Lubricants 2015, 3, 394-412; doi:10.

3390/lubricants3020394
OPEN ACCESS

lubricants
ISSN 2075-4442
www.mdpi.com/journal/lubricants
Review

The Synovial Lining and Synovial Fluid Properties after

Joint Arthroplasty
Michael Shang Kung 1, John Markantonis 1, Scott D. Nelson 2 and Patricia Campbell 1,*

1
J. Vernon Luck Sr. M.D. Orthopaedic Research Center, Orthopaedic Institute for Children/UCLA,
Los Angeles, CA 90007, USA; E-Mails: [email protected] (M.S.K.);
[email protected] (J.M.)
2
Departments of Pathology, and Orthopaedic Surgery, UCLA Santa Monica/Orthopaedic Hospital,
Santa Monica, CA 90404, USA; E-Mail: [email protected]

* Author to whom correspondence should be addressed; E-Mail: [email protected];

Tel.: +1-213-742-1134; Fax: +1-213-742-3253.

Academic Editors: J. Philippe Kretzer and Amir Kamali

Received: 18 February 2015 / Accepted: 5 May 2015 / Published: 18 May 2015

Abstract: The lubrication of the cartilaginous structures in human joints is provided by a

fluid from a specialized layer of cells at the surface of a delicate tissue called the synovial
lining. Little is known about the characteristics of the fluids produced after a joint
arthroplasty procedure. A literature review was carried out to identify papers that
characterized the synovial lining and the synovial fluids formed after total hip or knee
arthroplasty. Five papers about synovial lining histology and six papers about the
lubricating properties of the fluids were identified. The cells making up the re-formed
synovial lining, as well as the lining of interface membranes, were similar to the typical
Type A and B synoviocytes of normal joints. The synovial fluids around joint replacement
devices were typically lower in viscosity than pre-arthroplasty fluids but the protein
concentration and phospholipid concentrations tended to be comparable, suggesting that
the lining tissue function was preserved after arthroplasty. The widespread, long-term
success of joint arthroplasty suggests that the lubricant formed from implanted joint
synovium is adequate for good clinical performance in the majority of joints. The role the
fluid plays in component wear or failure is a topic for future study.
Lubricants 2015, 3 395

Keywords: synovial lining; synovial fluid; lubrication; histomorphological

characterization; arthroplasty; implant; systematic review

1. Introduction

1.1. Synovial Tissue

An appreciation of the microanatomy of the lining tissues of the human articular joints is important
to the understanding of the clinical performance of joint replacement devices used to restore function
to degenerative and painful joints. For the normal cartilage-capped articular joint, lubrication of the
bony, cartilaginous, ligamentous, and fibrous tissue structures is provided by a fluid produced from a
specialized layer of cells found at the surface of a delicate tissue called the synovial lining (from ovum
or egg, because of the similarity with egg white). This synovial tissue is composed of specialized
layers of cells starting at the outermost (joint-facing) level with the intima, which overlies a vascular
subintimal layer, and is supported by a fibrous stroma forming the joint capsule [13]. Of note, there is
a wide range of variability in the thickness of the intimal and subintimal layers, the degree of
continuity and cellularity of the intima and the presence of folds and villi (finger-like projections)
depending upon numerous factors including the health of the joint, the age of the host, and the location
of the sample within the joint.
The synovial tissue has been described as controlling the environment of the joint [4]. It does this
by participating in immune reactions to bacteria, phagocytosis (removal of detritus material),
lubrication, and cartilage nutrition [1,5]. In young, healthy synovium, the 12 cell thick intimal layer is
composed of macrophage-like type A cells which are mainly phagocytic and absorptive in nature, and
fibroblast-like type B cells which are secretory in nature [1,2,611]. This intimal layer shows low
degree of cell division, suggesting that many of the lining cells migrate to the synovial surface from
underlying blood vessels and that they are bone-marrow-derived cells [12]. The synovial tissue is able
to respond quickly to injury or disuse. The changes can include thickening of the intimal layer,
increased vascularity and inflammatory cell infiltration of the subsynovial layer, and the formation of
multiple finger-like projections (villi) into the joint. The synovial tissue of an aged individual with
osteoarthritis will look very different grossly and microscopically from that of a healthy
young individual.
The surgical removal of chronically inflamed synovial tissue from arthritic, tuberculoid or
traumatized joints was once a common orthopaedic procedure; fortunately, the synovial tissue can
regenerate [13,14]. It is this regenerative capacity that allows implanted artificial joints to be lubricated
despite surgical removal or damage to the synovial tissues. The microanatomy of the re-formed joint
lining tissue and the quality of the lubricating fluid that it produces are the subjects of the
present review.
Following joint replacement surgery, in addition to the regeneration of the lining of the joint
capsule, membranes that form between the bone and the cement, or bone and implant (known as the
interface) often show a synovium-like microstructure (so-called pseudo-synovial linings) [1519]. It
has been suggested that motion at the interface induces the production of this specialized tissue
Lubricants 2015, 3 396

because it was observed in association with loose implants [20,21]. More recently, it has been shown
that one of the constituents of synovial fluid, hyaluronan (also referred as hyaluronic acid), can induce
the synovial-like interface membrane features [22]. A synovial-like membrane can be produced in the
rat or mouse dorsal air pouch by disrupting the subcutaneous connective tissue by repeated injection of
air [23]. The resulting cavity develops a lining structure as early as six days which has many features
of a synovial lining, including the production of fluid, and this has become a useful experimental
model of the periprosthetic tissue [2325].

