). Many studies have examined periapical lesions to determine which cells and pro‐inflammatory mediators may be important in the development of cysts, and it is difficult and unnecessary to review them all. For example, Bernardi et al. (2015 ) undertook a review of the literature and identified 187 references to epithelium in periapical lesions between 1975 and 2014. After removing duplicates and simple case reports, they found 42 publications that reported the expression of various biological factors that might be involved in the formation of radicular cysts. Because the overall biological process is one of inflammation and the immune response, the factors are not specific to radicular cysts, and the whole field is constantly evolving. Readers interested in the finer points of the associated immunopathological processes should search for relevant up‐to‐date papers. There are a number of detailed reviews that have helped to inform the following discussion of the key points (Nair 2004 ; Silva et al. 2007 ; Lin et al. 2007 ; Graves et al. 2011 ; Graunaite et al. 2012 ; Marton and Kiss 2014 ; Bernardini et al. 2015 ).
It is now well accepted that the whole process is started by bacterial infection in the necrotic pulp, and that the key initiating event is the action of bacterial endotoxins (lipopolysaccharides, LPS) on periodontal ligament fibroblasts, which then secrete a range of pro‐inflammatory cytokines and chemokines. Meghji et al. (1996 ) studied cyst fluids and cultured cyst explants from radicular cysts, keratocysts, and follicular cysts. They showed high levels of endotoxins in radicular cysts compared with negligible levels in the other cyst types. These endotoxins were LPS derived from Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis, and Escherichia coli. Wayman et al. (1992 ) were able to cultivate bacteria from 51 periapical lesions, 23 of which had had previous endodontic therapy. They found a total of 50 different species, of which over 90% were anaerobes. Bacteria were cultured from both granulomas and cysts with equal frequency, but were seen histologically in only 8 (13%) cases. Tek et al. (2013 ) were only able to culture bacteria from about one‐third of periapical lesions. Nair (1987 ), however, in a detailed ultrastructural analysis, found bacteria in 100% of the root canals of teeth affected by periapical lesions, but in only 4 of 31 cases were bacteria found in the extraradicular tissues. In these cases, 1 was infected with Actinomyces and in 3 the bacteria were isolated on the dentinal wall close to the foramen and were associated with abscess formation with accumulations of PMNs. In a later review of the literature, Nair (2004 ) concluded that periapical lesions rarely harbour bacteria unless secondarily infected. This is supported by Ricucci and Siqueira (2010 ), who studied 106 biopsy specimens of tooth roots with apical lesions. They found evidence of bacteria in only 6% of extraradicular lesions. However, bacterial biofilms were found in 77% of the root canals and were more common in untreated (80%) than treated canals (74%). These authors also showed that biofilms were significantly associated with epithelialised lesions, with a frequency of 95%, 83%, and 69.5% in cysts, abscesses, and granulomas, respectively.
Taken together, these data indicate that periapical periodontitis is a result of endodontic infection, following death of the pulp due to caries. Although bacteria proliferate within the root canal, they are rarely encountered in the periapical lesions, confirming the view that cyst formation is initiated by leakage of endotoxins through the apical foramen.
With regard to the host response, all the expected immune cells have been reported in periapical granulomas, with most studies indicating that T lymphocytes predominate and that among these, T‐helper cells (CD4) predominate over cytotoxic/suppressor cells (CD8) (Skaug et al. 1984 ; Nilsen et al. 1984 ; Liapatas et al. 2003 ). However, the proportions of different cell types may vary between sites in the same lesion, or may depend on the stage (or ‘age’) of the lesion (Matthews and Browne 1987 ). Of particular importance, especially in the balance between protective and destructive tissue responses, is the relative number of the T‐helper cell subsets Th1, Th2, Th17, and regulatory T cells (Tregs) (Marton and Kiss 2014 ). Th1 cells, which secrete a range of pro‐inflammatory cytokines, always predominate in periapical granulomas. Th17 cells are the primary source of interleukin (IL)‐17, which is also pro‐inflammatory (Graunaite et al. 2012 ; Marton and Kiss 2014 ), but their action is moderated by Treg cells, which are a major source of anti‐inflammatory transforming growth factor (TGF)‐β. The Th17/Treg ratio seems to be an important regulator of the intensity of the inflammatory response and may thus determine the extent of tissue destruction (Marton and Kiss 2014 ). Kontiainen et al. (1986 ), Babál et al. (1987 ), and Gao et al. (1988a ) found that suppressor/cytotoxic cells (CD8) dominated. Gao et al. (1988a ), however, also noted that although suppressor/cytotoxic cells outnumbered T‐helper cells in periapical granulomas, T‐helper cells predominated in the cysts that they examined. A more recent study (Liapatas et al. 2003 ) has confirmed most of these findings. It showed that T cells predominated and that helper cells predominated over cytotoxic/suppressor cells in most periapical granulomas and cysts. In cysts, however, they found increased numbers of plasma cells, suggesting that humoral immune reactions may take on a more important role in cysts. These studies illustrate the complex nature of the immune interactions in periapical lesions and may explain the discrepancy in the results reported by different workers.
