Keywords

Introduction

Periprosthetic joint infection (PJI) is the most severe complication in terms of mortality, morbidity and functional disability. The cost of a septic revision surgery is 3–4 times greater than primary total hip arthroplasty , and double that of an aseptic revision surgery [1, 2]. Frequency varies depending on different etiopathogenic factors such as local contamination, hematogenous contamination and well as patient-associated factors. Local contamination is related with the type of surgical theater (laminar air flow and space suits, less than 1%), the existance of previous surgery [3], dislocations, surgical wound secretion [4], skin coverage defect [5], and antibiotic prophylactic non-use that can multiply the infection risk by seven [6]. Hematogenous contamination is related with the existance of distant septic foci (urinary, dental), and multiplies the risk of infection by three [6]. In regard to patient-associated risk factors, the existance of associated diseases, such as diabetes, rheumatoid arthritis, tuberculosis etc.; the abuse of some drugs and toxic-substances, such as corticotherapy, alcoholism, or drug addiction can also increase the risk of infection [7]. Older ages can also produce immunodepresion associated to progressive thymus gland atrophy (the site of T cell maduration), delayed hypersensibility and weaker lymphocyte response [8]. Nutritional alterations including obesity or deficiencies are also related with higher infection rates.

Diagnosis is challenging and it is a frequent cause of repeated surgeries that affect not only soft-tissues around a joint but also bone. Prosthetic joint infection is less frequent in the hip than in the knee, management is significantly different. Systemic antibiotic therapy after positive intraoperative cultures subsequent to a revision procedure, single- or two-stage procedures depending on the general status of the patient, the bone status (stock) receiving the new implant, and the microbiology analysis, all affect outcome. Controversial issues like temporary spacer use, have been reported. The low number of cases and the differences in patient’s characteristics make the analysis of clinical studies the surgeon’s decision very difficult.

Diagnosis of Infection

Risk Factors and Clinical Suspicion

During the last decades, many studies have reported some clinical and laboratory issues to help to identify risk factors in individual patients and enable post-operative evaluation of failed arthroplasty, determining the probability of infection in high versus low risk patients or situations; this information should help the surgeon decide on a more or less aggressive response to of possible infection [9].

Symptoms: Pain

When the surgeon evaluates a patient with a painful hip arthroplasty all important questions should be replanted including whether the pain had presented recently arises, if it is different than before surgery, or whether it could be related to other sites of infection. Pain in infection is not always “inflammatory”, many times “mechanical” pain occurs, when walking and standing, particularly if there is loosening, or a continuous pain that is not alleviated by rest. At this point, it is important to identify “over-demanding” patients (those expecting complete absence of pain and unlimited walking/activity) through the quantity and quality of analgesics, life style, their actual functional limitations, by the use of internationally-accepted patient reported outcomes.

Signs

During physical examination most International Scientific Societies and Consensus Systems include some signs as major criteria of infection like persistent drainage and a chronics in us with or without the abscence of associated issues such as edge necrosis or wound dehiscence [10, 11]. Most authors agree that these signs are critical even a case shows a negative culture. To date, some controversial studies have reported that simply culturing fistula samples is inadequate while others supports the 80% agreement between cultures of fistula and deep tissue infection [12]. Other inflammatory changes (redness, swelling, and/or increase of local temperature) are less specific and require confirmation. When inflammatory changes appear the clinician will have doubts; these signs need confirmation further study.

Blood Analysis

The most commonly-used laboratory markers are increases in of white cells blood counts, erithrocyte sedimentation rate (ESR) and C-reactive protein (CRP). Erithrocyte sedimentation rate is usually higher than 30 in theses patients, but ESR sensitivity is between 66% and 82%, and the specificity between 85% and 90%. A CRP above one has with a sensitivity between 77% and 96%, and specificity between 84% and 92% for infection. However, sensitivity and specificity increase when are combined showing a high positive (83%) and negative (92–100%) predictive value [13,14,15].

The normal elevation of CRP usually decreases after three weeks and ESR after 6 or even 12 weeks. Other comorbidity problems like another infection affecting the urinary or teeth or rheumatological conditions may also elevate these parameters, so the clinician needs to be aware. Other serum markers like procalcitonin or IL-6 have been investigated but further evidence and cost-efectiveness studies are recommended to confirm their usefulnes. Bottner et al. report a not high sensitivity for these markers: 87% for IL6, 33% for Procalcitonin, and 43% for TNF [16]. Recently, D-dimer seems to be another promising and easy to evaluate marker when compared to combined ESR and PCR values [17].