1.2. Synovial Fluid

The lubrication for the natural joint is provided by a small amount of viscous fluid consisting of
mainly hyaluronic acid, proteins and other factors such as lubricin and surface-active phospholipid
(SAPL) that allow for effective gliding of the joints [26]. As with the tissue, the characteristics and
properties of the fluid vary considerably as the result of joint health, injury status and other host factors
such as inflammation and hypersensitivity. Injury to the joint may result in an abundance of fluid
(a joint effusion), distension of the tissues and interruption of the normal functions of the synovial
tissue [13]. Under normal biological conditions, the coefficient of friction is only 0.001 to 0.006 and
wear is nearly zero because of the special tribologic properties of cartilage and synovial fluid [27].
Boundary lubrication is thought to be provided by surface-active phospholipid (SAPL) complexed
with lubricin and hyaluronic acid, which allows for load-bearing capacities for the joint [28,29].
Lubricin, a component found to be plentiful in synovial fluid, is thought to be the lubricating molecule
that facilitates the gliding motion of joints [30]. Lubricin may be a carrier of surface-active
phospholipids because the SAPL are super hydrophobic and adsorb to the joint surfaces. Hyaluronic
acid (HA), found abundantly in synovial fluid, while providing a lubrication effect, does not provide
the load-bearing capabilities for joints [31]. Instead, the HA portion is thought to maintain the
viscosity of the synovial fluid [3234] and facilitate the formation of lipid multilayers on the
lubricating surface by complexing with phospholipids, providing lubrication at the joint [35]. The
purpose of this review was to examine the state of knowledge regarding the changes that occur in the
synovial lining and lubrication properties of the surrounding fluid after implantation of a total joint
prosthesis. This review will (1) summarize the characteristics of the synovial lining tissue
histomorphology and (2) the composition of synovial fluid from published studies.

2. Methods

2.1. Synovial Lining Papers

A review of PubMed- and Embase-indexed studies up to November 2014 was carried out to identify
studies regarding the histological characterization of the synovial lining of joints following total hip or
knee replacement (hereafter called implanted tissue). This electronic search was supplemented by a
manual search of the reference lists of the included papers. Conference proceedings, single case
studies, and reviews were not included in the search and only studies in English language were
included. Search strings included various combinations of the terms synovium OR synovial lining OR
synovial-like membrane OR interface membrane OR synovial lining cells OR synovitis OR intima,
Lubricants 2015, 3 397

arthroplasty OR implant OR implanted, histological OR histology, and morphology OR

morphological. Articles of interest were selected if detailed histomorphological characterization of
the synovial lining or the interface membrane was performed and described. Articles were excluded for
the following reasons: in vitro studies, animal studies, studies involving infection or rheumatoid arthritis
exclusively, non-retrieval studies, and non-arthroplasty related studies. Articles that focused on the
changes of the subintimal layers due to wear debris, but without describing the intima, were excluded
from final selection. The resulting potential studies were reviewed, summarized, and critically evaluated
(Figure 1).
Five articles were selected which reported the histological characterization of a synovial lining or a
synovial-like membrane from implanted tissue and evaluated various bearing material combinations.
Tissue specimens from a total of 173 patients were histologically examined and the morphology of the
intimal layer was characterized in detail. Immunohistochemistry was performed using monoclonal and
polyclonal antibodies to identify the cell subsets or the presence of prostaglandins, fibronectin, and
other factors produced by the lining cells. All papers involved retrospective studies from implanted
tissue. A summary of the overall findings of each selected article are presented in Tables 1 and 2.

Figure 1. Flowchart of synovial lining article selection.

Lubricants 2015, 3 398

Table 1. Overall general findings for articles selected describing synovial lining characterization.
Characterization of Synovial-Like Membrane
Presence of
Author Bearing Type 1 Patients Layer Main Synovium Findings
Synovial Lining Cells Particulate Debris
Thickness
in Synovial Lining 2
Polygonal lining cells
Membrane at cement-bone interface with histological
Goldring et al. Large stellate cells with multiple thin
Metal on Polyethylene N = 20 cells 0 and histochemical characteristics of normal synovium
(1983) [17] dendritic processes
found in patients with implant loosening.
some scattered giant cells
Large polygonal cells with eccentric
Study confirmed formation of synovial-like lining at
Goldring et al. nuclei
Metal on Polyethylene N = 41 12 cells 1 bone-cement interface taken from patients with
(1986) [16] Few scattered multinucleated giant
loosened THA components
cells
Large, rounded or teardrop-shaped
Metal on Polyethylene; cells Cemented, press-fit, and biologic ingrowth prostheses
Lennox et al.
Ceramic on N = 61 13 cells Large Dendritic-shaped cells with 1 showed similar formation of pseudosynovial lining at the
(1987) [19]
Polyethylene long cellular processes extending implant-bone interface membrane.
toward surface
OKM1+/OKM5+ macrophage-like The newly formed bone-implant interface membrane
Metal on Polyethylene
Lalor and Revell N = 29, 23 hip; cells closely resembled true synovium and contained
and Polyethylene on 110 cells 1
(1993) [18] 6 knee macrophage-like type A cells and fibroblast type B
Delrin THA and TKA MAB67+ fibroblast-like cells
cells, but not necessarily always in distinct layers
CD163+ macrophage-like cells Tissues from patients revised due to suggested metal
hypersensitivity showed increased proliferation of
Metal on Polyethylene N = 22, 10 with
Burkandt et al. synovial lining cell layer similar to cases with
THA; Metal on synovitis; 12 with 15 cells 1
(2011) [15] D2-40+ fibroblast-like cells rheumatoid arthritis and high-grade synovitis, with
Metal HRA arthroplasty.
2 patients showing paucicellular synovial membrane
covered by a fibrinous exudate.
1 2
Specimens from revised hip arthroplasty unless otherwise noted and this column gives the bearing combinations; the relative amounts of debris noted histologically in
excised synovial lining layer: 0 = none, 1 = rare, 2 = minimal, 3 = moderate, 4 = severe; THA = total hip arthroplasty, HRA = hip resurfacing arthroplasty, TKA = Total
Knee Arthroplasty; Macrophage markers: OKM1, OKM5, CD163+; Fibroblast markers: MAB67, D2-406.
Lubricants 2015, 3 399

