Uncertainty about the causes and natural history of salivary stones (sialoliths) and other obstructions is being dispelled by clinical and experimental research. Sialoliths are now shown to be secondary to chronic obstructive sialadenitis. Microscopic stones (sialomicroliths) accumulate during secretory inactivity in normal salivary glands and produce atrophic foci by obstruction. Microbes ascend the main salivary duct during secretory inactivity and proliferate in atrophic foci and cause spreading inflammation, leading to inflammatory swelling and fibrosis that can compress large ducts. This leads to stagnation of secretory material rich in calcium that precipitates onto degenerating cellular membranes to form a sialolith.
Historical review
A brief historical review, which begins with Küttner in the late nineteenth century and finishes with research in progress, is necessary to appreciate current understanding. A detailed historical review has been presented elsewhere.
Clinical Investigations
Küttner in 1896 published the results of his microscopic examinations of two patients’ chronically swollen submandibular glands that attracted a clinical diagnosis of malignancy. He realized that the swelling was caused by chronic inflammation that, together with fibrosis, led to a clinical appearance of malignancy. He found a sialolith the size of a cherrystone associated with the gland in one of the patients, who had complained of a submandibular swelling for 10 years. He considered the sialolith secondary to the inflammation because the sialolith was far too small for a concretion of 10 years’ accumulation and because of the absence of a sialolith in the other case, in which the duration was only 1.5 months. Küttner was of the opinion that chronic sialadenitis is primary and arises by inflammation that ascends Wharton’s duct from the mouth. He considered three ways in which a sialolith could form: inflammation may roughen the lining of ducts and precipation occurs on this; inflammation may compress the intraglandular collecting ducts of lobules and precipitation occurs in the obstructed lobules; and precipitation may occur in bacterial deposits.
Küttner’s seminal publication in 1896 established chronic obstructive sialadenitis as an entity, which became known in continental Europe as Küttner’s tumor. Küttner’s subsequent practice and research confirmed his opinion that sialoliths are secondary to sialadenitis (see his Handbuch der Praktischen Chirurgie , 1926) ( Fig. 1 ). Rauch, in his monograph, Die Speicheldrüsen des Menschen , in 1959, considered Küttner’s tumor a mycosis and did not mention Küttner’s opinion about sialoliths. Instead, he described various other theories to explain the production of sialoliths.
Küttner’s work, however, was developed by Seifert and Donath in their 1977 clinicopathologic investigation of chronic submandibular sialadenitis. They divided it into progressive stages that ranged from focal sialadenitis to severe chronic sialadenitis with fibrosis. The first stage occurs when sialomicroliths cause obstruction of small intraglandular ducts, followed by an inflammatory reaction. In the subsequent stages, there is increasing atrophy, fibrosis, and inflammation. The decreased secretory activity of glandular atrophy facilitates ascending invasion by microbes that sustain the inflammation, thus creating a vicious circle.
A major problem of the classification into histologic stages is that the overall microscopic appearance is graded, although there can be great variation between the different features that make up this appearance, even within different parts of the same gland. Harrison and colleagues analyzed this problem by investigating 154 cases of chronic submandibular sialadenitis and statistically reviewing 18 different clinical and histologic features.
This investigation found that sialoliths, atrophy, fibrosis, and other histologic features are all related to inflammation. Inflammation is of the greatest importance in the progression of sialadenitis and the development of sialoliths. Inflammation, atrophy, fibrosis, and sialoliths are all related to the duration of symptoms, which supports Seifert and Donath’s concept of a chronologic progression through increasingly severe histologic stages with secondary production of sialoliths.
Previous investigations of chronic submandibular sialadenitis, however, did not reveal the etiologic factors that could transform a normal gland into a diseased gland. The 1970s, observations by Scott that sialomicroliths and foci of obstructive sialadenitis occur in normal submandibular glands were what eventually led Harrison and colleagues to find this missing link. A postmortem investigation found sialomicroliths in all normal submandibular glands and in a minority of normal parotids ( Fig. 2 ). These findings correspond to a higher concentration of calcium in the submandibular gland. Calcium is sequestered in secretory granules, where it is present as a cationic shield that allows the condensation of acidic secretory glycoprotein ( Fig. 3 ). The level of calcium can be far higher than in the serum and corresponds to the acidity of the glycoprotein in the secretory granules, which is greater in the submandibular gland.