The proportion of B lymphocytes has been reported to be about 20% (Kontiainen et al. 1986 ), with plasma cells varying from 2% (Kontiainen et al. 1986 ) to 13% (Stern et al. 1982 ). As would be expected, the vast majority (over 75%) of plasma cells express immunoglobulin (Ig)G, with 15–20% IgA and very few IgE or IgM (Stern et al. 1981 ; Smith et al. 1987 ). Other cell types that play a role in periapical lesions include macrophages, mast cells, and Langerhans cells (Pulver et al. 1978 ; Kontiainen et al. 1986 ; Drazic et al. 2010 ). Langerhans cells may be of particular interest because they have been found in the epithelial linings of radicular cysts (Contos et al. 1987 ; Matthews and Browne 1987 ; Gao et al. 1988a ; Liapatas et al. 2003 ) and are most prominent in areas of heavy inflammation. Although they are known to be important in antigen presentation, it has been suggested that they may also be associated with epithelial proliferation (Carillo et al. 2010 ). Non‐immune cells, including fibroblasts, endothelial cells, and epithelial cells, are also present and may be involved in producing relevant cytokines and growth factors. Lesions also contain PMNs and primary or secondary abscess formation may be a key factor in cyst formation (see later in this chapter).
Although the main initiating factor of a periapical lesion is bacterial LPS, the lesions are then sustained by a complex network of inflammatory mediators, including cytokines, chemokines, and the eicosanoids (mainly prostaglandins). These have multiple roles in cell activation and migration, in epithelial proliferation, and in bone resorption. In early lesions there is good evidence that the chemokine CXCL8/IL‐8 is important since it is absent in normal pulp, but found in about 95% of periapical lesions. LPS stimulates periodontal ligament fibroblasts to secrete CXCL8/IL‐8, which has a pivotal role in PMN migration and therefore in lesion initiation (reviewed in Silva et al. 2007 ; Graunaite et al. 2012 ; Marton and Kiss 2014 ). This chemokine is also important in bone resorption. Other pivotal cytokines are the interleukins, especially IL‐1 and IL‐6, which have roles in activating PMNs and lymphocytes, and in bone resorption (reviewed in Graunaite et al. 2012 ). Table 3.2 lists a number of selected biological factors that may be involved in the pathogenesis of radicular cysts. Mostly this list is derived from the reviews mentioned previously (Bernardini et al. 2015 ; Marton and Kiss 2014 ; Graunaite et al. 2012 ; Graves et al. 2011 ; Silva et al. 2007 ; Lin et al. 2007 ; Nair 2004 ). The actions of those factors thought to have important or specific roles in the formation of radicular cysts are discussed in the sections that follow.
Phase of Initiation
It is generally agreed that the source of epithelium for a radicular cyst is the epithelial cell rests of Malassez. During tooth development the epithelial root sheath of Hertwig maps out the shape of the roots and initiates dentine formation. When tooth formation is complete, the sheath disintegrates and epithelial remnants remain in the periodontal ligament as the rest cells of Malassez. There are two common misconceptions about these cells: first, that they are small islands; and second, that they have no normal function. It is now clear that the epithelial remnants form a network or mesh that lies in the periodontal ligament and surrounds or embraces the tooth root.