Table 2. Overall general findings for articles selected describing synovial fluid properties.
Author Arthroplasty Experiment Key Findings
Mass spectrophotometry and FTIR were performed Methyl esters of hexadecanoic acid, octadecanoic acid,
Costa et al. (2001)
10 UHMWPE hip implants after cyclohexane extraction for adsorbed products on squalene, and of cholesterol were found in the extracts
[36]
the liners. as well as a protein-like material at the surface.
Sufficient SF samples were obtained from 36 index SF from revision TKA tended to have lower viscosity
Mazzucco et al. 58 index TKA; 19 revision TKA; 2 effused
TKA, 14 revised TKA, and 2 effused previous TKA than that from index TKA. The difference was found to
(2002) [37] previous TKA
for flow property examination. not be statistically significant.
Protein and phospholipids were found to have a
SF from 24 index TKA and 7 revised TKA had their
Mazzucco et al. 77 index TKA; 20 revised TKA; 3 effused positive correlation in regards to each other. Protein and
composition of protein, phospholipids and HA
(2004) [38] previous TKA phospholipids were found to have a negative correlation
determined and correlated.
with HA.
8 species of phosphatidylcholine were identified. 3
Gale et al. 38 Metal on Polyethylene THA; 2 Metal on The bearing surfaces of the implants were rinsed and
species of unsaturated phosphatidylcholine
(2007) [39] Polyethylene TKA analyzed by HPLC for phospholipids.
predominated; PLPC, POPC, and SLPC.
Type 1 telemeterized cemented PE cup, 1 The highest peak temperature were observed in the
Patients were monitored doing various physical
Bergmann et al. temperature measurement at neck; Type 2 head of the implant and reached as high as 43.1 C,
activities and the temperatures inside their
Part 1 (2001) [40] telemeterized non-cemented, AC head, PE or greater than what is believed to affect the synovial
telemeterized implant were recorded.
AC cup, titanium shaft fluids lubrication ability.
The model shows that if the cup of an implant is made
Data from Bergmann et al. 2001 part 1 was used to
Bergmann et al. of a material with good conductivity, heat will be
(See above) generate a finite element model to calculate the
Part 2 (2001) [41] transferred away from the synovial fluid, capsule and
steady- state within the implant during walking.
stem towards the acetabular bone.
THA = total hip arthroplasty, TKA = Total Knee Arthroplasty, MS = Mass spectrometry, FTIR = Fourier transform infrared spectroscopy, AC = aluminum
oxide ceramic, PE = polyethylene, SF = synovial fluid, (PLPC) = Palmitoyl-linoleoyl-phosphatidylcholine, POPC = Palmitoyl-oleoyl-phosphatidylcholine,
SLPC = Stearoyl-linoleoyl-phosphatidylcholine (SLPC).
Lubricants 2015, 3 400

2.2. Synovial Lubrication Papers

A review of PubMed studies from the time period of 2000 through 2014 was performed to find
papers relevant to the lubrication characteristics of synovial fluid (SF) after joint arthroplasty. Search
strings included Synovial Fluid AND Prosthetic (136 results), Synovial Fluid AND Arthroplasty (391
results), Synovial Fluid AND Implant (91 results) and Synovial Fluid AND Revision (132 results) for
a total of 750 articles. The article titles and abstracts were examined and papers were excluded for the
following reasons: SF analysis of biomarkers for periprosthetic infections, ion/cytokine/particle/cell
levels, arthritic joint analysis before arthroplasty only, in vitro studies, pain only, wear/friction only,
animal studies, case studies, literature reviews, and surgical procedure and imaging studies only
(Figure 2).

Figure 2. Flowchart of synovial lubrication article selection.

Six suitable papers were selected. Of these papers, four dealt specifically with the composition of
the synovial fluid and the role this had on lubrication, while two discussed the role of increased
temperature on lubricating protein degradation.

3. Results

3.1. Synovial Lining Characterization

Very few papers specifically described the histomorphology of the synovial intima in the
periprosthetic capsular tissues and four of the five selected studies described the synovial-like
membranes at interfaces. A common finding from those four studies of a synovial-like membrane at
the bone-implant or bone-cement interface was the close resemblance to the true synovial lining
membrane in cellular structure and cell type within the surface layer of excised implanted tissue. A
range of histological presentations was also noted in each study, which was more pronounced in the
implanted tissues compared with non-implanted tissues.
Lubricants 2015, 3 401

The study by Goldring et al. [17] was the first which characterized the new membrane that formed
at the bone-cement interface into three distinct histological zones: (1) a synovial-like layer of lining
cells at the cement surface; (2) sheets of histiocytes and giant cells in the subintima; and (3) a fibrous
layer that blended into the bone and becomes continuous with the adjacent bone marrow. The
membrane was lined with large polygonal cells, often with nuclei pointing away from the surface and a
coating of fibrin around the surface. Cytoplasm was abundant and the nuclei were oval dark and
uniform. Occasionally, multinucleated giant cells appeared within the lining cells. In a later study of
prosthetic loosening after total hip arthroplasty, Goldring et al. confirmed the presence of a
synovial-like lining with the same morphology in lining tissues adjacent to the implant [16]. Tissue
cultures from implanted specimens showed three distinct cell types identified as stellate or dendritic
cells, macrophage-like cells with phagocytosed latex or India-ink particles, and elongated
fibroblast-like cells.
In a comparison of tissues from cemented, press-fit and biologic ingrowth prostheses revised due to
aseptic loosening, Lennox et al. [19] described variations in the histomorphology and described three
types of lining tissues. Type 1 linings comprised large, rounded or teardrop-shaped cells that were
13 cell layers thick. Type 2 membranes were composed of weakly stained collagenous matrix
containing large dendritic-shaped cells with eccentrically located nuclei and long cellular processes
extending toward the surface. Type 3 surfaces consisted of thick collagen fibers oriented perpendicular
to the surface. All three surfaces were present on interface membranes from cemented, press-fit, and
biologic ingrowth prostheses.
Lalor and Revell [18] compared the interface membrane of implanted tissue excised during revision
surgery to normal synovium and rheumatoid arthritis tissue controls. They found that the new interface
tissue had a layer of cells adjacent to the implant similar in appearance to normal synovial intima.
This surface layer varied in thickness from 1 to 10 cells deep in a palisading structure, depending on
each case and sample. Antibody staining for macrophage-like type A cells and fibroblast-like B cells
was positive in the implanted tissue and showed the intermixing of the two cell types within the lining,
similar to unimplanted tissue. Unlike the normal synovial lining and rheumatoid arthritis control
tissues, separate distinct layers of macrophages and deeper fibroblasts were not seen in the implanted
tissues, which could be because of the abundant macrophage infiltration of the subintimal area in
response to wear debris.
In another comparison study of non-implanted and implanted tissues by Burkandt et al. [15]
immunohistochemical staining was used to compare tissues from patients with low and high grade
synovitis (inflammation), metal or polyethylene (PE) wear particle-induced synovitis and proliferative
desquamative synovitis (associated with suspected metal hypersensitivity). The authors reasoned that,
since proliferation of the synovial lining cell layer is a characteristic feature of autoimmune joint
diseases, morphological changes of the synovial lining in periprosthetic tissues may indicate the
presence of an immune response. The intimal lining of the PE wear particle-induced synovitis group
was 12 cells thick, comparable to the unimplanted low grade synovitis group. By contrast the tissue
from the high grade unimplanted synovitis tissue lining was five or more cells thick while the tissues
from two metal-on-metal cases with extensive metal wear particle staining had a denuded cellular
lining. The tissues from five cases with suspected metal hypersensitivity had a proliferating, viable
synovial lining containing abundant fibroblast-like cells. Interestingly, the subsynovial layer of those
Lubricants 2015, 3 402

