Wednesday, August 2, 2017

Infected shoulder replacements - how should they be treated?

Two-stage reimplantation for the treatment of deep infection after shoulder arthroplasty

These authors report their results with two-stage implantation for deep periprosthetic infections in 35 shoulders (10 hemiarthroplasties, 24 anatomic total shoulder arthroplasties,1 reverse total shoulder arthroplasty).

Preoperative testing showed leukocytosis in 1 patient, elevated C-reactive protein concentration in 67%, elevated erythrocyte sedimentation rate in 61%, and positive preoperative aspiration in 69%.

At most recent follow-up (4.1 years) the patient had significant improvements in pain (from 4.4 to 2 on a 5-point scale; P < .0001), mean forward elevation (64°-118°; P < .0001), and mean external rotation (14°-41°; P < .0001).

Persistent infection, defined as positive cultures in samples obtained at the time of reimplantation, was identified in 5 shoulders (15%); 50% of persistent infections grew Propionibacterium acnes.

Reoperations for infection included irrigation and d├ębridement (1), a second 2-stage reimplantation (2), and resection arthroplasty (1); 2 additional patients were treated with chronic suppression. Reoperation for aseptic glenoid loosening was performed in 2 additional shoulders.

There was an overall rate of unsatisfactory results approaching 40%.

Reports of treatment of periprosthetic infections are complicated by (1) the lack of a useful definition of "periprosthetic infection" that includes both obvious presentations (drainage, erythema, elevated white blood counts, sedimentation rate, C-reactive protein) and stealth presentation (no clinical or blood lab evidence of infection, but substantially positive cultures of deep tissues obtained at revision arthroplasty) and (2) the well documented observation that clinical manifestations of infection can present years after a shoulder arthroplasty (so that the 'success' of treatment can be difficult to determine).

Because of the added morbidity of a two-stage approach to a suspected infection, we reserve the two-stage for cases of obvious infection that present, for example, with wound drainage, elevated serum markers of systemic inflammation (CBC, Sed Rate, C-repactive protein), and bacteria other than Propionibacterium and coagulase negative staph. In the remaining cases we prefer a single-stage revision as reported here:

Single-Stage Revision Is Effective for Failed Shoulder Arthroplasty with Positive Cultures for Propionibacterium

These authors point out that cultures taken at the tim of revision shoulder arthroplasty are often positive for Propionibacterium.  They tested the hypothesis that the functional outcomes of revising Propionibacterium culture-positive failed arthroplasties with a single-stage revision and immediate antibiotic therapy are not inferior to the clinical outcomes of revising failed shoulder arthroplasties that are not culture-positive.

Fifty-five shoulders without obvious clinical evidence of infection had a single-stage revision arthroplasty. Specifically all components (humeral and glenoid) were removed, a thorough debridement was carried out and a new humeral hemiarthroplasty was inserted with Vancomycin impregnated allograft. The residual glenoid bone was smoothed, but not bone grafted. No glenoid components were replaced.

Preoperative antibiotics were withheld until culture specimens were taken; a minimum of 5 tissue or explant specimens were obtained from each shoulder. Specimens were cultured for 21 days on blood agar (trypticase soy agar with 5% sheep blood), chocolate agar, Brucella agar (with blood, hemin, and vitamin K), and brain-heart infusion broth. Bacteria that were isolated received a full species-level identification by means of 16S rDNA sequencing.

After all culture specimens were obtained, 15 mg/kg of vancomycin and 2 g of ceftriaxone were administered intravenously. Patients were continued on antibiotics until the results of the cultures were finalized. Two or more cultures became positive, the infectious disease service started intravenous ceftriaxone and/or vancomycin through a PICC line with oral rifampin for 6 weeks followed by oral antibiotics in the form of amoxicillin and clavulanate or doxycycline for a minimum of 6months.

The patient self-assessed functional outcomes for those shoulders with ≥2 positive cultures for Propionibacterium (the culture-positive group) were compared with shoulders with no positive cultures or only 1 positive culture (the control group).

Below is an example of what is referred to as a 'stealth' presentation in which there were no preoperative symptoms or signs of infection, yet the cultures from revision surgery were strongly positive.

The culture-positive group were 89% male with a mean age of 63.5 ± 7.2 years. The mean Simple Shoulder Test (SST) scores for the 27 culture-positive shoulders improved from 3.2 ± 2.8 points before the surgical procedure to 7.8 ± 3.3 points at a mean follow-up of 45.8 ± 11.7 months after the surgical procedure (p < 0.001), a mean improvement of 49% of the maximum possible improvement. 

