Hidalgo I, Herrera ML, Bolaňos W, Alpizar J, Alvarez P.

Double blind, randomized, placebo-controlled study of dexamethasone therapy for hematogenous septic

arthritis in children. Pediatr Infect Dis J 2003; 22: 883-8.

77

Onyemelukwe G, Sturruck RD. Septic arthritis in Northern Nigeria. Rheumatol Rehabil 1979; 18: 13-7.

Parsch K, Savvidis E. Coxitis in the newborn and infant. Orthopade 1997; 26: 838-47.

Peek AC, Lavy CB, Thyoka M, Pitani A. Does long-term chloramphenicol cause anaemia in Malawi? Trop

Doct 2006: 36: 114-5.

Peltola H, Kallio MJT, Unkila-Kallio L. Reduced incidence of septic arthritis in children by Haemophilus

influenza type-b vaccination. Implications for treatment. J Bone Joint Surg 1998; 80-B: 471-3.

Peltola H, Unkila-Kallio L, Kallio MJ. The Finnish study group. Simplified treatment of acute

staphylococcal osteomyelitis of childhood. Pediatrics 1997; 99: 846-50.

Peltola H, Räsänen JA. Quantitative C-reactive protein in relation to erythrocyte sedimentation rate, fever,

and duration of antibiotic therapy in bacteraemic diseases of childhood. J Infect 1982; 5: 257-67.

Peltola H, Vahvanen V. Acute purulent arthritis in children. Scand J Infect Dis 1983; 15: 75-80.

Peltola H, Vahvanen V, Aalto K. Fever, C-reactive protein and to erythrocyte sedimentation rate in

monitoring recovery from septic arthritis. J Pediatr Orthoped 1984; 4: 170-4.

Perez-Trallero E, Martin-Herrero JE, Mazon A, Garcia-Delafuente C. Robles P, Iriarte V, Dal-Re R, Garcia-

de-Lomas J. Antimicrobial resistance among respiratory pathogens in Spain: latest data and changes over 11

years (1996-1997 to 2006-2007). Antimicrob Agents Chemother 2010; 54: 2953-9.

Perlman MH, Patzakis MJ, Kumar PJ, Holtom P. The incidence of joint involvement with adjacent

osteomyelitis in pediatric patients. J Pediatr Orthop 2000; 20: 40-3.

Ploug HH. Penicillin resistance of Staphylococcus aureus and its clinical implications. Am J Clin Pathol

1945; 15: 446-51.

Prado SMA, Lizama CM, Pena DA, Valenzuela MC, Viviani ST. Short duration of initial intravenous

treatment in 70 pediatric patients with osteoarticular infections. Rev Chilena Infectol 2008; 25: 30-6.

Prober CG. Oral antibiotic therapy for bone and joint infections. Pediatr Infect Dis 1982; 1: 8-10.

Reed L, Baskett A, Watkins N. Managing children with acute non-traumatic limp: the utility of clinical

findings, laboratory inflammatory markers and X-rays. Emerg Med Australas 2009; 21: 136-42.

Rehm SJ, Tice A. Staphylococcus aureus: methicillin-susceptible S.aureus to methicillin-resistant S.aureus

and vancomycin-resistant S.aureus. Clin Infect Dis 2010; 51 Suppl 2: S176-82.

78

Resti M, Moriondo M, Cortimiglia M, Indolfi G, Canessa C, Becciolini L, Bartolini E, Maria de Benedictis F,

de Martino M, Azzari C. Clin Infect Dis 2010; 51: 1042-9.

Riise OR, Handeland KS, Cvancarova M, Wathne KO, Nakstad B, Abrahamsen TG, Kirkhus E, Flato B.

Incidence and characteristics of arthritis in Norwegian children: a population based study. Pediatrics 2008;

121: e299-306.

Riise OR, Kirkhus E, Handeland KS, Flato B, Reiseter T, Cvancarova M, Nakstad B, Wathne KO.

Childhood osteomyelitis-incidence and differentation from acute onset musculoskeletal features in a

population based study. BMC Pediatr 2008; 8: 45.

Roine I, Faingezicht I, Arguedas A, Herrera JF, Rodriguez F. Serial serum C-reactive protein to monitor

recovery from acute hematogenous osteomyelitis in children. Pediatr Infect Dis J 1995; 14: 40-4.