tissues contained lymphocytes, an immune cell that was only seen in the high-grade rheumatoid
synovitis tissues but not in the PE hip tissues. The authors suggested that the morphological
similarities between tissues from patients revised for suspected metal hypersensitivity and patients
with rheumatoid arthritis adds to the evidence that metal hypersensitivity shares characteristic
morphological features with autoimmune diseases of the joints.
Figure 3 shows light micrographs of synovial tissue from an osteoarthritic joint prior to
metal-on-polyethylene joint replacement (Figure 3A) and then following revision of that joint
replacement (Figure 3B) in the same patient. Microscopic examination of both pre- and post-revision
specimens in this patient shows an intimal/synovial lining with an underlying, non-specific,
lymphoplasmacytic, chronic inflammatory infiltrate adjacent to, or centered around blood vessels. The
post-revision specimen displays a denser collagenous subintimal layer (Figure 3B). Also present in the
post-revision tissues are fragments of polarizable polyethylene with accompanying chronic
inflammation including foreign body giant cells (Figure 3C).

(A)

Figure 3. Cont.
Lubricants 2015, 3 403

(B)

(C)

Figure 3. (A) Pre-revision specimen showing synovial lining with intact intimal layer and
non-specific chronic inflammatory infiltrate (H&E 200); (B) post-revision specimen
showing intact intimal layer, denser subintimal collagenous tissue, and non-specific
chronic inflammatory infiltrate (arrows) (H&E 200); and (C) post-revision specimen,
deep to the synovial lining showing multiple fragments of polyethylene with
accompanying chronic inflammatory infiltrate including foreign body giant cells (H&E,
polarized, 200).
Lubricants 2015, 3 404

3.2. Synovial Fluid Properties

There is a large amount of information on the topic of synovial fluid properties and there is also a
substantial body of literature on the tribology of artificial prosthetic joints. One review and analysis of
the last decade of literature reiterated the complex nature of SF as a lubricant in both natural and
artificial joints [42]. Two major mechanisms are theorized to play a role in both natural and artificial
joints; boundary lubrication and fluid film lubrication. Macromolecules within the synovial fluid play a
key role in both forms of lubrication. Another important mechanism of lubrication not addressed by
the articles presented in this review is the theory of self-pressurized hydrostatic lubrication or biphasic
lubrication. Cartilage has a biphasic nature in which a porous elastic matrix is infiltrated by interstitial
fluid [43]. As the interstitial fluid is forced to flow through the permeable matrix a pressure gradient
develops. The cartilage interstitial water supports most of the joint contact load. Pressurization of this
fluid shifts most of the contact load away from the collagen-proteoglycan matrix resulting in a stress
shielding effect and a low friction coefficient [43,44]. It should also be noted that with articulating
cartilage and soft tissue, fluid pressurization is another distinct mechanism, besides boundary and fluid
film lubrication, which is important. However, this review showed that there is little information about
the characteristics of the fluid lubricating prosthetic joints or how alterations in implanted synovial
tissue affect lubricant production, composition or efficiency.
In a study by Costa et al. [36] the constituents of synovial fluids that had been absorbed into the
surfaces of ten revised PE hip implants (5 ethylene oxide/5 gamma ray sterilized in air) that were
implanted for degenerative arthritis and replaced because of aseptic failure after 623 years were
characterized. These products were extracted with cyclohexane then examined by mass
spectrophotometry and Fourier transform infrared (FTIR) analysis was performed on the cross sections
of the implants. Methyl esters of hexadecanoic and octadecanoic acids, of squalene, and of cholesterol
were found in the extracts and a protein-like material was found on the surface of the implants. The
diffusion of these molecules was postulated to cause the semi-crystalline polymer to become
plasticized, especially at the surface where the greatest amount of absorbed material was found. This
could cause changes to the ultimate tensile strength and ultimate elongation of the polymer, which
could lead to a softer surface and decreased resistance to abrasion. The authors noted that accelerated
simulated artificial joint testing may not allow time for the adsorption of synovial fluid constituents.
Joint fluid flow properties likely play an important role in fluid-film lubrication. A study by
Mazzucco et al. [37] examined the flow properties of synovial fluid at the time of total knee
arthroplasty (TKA) and at revision. Fifty-eight samples were obtained from patients undergoing index
TKA surgery for osteoarthritis, 19 from revision TKA in other patients and two samples from aspirated
effused joints that had undergone previous TKA. The volume of 22 samples from TKA, and five from
revision TKA were insufficient for testing. Of the 14 samples from revision TKA, seven were revised
because of wear and seven for mechanical problems not specific to wear. The average age of the tested
patients was 70 years old (42 to 89 years). Synovial fluid at revision had lower viscosity in comparison
to fluid at index TKA, which was more often within the normal range. However, the difference was
not found to be statistically significant by the authors. There was variation in terms of viscosity in both
pre- and post-revision groups and viscosity in both was decreased compared to synovial fluid from
healthy young individuals. The authors suggested that, since hyaluronic acid (HA) is considered the
Lubricants 2015, 3 405