Imaging assessment

Imaging assessment is also controversial, since simple radiographs can show signs of prosthetic radiographic loosening, such as migration of the components, subsidence of the stem, radiolucent lines wider 2 mm around all the contour of the socket and the stem. Suspicion of infection arises when these signs appear early, present a fast progression, and/or there is an absence of mechanical explanation. The appearance of bone cysts and osteolysis unrelated to radiographic polyethylene wear strongely suggest infection when they are focal, rapidly progressive and massive. Early periosteal reaction (periostitis) is also frequent in these patients [18, 19] (Fig. 12.1).

Fig. 12.1
figure 1

Anteroposterior radiographs of hip. Early signs of prosthetic radiographic loosening, such as component migration, stem subsidence, radiolucent lines wider than 2 mm all around the contour of the socket and the stem strongely suggest infection. Bone cysts and osteolysis are unrelated to significant radiographic polyethylene wear

Full size image

The use of Computed tomography(CT) associated with radiographs can be useful in helping to confirm acetabular and femoral bone loss particularly when combined with other signs like secuestrum and cortical defects. Periprosthetic fluid collections can also be diagnosed using ultrasound, CT, or MRI [10, 19].

An even more controversial topic is the use of conventional scintigraphy [20, 21]. Tecnecium and Galium scintigraphy become positive when bone turnover is increased and remain high during first year after any arthroplasty due to secondary implant fixation. Labeled leukocytes [21, 22] have shown high specificity but low sensitivity. Galium scintigraphy is positive when inflammation is present, so it is always positive during secondary fixation of THA. It has high sensitivity (95%) but very poor specificity (20%). This makes it useful for negative diagnosis (to confirm no infection is present). CT images of bone marrow with labeled leukocytes shows increased in sensitivity and specificity (accuracy: 89–98%).

The accuracy of fluorodeoxyglucose positron emission tomography (PET) in detecting differences between aseptic and septic failure has lead to an increased use of this technique, although it is still difficult to access and expensive [10, 23]. There also are doubts about its accuracy, which may be less than that of CT combined with labeled leukocytes.

Articular aspiration

A preoperative analysis of articular liquid has been also recommended. Some technical issues make this management somewhat difficult, requiring an aseptic preparation (operation room), and these are some imaging techniques like radiographs, ultrasonography or CT that can help determine the best sample [14]. A liquid sample in a blood-culture bottle is much better than a swab [24]. Another controversial issue with articular aspiration is that a needle can only obtain planctonical bacteria, and the biofilm is not analyzed.

The leukocyte level diagnostic ranges are quite controversial and many authors have proposed very different thresholds: >1800 [10, 11], >2700, >3000 in chronic cases [25], or >4350. The sensitivity of this parameter is 90% (>1800) and the specificity: 99% (>1800). The proportion of polymorphonuclear(PMN) cells is also very controversial because many authors again have proposed quite different thresholds: >73% [10, 11], >77%, >80%in chronic cases [25], and >85%. PMN level sensitivity is 83% (>73%), and its specificity 93% (>73%). The use of leukocyte esterase reagent strips, offers a good sensitivity (93%) with low specificity (77%), but 33% of the samples are unuseful because of blood or debris [26]. Alpha-defensin determination has a sensitivity of 97–100% and specificity of 95–97% [27, 28]. These new markers are not significantly better than the more traditional ESR, CRP, leukocyte count and/or PMN [27).

Gram staining has a very low sensitivity (50–75%). Aspirate cultures only show a sensitivity of 56–92%. The specificity is 95%, but very low with skin contaminants (S epidermidis, Propionibacterium sp) [14, 29,30,31,32,33,34].

Biopsy. Culture and Diagnosis of Infection

So, biopsy appears to be a better option than aspiration, offering a sensitivity and specificity ranging between 82% and 98% [35], which increases with the use of arthroscopic basketsup to 88–100% [36].

Intraoperative Diagnosis of Infection

Macroscopical Observations. Are They Useful?

