Bone and joint infections in hospitalized patients in France, 2008: clinical and economic outcomes

https://doi.org/10.1016/j.jhin.2012.04.025Get rights and content

Summary

Background

Adult bone and joint infections (BJIs) often require repeated and prolonged hospitalizations and are considered as a serious public health issue.

Aim

To describe the epidemiology and economical outcomes of BJI in France.

Methods

BJI hospitalizations with selected demographic, medical, and economic parameters from the French national hospital database for the year 2008 were identified. Overall patient characteristics and hospital stays for BJI underwent univariate analysis. Risk factors for device-associated infections were identified using multiple logistic regression modelling.

Findings

Of all hospitalizations in France, 0.2% were BJI-related, representing 54.6 cases per 100,000 population, with a higher prevalence in males (sex ratio: 1.54). BJIs were more often native (68%) than device-associated (32%). The mean age was 63.1 years. Only 39% of hospital discharges had microbiological information coded; Staphylococcus spp. were isolated in 66% of those cases. Obesity, Staphylococcus spp., male sex and age >64 years were important risk factors for device-associated infections, whereas diabetes and ulcer sores were significantly associated with native infections. The case fatality was 4.6%. Intensive care unit stays were needed in 6% of cases. Readmissions to hospital occurred in 19% of cases, with significantly longer stays for device-associated infections than for native BJIs (18.9 vs 16.8 days). The cost of BJIs was €259 million, or about €7,000 per hospitalization in 2008.

Conclusions

This is the largest BJI study to date. The high economic burden of BJIs was mostly associated with more frequent and prolonged hospitalizations, high morbidity, and complexity of care.

Introduction

Bone and joint infections (BJIs), including septic arthritis, osteomyelitis and spondylodiscitis, are a significant cause of morbidity and occasional mortality. These infections can occur on native joints or be device-associated.

Most of the current epidemiological literature on BJI is postoperative and periprosthetic (device-associated); these infections are increasingly more frequent due to higher numbers of arthroplasties.1, 2, 3, 4, 5 The incidence of septic arthritis is estimated in developed countries to be 4–10 per 100,000 population.6 Osteomyelitis, though rare in adults, occurs mainly in children (10 per 100,000).7, 8, 9 The incidence of spondylodiscitis, osteomyelitis of the spine, has been estimated at 2.4 per100,000 in France.10

BJI can become very costly, due to prolonged hospital stays, complicated treatment protocols, and frequent readmissions. The overall length of stay (LOS) for patients who develop BJI after arthroplasties is estimated to be more than twice that of patients with uninfected procedures.11, 12 The economic burden of BJI varies widely across studies, mainly due to differences in country-specific healthcare reimbursement systems.

A few studies have estimated the incidence of BJI; however, these rates are more often limited to reports of one hospital or area-specific hospital system. In France, each hospital discharge (HD), whether from a public or a private establishment, must be registered in the French National Hospital Discharge Database (NHDD). The NHDD is therefore a convenient medico-administrative database that can be mined for epidemiologic studies.

The objective of this study was to describe the epidemiology and the clinical and financial outcomes of hospitalized BJI cases in France using the 2008 NHDD.

Section snippets

Methods

Retrospective analysis of the NHDD for 2008 was conducted. An algorithm was used to create a case definition for BJI (Appendix 1). We selected any HD with a principal or secondary diagnosis of BJI appearing alone or in combination with either sepsis or a specific surgical procedure. International Classification of Diseases, Tenth Revision (ICD-10) codes included arthritis, osteomyelitis, spondylodiscitis and infectious complications of surgical care (T codes). The surgical procedures according

Results

Of the 16 million HDs in France in 2008, a total of 36,091 HDs met the case criteria for BJI hospitalization (0.2%). Ninety-nine percent of HDs were correctly linked to patient database, representing 28,579 patients. Septic arthritis represented 53% of BJIs, followed by osteomyelitis (34%), spondylodiscitis (9%), and infectious complications of orthopaedic device without a specific BJI type indicated (4%, ICD-10 codes T845, T846, and T847). The majority (68%) of BJI discharges was due to native

Discussion

This national study demonstrates that BJI have a major clinical and economic impact. There was an overall BJI prevalence of 54 per 100,000, which is in the range of other studies.9, 10, 14 BJI prevalence was age and sex dependent, increasing six-fold from those aged <50 years to those aged >70 years. Males had a 60% higher rate of infection as compared to females, but among individuals aged >64 years the risk of device-associated BJIs among females was 23% higher than among males, paralleling

Acknowledgements

We thank the French medical doctors who coded the HD, and Prof. P. Astagneau for reading and commenting on the manuscript. This study has been approved by the Commission nationale de l'informatique et des libertés, Paris, France, on 7 July 2010 (n° 1417805).

References (32)

  • R.A. Malinzak et al.

    Morbidly obese, diabetic, younger, and unilateral joint arthroplasty patients have elevated total joint arthroplasty infection rates

    J Arthroplasty

    (2009)
  • A. Tarkowski

    Infection and musculoskeletal conditions: infectious arthritis

    Best Pract Res Clin Rheumatol

    (2006)
  • J. Partanen et al.

    Impact of deep infection after hip fracture surgery on function and mortality

    J Hosp Infect

    (2006)
  • J.N. Katz

    Total joint replacement in osteoarthritis

    Best Pract Res Clin Rheumatol

    (2006)
  • L. Bernard et al.

    Trends in the treatment of orthopaedic prosthetic infections

    J Antimicrob Chemother

    (2004)
  • E.F. Berbari et al.

    Risk factors for prosthetic joint infection: case–control study

    Clin Infect Dis

    (1998)
  • Cited by (97)

    View all citing articles on Scopus

    Presented at the Boston and Chicago Interscience Conference on Antimicrobial Agents and Chemotherapy, 14 September 2010 (paper K-1676, session 185) and 19 September 2011 (paper K-1566, session 200).

    View full text