major component of SF that affects its viscosity, the addition of HA to the lubricants used in joint
simulation studies could provide an artificial joint fluid with viscosity more similar to natural SF.
Previous studies performed on animals [45] have shown decreased HA concentrations after joint
replacement which may explain the observations in these revision fluids.
Having explored fluid-film lubrication previously, Mazzucco et al. [38] explored the correlation of
certain molecules associated with boundary lubrication and their association with flow properties.
These boundary lubricants included phospholipids and proteins that passively diffuse through the joint
capsule to reach the synovial fluid via filtration, and lubricin and superficial zone proteins (SZP) that
are actively secreted into the joint fluid by synoviocytes. One hundred joint fluid samples were
obtained, 77 from patients undergoing TKA for osteoarthritis (except 1 from post-traumatic arthritis),
20 from revision TKA, and 3 samples from aspirated effused joints that had previously undergone
TKA. However, the majority of the samples (69) were insufficient in volume for analysis. Of the 14
revision samples examined, 10 were revised for wear related osteolysis and 4 for non-wear specific
mechanical problems. A positive correlation was found between protein concentration and
phospholipid concentration in pre and post-TKA synovial fluid. This indicated that the joint capsule
functioned the same as before surgery and allowed diffusion of these macromolecules from interstitial
fluid to the joint SF. A negative correlation was seen between these two components of SF and
hyaluronic acid i.e., when HA levels were low, protein and phospholipids were high, or when HA
levels were high, protein and phospholipids were low. The authors offered two explanations for this.
One was that when HA was low, the synovial membrane could compensate by allowing increased
entry of proteins into the synovial fluid. Another explanation was that when protein content was high
synoviocytes down-regulate production of HA. Seemingly the ability of the joint capsule to synthesize
SF components and regulate macromolecule filtration is linked to one another. These findings were
found in both pre and post-TKA specimens.
The surface-active phospholipids (SAPLs) in lubricin are believed to play an important role
in boundary lubrication and analyzing the surfaces of revised implants for their presence was
performed by Gale et al. [39]. The bearing surfaces of 38 revised total hips and 2 total knees, all with
metal-on-polyethylene bearings were rinsed and analyzed by HPLC for phospholipids. In total
eight different species of phosphatidylcholines were identified by the study. Palmitoyl linoleoyl
phosphatidylcholine palmitoyl oleoyl phosphatidylcholine and stearoyl linoleoyl phosphatidylcholine
were identified as the phospholipids most likely responsible for boundary lubrication. The authors
recommended that a combination of saturated and unsaturated phosphatidylcholines be added to joint
simulator lubricants to produce an artificial joint fluid that more closely simulated the boundary
lubrication properties of normal SF.
In a novel study using an instrumented total hip prosthesis with a metal-on-polyethylene bearing,
Bergmann et al. [40] measured the temperature resulting from friction at the bearing surfaces caused
by walking and other physical activities. In part 1 of the study, temperatures inside two types of
telemeterized hip prostheses were recorded at 9 locations along the prosthesis length. The implant
comprised a titanium shaft, aluminum oxide ceramic head and a polyethylene cup. This non-cemented
prosthesis was used in 4 patients, one of which has a second contralateral instrumented implant with a
ceramic cup. Steady state temperatures were reached after 60 min of walking. The temperatures often
rose above the critical level (42 C) needed to cause synovial fluid protein degradation and
Lubricants 2015, 3 406

precipitation. The joint fluid components lost to precipitation could result in decreased
lubrication properties.
The authors noted that the volume of the synovial fluid and its lubricating function play a large role
in the generation of heat in the active joint. If the acetabular cup material has good conductivity such
as ceramic and metal, it would facilitate transfer of heat to the acetabular bone and away from the
synovial fluid, capsule, and stem. The same is true of the stem, where a good heat conducting material
like cobalt chromium alloy transfers heat away from the femoral head to the colder part of the implant.
The authors suggested that an implant with better head and cup separation during the swing phase of
walking would allow for better lubrication and heat dissipation. Bergmann et al. [41] noted that if hip
simulators use a constant controlled temperature of 37 C, such conditions may not mimic the rising
and falling temperatures related to variable levels of in vivo joint activity.

4. Discussion

The purpose of this review of the literature was to examine (1) what is known about the synovial
lining tissue histomorphology and (2) the composition of synovial fluid in joints following
implantation of prosthetic implants.
Specifically, the primary focus of this review was to find papers in which the intimal layer of the
synovial lining was characterized because type A and type B synoviocytes reside there, and they are
the cells that contribute to the lubrication properties of the joint. Therefore, stringent criteria were set
in our search for articles of interest. Many key implant retrieval articles described changes in the
synovial tissue in general but were excluded because they focused on the subintimal layer and the
inflammatory responses due to wear debris. Of note, several studies of the histological features of
tissues from metal-on-metal total hips have reported the partial or complete loss of the synovial lining
cells and their replacement by fibrin or by necrotic tissue [4650]. How this might affect the
lubrication of those bearings has not been addressed. In addition, many papers described the synovial
lining in great detail but were excluded because the studies were focused on tissues excised from
arthritic or diseased joints, not implanted joints [13,51].
The membranes that form at the interfaces of implants have been consistently described as
resembling the normal synovial lining. With immunohistochemical methods, implanted synovial-like
membrane cells tested positive for macrophage-like and fibroblast-like markers [1519]. These
findings imply that the formation of an interfacial membrane after implantation is a natural response to
the influence of micromotion or from chemical mediators such as hyaluronic acid. The similarity in the
intimal cell types and arrangement in membranes and joint linings has led to the conclusion that these
tissues are likely to be capable of inflammatory cytokine production leading to bone loss and local
tissue damage in the same way that the rheumatoid synovial pannus is responsible for local tissue
destruction [17].
Under conditions of osteoarthritis and rheumatoid arthritis, synovitis is present and the increase in
cell layer thickness, the excess of synoviocytes and increased fluid production are thought to contribute
to the destruction of articular cartilage, and the formation of bone cysts. The thickening of fibrotic
tissue in the synovial lining and dense cellular infiltrates of lymphocytes and monocytes are common
morphological observations of the diseased synovium [52]. These changes in the synovial lining
Lubricants 2015, 3 407