An intraoperative diagnosis will usually confirm the preoperative diagnoses. Subjective, macroscopic observations based on the surgical teams experience and sensorial perceptions (eyes and noses) are important. Photographs ought to be taken. The Musculoskeletal Infection Diseases Society criteria includes gross purulence, great amount of liquid, necrotic and devitalized tissues or smell characteristics as signs of infection [10, 25]. To date, some types of bacteria can be suspected based on the appearance of the purulence. For example, a creamy, yellowish-white, abundant pus, suggests S. Aureus; scarce purulence, dirty interfaces, no smell, and undefinite color suggest S. Epidermidis; abundant exudation and “urine-like” smell suggest E Coli, Proteus spp.; a fecal appearance and/or smell suggest Enterobacter, Enterococcus, Klebsiella and Serratia; asewer smell, Pseudomonas spp.; and strong acid/sour smell: Streptococcus spp.

Gram staining has a sensitivity below 17%, so, itis not useful at all [30]. The use of frozen sections can supply the number of PMN cells per high-power microscopic field [37,38,39,40,41,42,43,44,45]. A figure higher than 10 PMN suggest probable infection (implant extraction). From 5 to less than 10 PMN: possible infection. A figure below than 5 PMN: very low probability of infection. However, Gram staining has very low sensitivity in suspected aseptic loosening (50%) and reimplantation surgery (second stage of two-stage exchange) (29%) [38].

Postoperative Diagnosis of Infection

Conventional Cultures

Conventional recommendations for culturing include samples obtained from synovial fluid and tissues (synovial, devitalized soft-tissues, periprosthetic membranes, intramedullary bone content) [31]. Liquids only have planctonic bacteria, while tissuescan provideintracellular as well as sessile (from biofilm) bacteria [46, 47]. At least three periprosthetic tissue specimens, ideally five to six samples must be taken before irrigation is started [11, 14, 30, 31, 48].

Sonication of Retired Implants

Sonication of retired implants has been reported to be of great value due to its sensitivity, specificity and ability to supply material for quantitative cultures and culture-media for slow-growing microorganisms. It is always recommended and the resulting cultures show a sensitivity of 78%, and specificity of 99% [15]. Associated with quantitative cultures and culture-media for slow-growing microorganisms, sensitivity improves but specificity is worse [47]. There is no consensus if as yet on usually-non-pathogenic bacteria obtained by sonication are actually responsible for clinical syndromes? [49, 50].

Pathology of Intraoperative Samples

Pathologists can be very helpful since frozen sections are still probably the most accurate intraoperative test giving the number of PMN cells per high-power microscopic field [51]. Also quantification of sinovial CRP hace been reported to be evaluable as study of frozen sections [52].There are some doubts as to when the neutrophils are the immune response against plantonic bacteria, and macrophages are the immune response against biofilm. But, how do we differentiate between macrophages against wear particles and those against biofilm?. The pathological report on intraoperative samples is affected by the macrophage response to wear particles and biofilm.

Infection Criteria

Different infection criteria have been postulated based on an International Consensus in order to clearly define this complication [25]. Any of the following will indicate a deep infection: (1) There is a sinus tract communicating with the prosthesis; (2) A pathogen is isolated by culture from at least two separate tissue or fluid samples obtained from the affected prosthetic joint; or (3) Four of the following six criteria exist:

  1. 1.

    Elevated serum erythrocyte sedimentation rate (ESR) and serum C-reactive protein (CRP) concentration,

  2. 2.

    Elevated sinovial leukocyte count,

  3. 3.

    Elevated sinovial neutrophil percentage (PMN%),

  4. 4.

    Presence of purulence in the affected joint,

  5. 5.

    Isolation of a microorganism in one culture of periprosthetic tissue or fluid, or

  6. 6.

    More than five neutrophils per high-power field in five high-powe fields observed from histologic analysis of periprosthetic tissue at ×400 magnification.

The presence of a sinus tract, visible pus surrounding the joint without other explanation (e.g. no crystals), acute inflammation on histopathological examination (>5 neutrophils/high-powerfield), >4200 leukocyte per μl and/or >80% polymorphonuclear leukocytes in synovial fluid, growth of the same microorganism in at least two cultures of synovial fluid, peri-prosthetic issue and/or sonication fluid. Patients were classified into acute postoperative (≤1 month after implantation), acute hematogenous (≤3 weeks of infectious symptoms), and chronic PJI (all other situations).