Roos E. Hiukan vertaavaa operatsioonitilastoa lasaretiolojen valaisemiseksi Suomessa. Duodecim 1894; 10:

149.

Ross JJ, Saltzman CL, Carling P, Shapiro DS. Pneumococcal septic arthritis: review of 190 cases. Clin Infect

Dis 2003; 36: 319-27.

Ross AC, Toltzis P, OʼRiordan MA, Millstein L, Sands T, Redpath A, John C. Frequency and risk factors for

deep focus of infection in children with Staphylococcus aureus bacteremia. Pediatr Infect Dis J 2008; 27:

396-9.

Roth S, Sestan B, Tudor A, Dapić T, Cicvarić T, Miletić D. Bilateral fibular stress fracture in young female

basketball player. J Pediatr Orthop B 2008; 17: 195-8.

Ruebner R, Keren R, Coffin S, Chu J, Horn D, Zaoutis Te. Complications of central venous catheters used for

the treatment of acute hematogenous osteomyelitis. Pediatrics 2006; 117: 1210-5.

Rueda AM, Serpa JA, Matloobi M, Mushtaq M, Musher DM. The spectrum of invasive pneumococcal

disease at an adult tertiary care hospital in the early 21st century. Medicine (Baltimore) 2010; 89: 331-6.

Rusak PS, Vyshpinskii IM, Brodskii BF. Treatment of acute hematogenous osteomyelitis in children.

Khirurgiia (Mosk) 1991; 8: 136-8.

Saphyakhajon P, Joshi AY, Huskins WC, Henry NK, Boyce TG. Empiric antibiotic therapy for acute

osteoarticular infections with suspected methicillin-resistant Staphylococcus aureus or Kingella. Pediatr

Infect Dis J 2008; 27: 765-7.

Schneider L, Ehlinger M, Stanchina C, Giacomelli MC, Gicquel P, Karqer C, Clavert JM. Salmonella

enterica subsp. arizonae bone and joint sepsis. A case report and literature review. Orthop Traumatol Surg

Res 2009; 95: 237-42.

79

Sebastian MR, Lodha R, Kabra SK. Chikungunya infection in children. Indian J Pediatr 2009; 76: 185-9.

Shaw BA, Kasser JR. Acute septic arthritis in infancy and childhood. Clin Orthop Rel Res 1990; 257 :212-5.

Shetty AK, Gedalia A. Management of septic arthritis. Indian J Pediatr 2004; 71: 819-24.

Shetty AK, Kumar A. Osteomyelitis in adolescents. Adolesc Med State Art Rev 2007; 18: 79-94.

Sigerist HE: Grosse Ärzte. Eine Geschichte der Heilkunde in Lebensbildern. J. F. Lehmanns Verlag,

München. Freising – München 1931.

Silverman, M.: Ihmeitä pullossa. Kertomus kymmenestä ihmeitä tekevästä lääkeaineesta ja eräistä

lahjakkaista miehistä, jotka tieten tai tietämättään keksivät ne. Gummerus. Jyväskylä 1944 (Article in

Finnish).

Slabbekoorn M, Horlings HM, van der Meer JT, Windhausen A, van der Sloot JA, Lagrand WK. Neth J Med

2010; 68: 341-7.

Sluga M, Windhager R, Pfeiffer M, Ofner P, Lang S, Dominkus M, Nehrer S, Zoupeck A, Kotz R.

Osteosarcoma and Ewingʼs sarcoma—The most frequent malignant bone tumors in children—therapy and

outcome. Z Orthop Ihre Grenzgeb 2002; 140: 652-5.

Smith G. Alexander Ogston (1844-1929). British Journal of Surgery 1965; 52: 917-20.

Smith JW, Piercy EA. Infectious arthritis. Clin Infect Dis 1995; 20: 225-31.

Smith N. The detection of spinal caries by the Roentgen process. Br Med J 1896; 1849: 1382-3.

Smith SP, Thyoka M, Lavy CBD, Lavy CBD, Pitani A. Septic arthritis of the shoulder in children in Malawi.

A randomised, prospective study of aspiration versus arthrotomy and washout. J Bone Joint Surg Br 2002;

84-B: 1167-72.

Smith T. On acute septic arthritis in infants. St. Bartholomewʼs Hosp Rep 1874; 10: 189.