structure may transform the natural lubrication properties of the synovial fluid after joint replacement
surgery. Delecrin et al. found that the level of total synovial fluid after total arthroplasty increased in a
rabbit model, but the level of hyaluronic acid remained significantly lower than the internal controls [45]
and the decrease in HA after implantation was found to be similar in humans as well [53].
For the selection of papers describing or characterizing implanted joint fluids, we excluded a large
number of articles that focused on clinical management, synovial biomarkers, in vitro simulation
studies and others that mentioned synovial fluid but not from the implanted joint. This left only a
handful of studies of synovial fluids after arthroplasty. These studies reported that even though
post-implantation SF properties are similar to pre-arthroplasty SF properties in some regards,
significant alterations in composition and properties exist after arthroplasty surgery. Phospholipid
overall contribution to boundary lubrications still remains unclear. Initial studies into its role utilized
phospholipases, but it appears these early studies were contaminated with low levels of proteolytic
enzymes. Later studies with proteolytic inhibitors did not show this increase in friction after
phospholipase digestion. Studies performed in vivo in rabbits, however, have shown a reduced
coefficient of frictions when DPPC liposomes and hyaluronic acid were delivered via intraarticular
injections to damaged joints in comparison to only hyaluronic acid injections [54]. Studies have found
that hyaluronic acid plays a key role in fluid film lubrication and the viscosity of the synovial fluid.
Decreased production of HA by type B synoviocytes could explain the decreased viscosity in post
arthroplasty fluid and a reciprocal increase of joint wear [37,38]. Lubricin has previously been
identified as a key protein in boundary lubrication [55]. The role of species other than HA and lubricin,
such as the various phosphatidylcholines, in post arthroplasty lubrication should be explored in future
studies as they may have significance in boundary lubrication. These include hexadecanoic acid,
octadecanoic acids, squalene, and cholesterol. [36] These proteins adsorb at the cobalt chromium
surface to form thin, discontinuous deposited films in vitro. Because increasing protein content
increases film thickness and can be directly correlated with femoral head wear, it seems likely that
patient SF chemistry plays an important role in lubrication [56]. The choice of the optimal implant
bearing materials for a given patient may, in future, be based on a better understanding of this
individual SF chemistry.
One of the themes throughout the small number of studies characterizing post arthroplasty
lubricants was the comparison with artificial joint simulator lubricants. Joint simulator studies have
been performed for decades but there are often large differences between labs in the properties of the
lubricants used to conduct the tests [42,5759]. The conditions that these simulations are run under
also differ markedly from joint function in vivo including the volume and temperature of the fluids.
Several authors noted that artificial lubricant differs from post arthroplasty synovial fluid in many
regards and suggested ways that this could be improved. For example, supplementation of hyaluronic
acid, and a mixture of saturated and unsaturated phosphatidylcholine could lead to an artificial
synovial fluid with properties more similarly seen in vivo. Brandt et al. suggest the addition of
hyaluronic acid and phosphate-buffered saline to alpha-calf serum to be essential constituents in
artificial lubricants [60]. Running the simulators with temperatures measured in vivo and for a longer
duration could lead to tribological conditions that more closely resemble what an artificial joint
experiences in the body [40].
Lubricants 2015, 3 408

In addition, higher serum degradation and larger wear particles were observed with smaller fluid
volumes used for testing [61]. Despite the differences between in vitro and in vivo joint lubrication, the
overall success of joint replacement components shows that the fundamental requirements for clinical
use have been met.

5. Conclusions

There was a surprising lack of studies on implanted synovial tissue and the lubricating fluid it
produces, although that tissue presumably plays an important role in the tribology and success of joint
arthroplasty. The synovial lining tissue regenerates after implantation and produces a lubricating fluid
that is sufficient in volume and lubricating constituents to allow the majority of joint replacements to
function successfully, possibly for decades. The exact conditions for well-functioning implants are
unclear; however, preserving the integrity of the joint including the synovial lining and natural
lubricating fluid properties for joint arthroplasty may be key to successful implant survivorship.
Whether the degree of implant wear or if some cases of failure can be attributed to any synovial tissue
or lubricant deficiency are questions that remain to be answered and warrant future investigation.

Author Contributions

Michael Shang Kung and John Markantonis performed the systematic review. Patricia Campbell
and Scott D. Nelson designed the study, assisted with the literature search and reviewed the histology.
All authors were involved in the manuscript preparation and editing for final submission.

Conflicts of Interest

The authors declare no conflict of interest.

References

1. Ghadially, F.N. Fine Structure of Synovial Joints: A Text and Atlas of the Ultrastructure of
Normal and Pathological Articular Tissues; Butterworths: London, UK; Boston, MA, USA, 1983.
2. Ghadially, F.N.; Roy, S. Ultrastructure of Synovial Joints in Health and Disease; Butterworths:
London, UK, 1969.
3. Key, J.A. The Synovial Membrane of Joints and Bursae, 2nd ed.; Paul, B., Hober: New York, NY,
USA, 1932.
4. Bronner, F.; Farach-Carson, M.C. Bone and Osteoarthritis; Springer-Verlag: London, UK, 2007;
Volume 4.
5. Pavlovich, R.I.; Lubowitz, J. Current concepts in synovial tissue of the knee joint. Orthopedics
2008, 31, 160163.
6. Athanasou, N.A.; Quinn, J.; Heryet, A.; Puddle, B.; Woods, C.G.; McGee, J.O. The
immunohistology of synovial lining cells in normal and inflamed synovium. J. Pathol. 1988, 155,
133142.
7. Barland, P.; Novikoff, A.B.; Hamerman, D. Electron microscopy of the human synovial
membrane. J. Cell Biol. 1962, 14, 207220.
Lubricants 2015, 3 409