The control group were 39% male with a mean age of 67.1 ± 8.1 years. The mean SST scores for the 28 control shoulders improved from 2.6 ± 1.9 points preoperatively to 6.1 ± 3.4 points postoperatively at a mean follow-up of 49.6 ± 11.8 months (p < 0.001), a mean improvement of 37% of the maximum possible improvement. 

Subsequent procedures for persistent pain or stiffness were required in 3 patients (11%) in the culture-positive group and in 3 patients (11%) in the control group; none of the revisions were culture-positive. 

The authors concluded that the clinical outcomes after single-stage revision for Propionibacterium culture-positive shoulders were at least as good as the outcomes in revision procedures for control shoulders. Two-stage revision procedures may not be necessary in the management of these cases. 

Fourteen patients reported side effects to antibiotics, indication that patients should be educated with regard to potential antibiotic side effects.

While a two-stage revision may be indicated for the 'obvious' infections, this article suggests that a single stage revision may be sufficient for the management of 'stealth' presentations.

This article carefully distinguishes between (a) the 'obvious' presentation of a shoulder infection with findings such as abnormal blood tests (WBC, ESR, C-reactive protein), erythema, fever, and/or wound drainage from (b) the 'stealth' presentation in which none of these findings are present in shoulder arthroplasties revised for pain, stiffness or component loosening combined with cultures positive for organisms such as Propionibacterium. While in the past some have referred to the second group of  cases as "unexpected positive cultures in revision shoulder arthroplasty", it is now preferable simply to report the clinical findings (i.e. is there obvious clinical evidence of infection?), the number of specimens, and the culture results. Furthermore, since it is not currently possible to distinguish "true infections", "contamination", "false-positive cultures", "non-pathogenic Propionibacterium growth", it is preferable to avoid these terms and, again, reporting the clinical findings, the number of specimens, and the culture results.

Again it is important to recognize that each report regarding periprosthetic infections needs to grapple with the question "What is an infection of a shoulder joint replacement?" as discussed here:

What is a "periprosthetic shoulder infection"? A systematic review of two decades of publications.

These authors point out that while as many as 50% of revision shoulder arthroplasties are culture positive, a consistent, clinically useful definition of a "periprosthetic shoulder infection" is lacking. They conducted a systematic review of the published literature with respect to (1) the definition of a "periprosthetic shoulder infection", (2) the pre-operative evaluation for possible infection, and (3) the harvesting and culturing of specimens at the time of surgical revision.

They found a remarkable lack of consistency in the way different authors defined an 'infection', in the way authors evaluated patients with possible infections before surgery and in the way authors obtained and analyzed specimens obtained for culture harvested at the time of the surgical revision of failed shoulder joint replacements.

This inconsistency makes it very difficult to compare different treatment approaches to failed shoulder joint replacements, recognizing that some of them will have substantial bacteria in the joint, the presence of which may go unrecognized until the culture results are finalized at 2 to 3 weeks after surgery.

Comment: It is critically important not to combine, confuse or commingle data from "obvious infections" (i.e. those with swelling, redness, drainage, fever, chills, elevated serum markers of inflammation) where the diagnosis of infection is apparent with cases of "stealth" presentation (i.e. the unexplained onset of pain and stiffness of the shoulder after a 'honeymoon' of good function in which specimens obtained at revision surgery are strongly positive for organisms such as Propionibacterium).

Here's an example of a stealth presentation:

A 50 year old patient presented desiring a ream and run arthroplasty for severe glenohumeral arthritis

After surgery, the shoulder progressively regained comfort and function. Subsequently, however it started to become stiff and painful without obvious explanation. Eight years after his shoulder arthroplasty, the patient returned to the office with no clinical, laboratory, or radiographic evidence of infection.

A single stage revision was performed (soft tissue releases, prothesis exchange) without any evidence of inflammation, joint fluid, loosening, or osteolysis. Five explant and tissue cultures were sent. The patient was discharge on the yellow protocol (Augmentin) until the results of the cultures were final.

The culture results were
Humeral head explant: 3+ Propionibacterium
Humeral stem explant: no growth
Collar membrane: 1+ Propionibacterium
Humeral periosteum: 1+ Propionibacterium
Joint capsule: no growth

At this point the red protocol (IV ceftriaxone) was started and continued for 6 weeks followed by a 6 month course of Augmentin. The patient has a comfortable shoulder and has regained most of the lost shoulder motion.
The reader may also be interested in these posts:

Information about shoulder exercises can be found at this link.

Use the "Search" box to the right to find other topics of interest to you.

You may be interested in some of our most visited web pages including:shoulder arthritis, total shoulder, ream and runreverse total shoulderCTA arthroplasty, and rotator cuff surgery as well as the 'ream and run essentials'