DIAGNOSIS OF INFECTION AFTER TOTAL HIP REPLACEMENT

*Savarino, L; *Tarabusi, C; *Pellacani, A; *Giunti, A; +*Baldini, N +Laboratory for Pathophysiology, Istituti Ortopedici Rizzoli, Bologna, Italy

INTRODUCTION Subclinical infection in patients with pain following total hip replacement (THR) is an understimated condition that needs proper consideration, since it mimics aseptic loosening, contributes to periprosthetic osteolysis and implant failure, and necessitates proper treatment.1 We aimed to define the diagnostic reliability of current clinical, imaging and laboratory parameters that are routinely used before revision surgery for the assessment of infection. METHODS A continuous series of 26 subjects who underwent THR revision surgery was considered, including 21 cases diagnosed and treated as aseptic loosening (Group A), and 5 hip revisions with a clinical diagnosis for infection (Group B). Seven subjects were studied at the time of the primary arthroplasty, and used as negative controls (Group C). Table 1 shows the characteristics of the patients. Tc-99m labeled hydroxymethylene diphosphonate [(99m)Tc-HDP] and technetium -99m hexamethylpropyleneamine oxide [(99m)Tc-HMPAO)]-labeled granulocyte scintigraphy, semi-quantitative histology of peri-implant tissues,2 laboratory tests for inflammation (Erytrocyte Sedimentation Rate, Reactive C Protein, Fibrinogen, Peripheral blood cell count), and microbiology were performed. Table 1. Characteristics of the patients with aseptic loosening (Group A), infection (Group B), and pre-surgery (Group C).

Group Group A (#21)

Group B (#5)

Group C (#7)

Sex Males Females

7 14

2 3

2 5

Age (years) (m±sd) 63±12 63±10 54±11 Diagnosis Osteoarthritis

CHD Trauma Osteonecrosis

15 2 3 1

3 1 1

____

2 2 2 1

Follow-up (months) (m±sd) 87±78 39±35 ___ RESULTS Laboratory tests (Table 2) did not give any relevant information for the diagnosis of sub-clinical infection, with the notable exception of fibrinogen blood levels, that, among Group A subjects, were found to be significantly higher (p<0.01) in culture-positive compared to negative cases. Scintigraphy was positive for loosening but negative for infection in all of Group A patients, whereas in 11 cases (52%) a positive culture was unexpectedly obtained. Histology showed conflicting results: PMNs were found only in five out of the eleven culture-positive patients; on the contrary, in two cases the presence of PMNs did not correspond to a positive culture. In Group B patients, both scans were found to be positive, and the diagnosis of infection accordingly confirmed by the microbiological analysis (3 St.aureus and 2 St. epidermidis). All control subjects (Group C) had negative cultures. Table 3 summarizes the results of Group A subjects. Table 2. Comparison (Mann Whitney U test, P<0.05) between the patients with aseptic loosening (group A) and infection (group B), and controls (group C). All the hematological values are expressed as m±SD. P=group A vs B; P*=group A vs C; P**=group B vs C Normal values Group A

#21 Group B

#5 Group C

#7 P P* P**

ESR =15mm/h 29.2±21.6 74.0±26.6 22.4±7.3 <0.05 n.s. <0.05 CRP <0.5mg/dL 0.84±0.98 5.65±4.88 __ <0.05 __ ___ WBCs 4.5-9.5 cell x 103/mm3

6.7±1.6 7.5±1.9 6.6± 1.1 n.s. n.s. n.s.

Fibrinogen 150-400 mg/dL

409.6±71.6

566.8±97.4

371.1±59.8<0.05 n.s. <0.05

Table 3. Diagnostic profile of “non septic” loosened implants Histology

Wear Inflamm. cell score Culture

PMN Lymph Plasma cells

Laboratory tests*

St.epiderm. + Neg + Neg ↑ ESR; fibrinogen St.epiderm. +++ Neg Neg Neg ↑ C-RP; fibrinogen

St.epiderm. +++ Neg Neg Neg ↑ ESR; C-RP Propion.

acnes +++ Neg ++ ++ ↑ fibrinogen

Neg +++ Neg +++ ++ ↑ ESR; C-RP; fibrinogen St.epiderm. +++ ++ ++ Neg ↑ fibrinogen

Neg +++ + + Neg ↑↑ C-RP St.epiderm. +++ + + Neg ↑↑ESR;fibrinogen

Neg ++ Neg Neg Neg ↑↑ ESR; C-RP; fibrinogen St.epiderm. + Neg Neg Neg ↑ ESR; C-RP

Neg +++ Neg ++ Neg ↑ ESR Neg +++ Neg ++ Neg ↑ ESR; fibrinogen Neg +++ Neg Neg Neg Normal values

St.epiderm. +++ +++ +++ + Normal values Neg +++ Neg + Neg Normal values Neg Neg Neg Neg Neg Normal values

St.epiderm. +++ Neg ++ ++ Normal values St.epiderm. Propionib.

+++ ++ ++ Neg ↑↑ ESR; fibrinogen

Neg + + Neg Neg ↑ ESR Neg +++ Neg Neg Neg ↑ ESR

St. aureus + + ++ Neg Normal values Symbols: ESR: ↑15-50,↑↑>50; fibrinogen: ↑400-450, ↑↑>450; C-RP: ↑0.5-2, ↑↑>2; leucocytes: ↑9500-12000, ↑↑>12000. DISCUSSION Infection in THR continues to be a challenging problem: it leads to a long and difficult course for the patient and frequently to a suboptimal functional outcome. In our opinion, a significant portion of failed hip implants that are currently diagnosed as “non septic” may be a result of subclinical infection, mainly due to gram-positive organisms; bacteria can enhance local bone resorption induced by wear debris, in the absence of clinical, radiological and laboratory evidence of infection, contributing to the pathogenesis of periprosthetic osteolysis. The present study was undertaken to evaluate the value of the routine tests for the diagnosis of sub-clinical infections in a continuous series of patients who underwent a THR revision surgery for ‘presumed’ aseptic loosening. We demonstrated that scintigraphy, microbiology, histology and lab tests often show conflicting results, and no single test proves to be sensitive or specific enough to reliably distinguish between aseptic loosening and loosening due to sub-clinical infection. REFERENCES 1. Hanssen AD, Osmon DR, Nelson CL. Prevention of deep

periprosthetic joint infection. Instr Course Lect 1997;46:555-567. 2. Pizzoferrato A, Ciapetti G, Savarino L, Stea S, Tarabusi C. Results

of histopathological grading on 100 cases of HIP prosthesis failure. Biomaterials 1988;9:314-318.

50th Annual Meeting of the Orthopaedic Research Society

Poster No: 1068