8. Iwanaga, T.; Shikichi, M.; Kitamura, H.; Yanase, H.; Nozawa-Inoue, K. Morphology and
functional roles of synoviocytes in the joint. Arch. Histol. Cytol. 2000, 63, 1731.
9. Mapp, P.I.; Revell, P.A. Fibronectin production by synovial intimal cells. Rheumatol. Int. 1985, 5,
229237.
10. Matsubara, T.; Spycher, M.A.; Ruttner, J.R.; Fehr, K. The ultrastructural localization of
fibronectin in the lining layer of rheumatoid arthritis synovium: The synthesis of fibronectin by
type B lining cells. Rheumatol. Int. 1983, 3, 7579.
11. Yielding, K.L.; Tomkins, G.M.; Bunim, J.J. Synthesis of hyaluronic acid by human synovial
tissue slices. Science 1957, 125, 1300.
12. Hogg, N.; Palmer, D.G.; Revell, P.A. Mononuclear phagocytes of normal and rheumatoid
synovial membrane identified by monoclonal antibodies. Immunology 1985, 56, 673681.
13. Depalma, A.F. Diseases of the Knee: Management in Medicine and Surgery, 1st ed.; J.B.
Lippincott Company: Philadelphia, PA, USA, 1954.
14. Key, J.A. The reformation of synovial membrane in the knees of rabbits after synovectomy. J.
Bone Joint Surg. 1925, 7, 793813.
15. Burkandt, A.; Katzer, A.; Thaler, K.; Von Baehr, V.; Friedrich, R.E.; Ruther, W.; Amling, M.;
Zustin, J. Proliferation of the synovial lining cell layer in suggested metal hypersensitivity. In Vivo
2011, 25, 679686.
16. Goldring, S.R.; Jasty, M.; Roelke, M.S.; Rourke, C.M.; Bringhurst, F.R.; Harris, W.H. Formation
of a synovial-like membrane at the bone-cement interface. Its role in bone resorption and implant
loosening after total hip replacement. Arthritis Rheum. 1986, 29, 836842.
17. Goldring, S.R.; Schiller, A.L.; Roelke, M.; Rourke, C.M.; ONeil, D.A.; Harris, W.H. The
synovial-like membrane at the bone-cement interface in loose total hip replacements and its
proposed role in bone lysis. J. Bone Joint Surg. 1983, 65, 575584.
18. Lalor, P.A.; Revell, P.A. The presence of a synovial layer at the bone-implant interface: An
immunohistological study demonstrating the close similarity to true synovium. Clin. Mater. 1993,
14, 91100.
19. Lennox, D.W.; Schofield, B.H.; McDonald, D.F.; Riley, L.H., Jr. A histologic comparison of
aseptic loosening of cemented, press-fit, and biologic ingrowth prostheses. Clin. Orthop. Relat.
Res. 1987, 171191.
20. Drachman, D.B.; Sokoloff, L. Role of movement in embryonic joint development. Dev. Biol.
1966, 14, 401420.
21. Engh, C.A.; Oconnor, D.; Jasty, M.; Mcgovern, T.F.; Bobyn, J.D.; Harris, W.H. Quantification of
implant micromotion, strain shielding, and bone-resorption with porous-coated anatomic
medullary locking femoral prostheses. Clin. Orthop. Relat. Res. 1992, 285, 1329.
22. Konttinen, Y.T.; Li, T.F.; Mandelin, J.; Ainola, M.; Lassus, J.; Virtanen, I.; Santavirta, S.;
Tammi, M.; Tammi, R. Hyaluronan synthases, hyaluronan, and its CD44 receptor in tissue around
loosened total hip prostheses. J. Pathol. 2001, 194, 384390.
23. Edwards, J.C.; Sedgwick, A.D.; Willoughby, D.A. The formation of a structure with the features
of synovial lining by subcutaneous injection of air: An in vivo tissue culture system. J. Pathol.
1981, 134, 147156.
Lubricants 2015, 3 410

24. Pap, G.; Machner, A.; Rinnert, T.; Horler, D.; Gay, R.E.; Schwarzberg, H.; Neumann, W.;
Michel, B.A.; Gay, S.; Pap, T. Development and characteristics of a synovial-like interface
membrane around cemented tibial hemiarthroplasties in a novel rat model of aseptic prosthesis
loosening. Arthritis Rheum. 2001, 44, 956963.
25. Sedgwick, A.D.; Sin, Y.M.; Edwards, J.C.W.; Willoughby, D.A. Increased inflammatory
reactivity in newly formed lining tissue. J. Pathol. 1983, 141, 483495.
26. Radin, E.L.; Paul, I.L.; Swann, D.A.; Schottstaedt, E.S. Lubrication of synovial membrane.
Ann. Rheum. Dis. 1971, 30, 322325.
27. Konttinen, Y.T.; Zhao, D.; Beklen, A.; Ma, G.; Takagi, M.; Kivela-Rajamaki, M.; Ashammakhi, N.;
Santavirta, S. The microenvironment around total hip replacement prostheses. Clin. Orthop.
Relat. Res. 2005, 430, 2838.
28. Hills, B.A.; Butler, B.D. Surfactants identified in synovial fluid and their ability to act as
boundary lubricants. Ann. Rheum. Dis. 1984, 43, 641648.
29. Seror, J.; Zhu, L.; Goldberg, R.; Day, A.J.; Klein, J. Supramolecular synergy in the boundary
lubrication of synovial joints. Nat. Commun. 2015, 6, doi:10.1038/ncomms7497.
30. Radin, E.L.; Swann, D.A.; Weisser, P.A. Separation of a hyaluronate-free lubricating fraction
from synovial fluid. Nature 1970, 228, 377378.
31. McCutchen, C.W. Joint lubrication. Bull. Hosp. Jt. Dis. Orthop. Inst. 1983, 43, 118129.
32. Balazs, E.A. The Physical Properties of Synovial Fluid and the Specific Role of Hyaluronic Acid;
J B Lippincott: Philadelphia, PA, USA, 1982.
33. Balazs, E.A.; Watson, D.; Duff, I.F.; Roseman, S. Hyaluronic acid in synovial fluid. I. Molecular
parameters of hyaluronic acid in normal and arthritis human fluids. Arthritis Rheum. 1967, 10,
357376.
34. Ogston, A.G.; Stanier, J.E. The physiological function of hyaluronic acid in synovial fluid;
viscous, elastic and lubricant properties. J. Physiol. 1953, 119, 244252.
35. Wang, M.; Liu, C.; Thormann, E.; Dedinaite, A. Hyaluronan and phospholipid association in
biolubrication. Biomacromolecules 2013, 14, 41984206.
36. Costa, L.; Bracco, P.; del Prever, E.B.; Luda, M.P.; Trossarelli, L. Analysis of products diffused
into UHMWPE prosthetic components in vivo. Biomaterials 2001, 22, 307315.
37. Mazzucco, D.; McKinley, G.; Scott, R.D.; Spector, M. Rheology of joint fluid in total knee
arthroplasty patients. J. Orthop. Res. 2002, 20, 11571163.
38. Mazzucco, D.; Scott, R.; Spector, M. Composition of joint fluid in patients undergoing total knee
replacement and revision arthroplasty: Correlation with flow properties. Biomaterials 2004, 25,
44334445.
39. Gale, L.R.; Chen, Y.; Hills, B.A.; Crawford, R. Boundary lubrication of joints: Characterization of
surface-active phospholipids found on retrieved implants. Acta Orthop. 2007, 78, 309314.
40. Bergmann, G.; Graichen, F.; Rohlmann, A.; Verdonschot, N.; van Lenthe, G.H. Frictional heating
of total hip implants. Part 1: Measurements in patients. J. Biomech. 2001, 34, 421428.
41. Bergmann, G.; Graichen, F.; Rohlmann, A.; Verdonschot, N.; van Lenthe, G.H. Frictional heating
of total hip implants. Part 2: Finite element study. J. Biomech. 2001, 34, 429435.
Lubricants 2015, 3 411