Major criteria are two positive periprosthetic cultures with identical microroganisms, and sinus tract communication with the joint. Minor criteria are: elevated serum ESR and CRP, elevated synovial fluid white blood cell count or leucocyte esterase test strip, elevated PMN synovial fluid, positive histological analysis, a single positive culture.

This complex problem is critically important to making the right decision regarding management.

Treatment of the Infected Total Hip Arthroplasty

Once the infection is diagnosed, the surgeon must keep several considerations in mind. Particularly is important the time element (early or late), the severity of the infection according to the microbiological analysis and the comorbidity and medical situation of the patient. Once surgery is decided, the question of a single- or two-stage procedure, the local status of soft tissues, surgical and hospital environment must be taken into account in order to manage the most important aspect in hip revision reconstruction: bone loss.

A multidisciplinary approach with the Microbiology and Infectious Disease Department is recommended to properly manage the antibiotherapy. In a few cases a suppressive therapy (unexpected intraoperative cultures in an aseptic revision procedure or very fragile patients with previous repeated surgeries) may be indicated, but, in most cases surgery should be done.

According to the origin of infection and biofilm formation, the implant must be removed whenever possible. A single surgical debridement is often considered controversial due to its inefficacy because bacteria create biofilm in less than 24 h, and thus a single- or two-stage procedure may be indicated. Even more controversial, the surgeon must decide which components should be changed: First it was the polyethylene, then the femoral head, more recently all the uncemented and loose components (it is supposed that in the first 2–4 weeks they are not biologically fixed, but this may be not true with hydroxyapatite-coated pieces). The question arises immediately from a formal point of view: Is it a debridement, or is its a “one-stage” exchange?

Despite deciding on whether to do a single- or a two-stage procedure the management of bone loss is critical for the success of the problem. The surgeon must be aware that a bone defect is created after the implant’s removal so while preoperative planning may be helpful it is not definitive. Currently, aggressive debridement is recommended to remove all membranes from the bone as well as all necrotic and devitalized tissue so the infection can be healed before re-implanting the new prosthesis.

Acetabular Bone Defect in an Infected Total Hip Arthroplasty

As previously mentioned, it is critical to remove all necrotic soft-tissue and membranes. The bone defect will be determined after explanting the acetabular and femoral components, a point that will determine the re-implantation stage [53]. Another issue, when a two-stage procedure is performed is the use of spacer. Although widely used for their theoretical advantages in terms of soft tissue management and infection control [54,55,56,57,58,59,60], the high number of intrinsic complications can make considering the good results of not using attractive. The possibility of acetabular erosion when the spacer is left over long times is also a matter for concern [61, 62].

Girdlestone resection-arthoplasty (RA) can be a useful procedure to solve difficult cases such as severe infections with very severe bone defects, however, numerous studies have reported some unsatisfactory outcomes, including painful hips, the necessity of crutch use, severe limps, hip instability, leg length discrepancy (LLD) and increased oxygen consumption.

Considering the published results, the indications for RA of the hip must be strictly limited to non-ambulatory patients, intravenous drug abuse and impossible reimplantation because of new medical comorbidities or technical difficulties in medically compromised patients. Given these difficulties, some patients do not accept their condition as definitive and request conversion of their RA to a THA, hoping to alleviate symptoms and enhance their quality of life. The clinical outcome after conversion to a THA is unpredictable, so patients must be well informed regarding the expected results of this kind of complex procedure. It is uncommon, technically demanding and frequently associated with difficulties that can lead to complications resulting from some preoperative conditions like severe bone defects, osteoporotic bone, or LLD, which may make a shortening femoral osteotomy necessary, and abductor weakness.

Although the surgery may be technically demanding and the number of complications is not low, the clinical outcome for conversion from RA to THR is comparable to conventional revision surgery. Since the surgery is adapted to the bone defect in both groups, the defect may influence outcome more than the conversion procedure itself.

In fact, conversion of an RA, even after several years, can be done with an acceptable rate of complications (Fig. 12.2). Although dislocation is one of the concerns [63], this complication may be the result of inadequate cup positioning or soft tissue tension or both; at conversion, the soft tissues have invariably tightened, making proper testing of hip instability difficult and thus reconstruction of the hip rotation centre and proper management of the bone defect are more determinant of outcome rather than the use of constraint or dual mobility cups [64]. The number of peri-prosthetic fractures is usually low, and limited to intraoperative proximal cracks, which can be solved during the surgery. The circumstance that there is no need to remove a previous implant in an RA conversion may be one of the explanations for the relatively low number of complications, particularly for intraoperative fractures.