Smoot EC, Graham DR, Fisk JR, Kucan JO. Development of septic arthritis by hematogenous seeding in a

pediatric patient with burns. J Burn Care Rehabil. 1993; 14: 55-7.

Smythe PM. Haemophilus influenzae meningitis treated with streptomycin. Lancet 1948; 6526: 485-91.

Solari N, Viola S, Pistorio A, Magni-Manzoni S, Vitale R, Ruperto N, Ullmann N, Filocamo G, Martini A,

Ravelli A. Assessing current outcomes of juvenile idiopathic arthritis: a cross-sectional study in a tertiary

center sample. Arthritis Rheum 2008; 59: 1571-9.

80

Spiegel DA, Shestra OP, Rajbhandary T, Bijukachhe B, Sitoula P, Banskota B, Banskota A. Epidemiology of

surgical admissions to a childrenʼs disability hospital in Nepal. World J Surg 2010; 34: 954-62.

Spindler T, Huenges R, Hoppe JE. Acute haematogenous osteomyelitis in childhood – correlation of clinical

aspects, diagnostic parameters, therapy and prognosis. Klin Pädiatr 1998; 210: 56-60.

Spoof A. Asevelvollisuuskatsaus vuonna 1885 Turun ja Porin läänin ensimmäisessä piirissä. Duodecim 1885;

1; 142 (article in Finnish).

Suger G, Schmeltz A, Kinzl L, Liener U. Correction of complex deformities in infectious conditions.

Orthopade 2000; 29: 9-17.

Sulamaa M. Varhaisdiagnoosin ja penisilliinin vaikutus hematogeenisen osteomyeliitin prognoosiin.

Duodecim 1945; 9: 651-60.

Suljagic V, Jevtic M, Djordjevic B, Jovelic A. Surgical site infections in a tertiary health care center:

prospective cohort study. Surg Today 2010; 40: 763-71.

Stranger-Jones YK, Bae T, Schneewind O. Vaccine assembly form surface proteins of Staphylococcus

aureus. Proc Natl Acad Sci USA 2006; 103: 16942-7.

Syrogiannopoulos GA, Nelson JD. Duration of antibiotic therapy for acute suppurative osteoarticular

infections. Lancet 1988; 1: 37-40.

Tekou H, Foly A, Akue B. Current profile of hematogenous osteomyelitis in children at the Tokoin

University Hospital Center in Lome, Togo. Report of 145 cases. Med Trop (Mars) 2000; 60: 365-8.

Tempera G, Furneri PM, Carlone NA, Cocuzza C, Rigoli R, Musumeci R, Pilloni AP, Prenna M, Tufano MA,

Tullio V, Vitali LA, Nicoletti G. Antibiotic susceptibility of respiratory pathogens recently isolated in Italy:

focus on cefditoren. J Chemother 2010: 22: 153-9.

Tetzlaff TR, McCracken GH Jr, Nelson JD. Oral antibiotic therapy for skeletal infections of children. II.

Therapy of osteomyelitis and suppurative arthritis. J Pediatr 1978; 92: 485-90.

Tice AD, Hoaglund PA, Shoultz DA. Risk factors and treatment outcomes in osteomyelitis. J Antimicr

Chemother 2003; 51: 1261-8.

Tiitinen E. Veriteitse syntyvästä luumädästä. Duodecim 1944, Suppl. No 4.

Tindimwebwa G, Adeyokunnu AA, Effiong CE. Acute hematogenous osteomyelitis in childhood. Nigerian

Journal of Paediatrics 1983; 10: 93-7.

81

Torppi P, Uurasmaa P. Akuutin ja kroonisen luumädän nykyisestä hoidosta. Duodecim 1962; 78: 440.

Toziano RR, Roncoroni JM, Lopardo H, Unchalo F, Speranza AM, Seoane S, Bologna R, Groiso JA. Artritis

infecciosa en el paciente pediatrico. Experiencia sobre 135 casos. Medicina Infantil 1993; 1: 64-71.

Tröhler U.: Surgery (modern). Teoksessa: Companion Encyclopedia of the History of Medicine, Volume 2.

Ed. by W. F. Bynum and R. Porter. London and New York 1993, Reprinted 1994.