42. Ghosh, S.; Choudhury, D.; Das, N.S.; Pingguan-Murphy, B. Tribological role of synovial fluid
compositions on artificial jointsA systematic review of the last 10 years. Lubr. Sci. 2014, 26,
387410.
43. Bonnevie, E.D.; Baro, V.J.; Wang, L.; Burris, D.L. Fluid load support during localized indentation
of cartilage with a spherical probe. J. Biomech. 2012, 45, 10361041.
44. Caligaris, M.; Ateshian, G.A. Effects of sustained interstitial fluid pressurization under migrating
contact area, and boundary lubrication by synovial fluid, on cartilage friction. Osteoarthritis
Cartilage 2008, 16, 12201227.
45. Delecrin, J.; Oka, M.; Takahashi, S.; Yamamuro, T.; Nakamura, T. Changes in joint fluid after
total arthroplasty. A quantitative study on the rabbit knee joint. Clin. Orthop. Relat. Res. 1994,
307, 240249.
46. Campbell, P.; Ebramzadeh, E.; Nelson, S.; Takamura, K.; De Smet, K.; Amstutz, H.C.
Histological features of pseudotumor-like tissues from metal-on-metal hips. Clin. Orthop. Relat.
Res. 2010, 468, 23212327.
47. Davies, A.P.; Willert, H.G.; Campbell, P.A.; Learmonth, I.D.; Case, C.P. An unusual lymphocytic
perivascular infiltration in tissues around contemporary metal-on-metal joint replacements. J.
Bone Joint Surg. 2005, 87, 1827.
48. Grammatopoulos, G.; Pandit, H.; Kamali, A.; Maggiani, F.; Glyn-Jones, S.; Gill, H.S.; Murray,
D.W.; Athanasou, N. The correlation of wear with histological features after failed hip resurfacing
arthroplasty. J. Bone Joint Surg. 2013, 95, e81.
49. Howie, D.W.; Cain, C.M.; Cornish, B.L. Pseudo-abscess of the psoas bursa in failed double-cup
arthroplasty of the hip. J. Bone Joint Surg. 1991, 73, 2932.
50. Willert, H.G.; Buchhorn, G.H.; Fayyazi, A.; Flury, R.; Windler, M.; Koster, G.; Lohmann, C.H.
Metal-on-metal bearings and hypersensitivity in patients with artificial hip joints. A clinical and
histomorphological study. J. Bone Joint Surg. 2005, 87, 2836.
51. Hirohata, K.; Kobayashi, I. Fine structures of the synovial tissues in rheumatoid arthritis. Kobe J.
Med. Sci. 1964, 10, 195225.
52. Haraoui, B.; Pelletier, J.P.; Cloutier, J.M.; Faure, M.P.; Martel-Pelletier, J. Synovial membrane
histology and immunopathology in rheumatoid arthritis and osteoarthritis. In vivo effects of
antirheumatic drugs. Arth. Rheum. 1991, 34, 153163.
53. Walker, P.S. A comparison of normal and artificial human joints. Acta Orthop. Belg. 1973, 39,
4354.
54. Jay, G.D.; Waller, K.A. The biology of lubricin: Near frictionless joint motion. Matrix Boil. J. Int.
Soc. Matrix Biol. 2014, 39, 1724.
55. Swann, D.A.; Silver, F.H.; Slayter, H.S.; Stafford, W.; Shore, E. The molecular structure and
lubricating activity of lubricin isolated from bovine and human synovial fluids. Biochem. J. 1985,
225, 195201.
56. Fan, J.; Myant, C.; Underwood, R.; Cann, P. Synovial fluid lubrication of artificial joints: Protein
film formation and composition. Faraday Discuss. 2012, 156, 6985.
57. Gispert, M.P.; Serro, A.P.; Colaco, R.; Saramago, B. Friction and wear mechanisms in hip
prosthesis: Comparison of joint materials behaviour in several lubricants. Wear 2006, 260,
149158.
Lubricants 2015, 3 412

58. Roba, M.; Bruhin, C.; Ebneter, U.; Ehrbar, R.; Crockett, R.; Spencer, N.D. Latex on glass: An
appropriate model for cartilage-lubrication studies? Tribol. Lett. 2010, 38, 267273.
59. Wang, A.; Essner, A.; Schmidig, G. The effects of lubricant composition on in vitro wear testing
of polymeric acetabular components. J. Biomed. Mater. Res. B Appl. Biomater. 2004, 68, 4552.
60. Brandt, J.M.; Briere, L.K.; Marr, J.; MacDonald, S.J.; Bourne, R.B.; Medley, J.B. Biochemical
comparisons of osteoarthritic human synovial fluid with calf sera used in knee simulator wear
testing. J. Biomed. Mater. Res. A 2010, 94, 961971.
61. Reinders, J.; Sonntag, R.; Kretzer, J.P. Synovial fluid replication in knee wear testing: An
investigation of the fluid volume. J. Orthop. Res. 2015, 33, 9297.

2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).