Fig. 12.2
figure 2

(a) Anteroposterior radiograph of a hip in a 72 year-old woman shows an infected cementless hip prosthesis. (b) Resection-arthroplasty. (c) Three months later, the hip was reconstructed using a cementless cup and impacting bone grafting with a cemented femoral long-stem showing a good outcome at 10 years

Full size image

Functional outcome is commonly affected in many patients. Limp is frequent due to destruction of the soft tissue and poor active joint movement that may result in only fair improvement after re-implantation. When compared by bone defect and age the clinical outcome can be similar to that of an aseptic revision surgery, but nevertheless, pain, function, and motion are improved after conversion in most series [63, 65].

Age and bone defect, and/or physical status determine acetabular and femoral reconstruction [53, 66, 67]. RA in older low-demand patients may be adequate for some years, but muscular weakness and excessive oxygen consumption can be very disabling in this population [63]. Acetabular bone defect is the most important factor to affect postoperative clinical outcome, function, range of mobility and LLD so adequate reconstruction is critical to improve results in this complex situation. The low number of cases in most series makes it difficult to recommend the optimal interval between RA and conversion [68,69,70,71,72].

Despite THA after RA showing a clinical outcome and radiographic results similar to those obtained in aseptic revision surgery when the hips have similar bone defects, preoperative counseling must stress the influence of the intraoperative bone defect that will determine the surgical procedure and the risk of complications. Nevertheless, the fact that the functional level will be related to any resulting limp and LLD, mean a patients’ high expectations of improved hip function should be tempered to a realistic level.

Reconstruction of the Acetabular Bone Defect in the Infected Hip

Although an infected THA has been considered as a contraindication for bone graft due to the possibility of a potential sequestrum [73], many series have reported acceptable results and contributed to this option now being considered in some cases.

Filling with antibiotic-loaded cement should be recommended only for small cavitary defects and low-demand patients. Despite its efficacy in controlling infection, mechanically done filling can be less safe in a more active population.

Cementless implants are more widely used for small or moderate acetabular bone defects of less than 30% [67] since they provide better mechanical fixation in most cases. Infection healing has been reported to be similar to cemented fixation with contemporary cups, including “Jumbo” cups, and hemispherical tantalum cups (Fig. 12.3); tantalum supplements [19, 74,75,76,77,78]; and antiprotrusio cage plus tantalum augments [79]. Currently, It is not yet established how much healthy bone is necessary for integration using tantalum prosthesis [80].

Fig. 12.3
figure 3

(a) Anteroposterior radiograph of the hip of a 70-year-old woman shows an infected cementless total hip arthroplasty. (b) Resection arthroplasty. (c). The hip was reconstructed using impacting bone grafting with cemented components., with a good clinical outcome at 5 years

Full size image

The most serious cases with large bone acetabular defects will be managed with similar techniques in infected cases and aseptic cases (Fig. 12.4). Reconstruction of large bone defects with impacting bone grafting technique associated with a cemented cup has obtained very good results and low recurrence figures (from 0% to 8%) (Fig. 12.5) [47, 75, 81]. Antibiotic-loaded cements and vancomycin-loaded graft chips are recommended by the original and other centres [82, 83]. As in aseptic cases, the biological restoration of bone stock makes this technique very attractive, particularly in the most active and young patients. Other options like the use of reinforcement rings, trabecular metal augments or custom-made implants along with the allograft have also provided acceptable early results in patients with infection [84].

Fig. 12.4
figure 4

(a) Anteroposterior radiograph of the pelvis of a 83 year-old woman shows a resection-arthroplasty done 28 years earlier due to infection of a total hip arthroplasty. (b) The hip was reconstructed using a tantalum augment associated to bone impacting grafting technique with a cemented cup and bone impacting technique in the femur

Full size image
Fig. 12.5
figure 5

(a) Radiograph shows an infected bipolar tumoral femoral stem. (b) The prosthesis was removed and a spacer implanted. (c) A new femoral stem was implanted associated with a cup using impacting bone grafting technique

Full size image