Tykslab (Prices quoted from the laboratory of Turku University Hospital website). http://www.tykslab.fi/fi/.

Uckay I, Assal M, Legout L, Rohner P, Stern R, Lew D, Hoffmeyer P, Bernard L. Recurrent osteomyelitis

causesd by infection with different bacterial strains without obvious source of reinfection. J Clin Microbiol

2006; 44: 1194-6.

Umer M, Hashmi P, Ahmad T, Ahmed M, Umar M. Septic arthritis of the hip in children—Aga Khan

University Hospital experience in Pakistan. J Pak Med Assoc 2003; 53: 472-8.

Unkila-Kallio L, Kallio MJT, Eskola J, Peltola H. Serum C-reactive protein, erythrocyte sedimentation rate,

and white blood cell count in acute hematogenous osteomyelitis of children. Pediatrics 1994; 93, 59-62. [1]

Unkila-Kallio L, Kallio MJ, Peltola H. The usefulness of C-reactive protein levels in the identification of

concurrent septic arthritis in children who have acute hematogenous osteomyelitis. A comparison with the

usefulness of the erythrocyte sedimentation rate and the white blood cell count. J Bone Joint Surg Am 1994;

76: 848-53. [2]

Van Rossum AM, Wulkan RW, Oudesluys-Murphy AM. Procalcitonin as an early marker of infection in

neonates and children. Lancet Infect Dis 2004; 10: 620-30.

Vidigal Junior EC, Vidigal EC, Fernandes JL. Avascular necrosis as a complication of septic arthritis of the

hip in children. Int Orthop 1997; 21: 389-92.

Vinod MB, Matussek J, Curtis N, Graham HK, Carapetis JR. Duration of antimicrobials in children with

osteomyelitis and septic arthritis. J Paediatr Child Health 2002; 38: 363-7.

Vuorinen HS. Nivelrikko ja muita tuki- ja liikuntaelinten tauteja. In: Tautinen historia, 1st Ed. Tampere:

Vastapaino, 2002: 206-7.

Wang CL, Wang SM, Yang YJ, Tsai CH, Liu CC. Septic arthritis in children: relationship of causative

pathogens, complications, and outcome. J Microbiol Immunol Infect 2003; 36: 41-6.

Waris W. Kaksi polvilumpion luumätätapausta. Duodecim 1938; 54: 1062.

82

Wei HH, Wu Kg, Sy LB, Chen CJ, Tang RB. Infectious endocarditis in pediatric patients: analysis of 19

cases presenting at a medical center. J Microbiol Immunol Infect 2010; 43: 430-7.

Weichert S, Sharland M, Clarke NM, Faust SN. Acute haematogenous osteomyelitis in children: Is there any

evidence for how long we should treat? Curr Opin Infect Dis 2008; 146: 375-80.

Welkon CJ, Long SS, Fischer MC, Alburger PD, Pyogenic arthritis in infants and children; a review of 95

cases. Pediatr Infect Dis 1986; 6: 669-76.

Yagupsky P, Bar-Ziv Y, Howard CB, Dagan R. Epidemiology, etiology and clinical features of septic

arthritis in children younger than 24 months. Arch Pediatr Adolesc Med 1995; 35: 2410-2.

Yagupsky P, Katz O, Peled N. Antibiotic susceptibility of Kingella kingae isolates from respiratory carriers

and patients with invasive infections. J Antimicrob Chemother 2001; 47: 191-3.

Yuan HC, Wu KG, Chen CJ, Tang RB, Hwang BT. Characteristics and outcome of septic arthritis in children.

J Microbiol Immunol Infect 2006; 39: 342-7.

Zaki A, Schreiber EC, Weliky I, Knill JR, Hubsher JA. Clinical pharmacology of oral cephradine. J Clin

Pharmacol 1974; 14: 118-26.

Zaoutis T, Localio AR, Leckerman K, Saddlemire S, Bertoch D, Keren R. Prolonged intravenous therapy

versus early transition to oral antimicrobial therapy for acute osteomyelitis in children. Pediatrics 2009; 123:

636-42.

Zavin JK, Hoffer FA, Rand FF, Teele RL. Joint effusion in children with an irritable hip: US diagnosis and

aspiration. Radiology 1993; 187: 459-63.

ORIGINAL PUBLICATIONS (printed with permission)