Harrison and colleagues investigated the hypothesis that sialomicroliths impact in ducts and accrete to form sialoliths. A search for sialomicroliths in cases of chronic submandibular sialadenitis was successful ( Figs 4 and 5 ). Although this success supported the hypothesis, Harrison and colleagues were unable to find any relation between sialomicroliths and sialoliths or between sialomicroliths and duration of symptoms in chronic submandibular sialadenitis. This was a surprise and did not support the hypothesis that sialomicroliths are inchoate sialoliths.
The mystery was solved only after experimental investigations.
Experimental Investigations
Animal experiments led to major breakthroughs in understanding the origins and pathogenesis of sialadenitis and sialolithiasis and much of this work relates to sialomicrolith formation.
The earliest model for the investigation of the origins and pathogenesis of sialadenitis and sialolithiasis was the salivary glands of rat. Sialomicroliths and obstructive sialadenitis were produced in the submandibular and parotid glands of rats made hypercalcemic and given repeated high doses of isoprenaline. Isoprenaline given in repeated high doses soon produces a great increase in the size and weight of the submandibular and parotid glands of rats as a result of hyperplasia and hypertrophy of the acinar cells. The acinar enlargement is sufficient to result in compression of the intraglandular ducts. Every dose of isoprenaline is followed by an explosive release of secretory material from the acinar cells, which is unable to flow freely through the lumina of the compressed ducts, and the resultant increase of luminal pressure damages acinar cells. The partial obstruction thus results in a mixture of stagnant secretory material, which is particularly rich in calcium because of the hypercalcemia, and cellular debris. An important component of the cellular debris appertaining to calcification is damaged membranes. Cellular membranes contain phospholipid that becomes exposed when they are damaged. This exposed phospholipid is the potent nucleator of calcification. The combination of stagnant calcium-rich secretory material together with phospholipid allows the calcium to precipitate on the phospholipid to form sialomicroliths. This occurs in the small intraglandular ducts and gives rise to obstructive sialadenitis ( Fig. 6 ).
Harrison and colleagues found the salivary glands of cat to be a better experimental model than those of rat for the investigation of the origins and pathogenesis of sialadenitis and sialolithiasis, and investigations included ductal ligation, stimulation of the parasympathetic and sympathetic nerves, parasympathectomy, and sympathectomy. Sialomicroliths were detected in 1 out of 75 normal parotids, 9 out of 93 normal submandibular glands, and 17 out of 63 normal sublingual glands ( Fig. 7 ). A greatly increased occurrence of sialomicroliths in submandibular glands that had been parasympathectomized, in which sialomicroliths were found in 31 out of 41 glands, led to the pivotal realization that the lack of parasympathetic secretory stimulation had caused the pathologic accumulation of sialomicroliths ( Fig. 8 ).
The acinar secretory granules of the submandibular gland of cat contain a high level of sequestered calcium associated with acidic glycoprotein. The sequestered calcium is released in an ionized form during the normal release of glycoprotein from secretory granules or during the degradation of secretory granules in autophagosomes when there is secretory inactivity, such as that caused by parasympathectomy. The phospholipid of degraded cellular membranes becomes exposed and the ionized calcium precipitates on the phospholipid to form calcified sialomicroliths (see Fig. 7 ). The cell thereby is saved from toxic death owing to an overwhelming release of ionized calcium. Sialomicroliths may be expelled from the cells and pass into the lumina. There they may also be formed in stagnant secretory material (see Fig. 8 ). Luminal sialomicroliths may be flushed away in the saliva, although if they impact in a small intraglandular duct, a focus of obstructive atrophy may be produced. This is more likely when a pathologic accumulation of sialomicroliths occurs, such as when there is secretory inactivity. Sialomicroliths are also removed by macrophages, which helps prevent an accumulation under normal conditions.
The contrast between the accumulation of sialomicroliths in the parasympathectomized submandibular glands and the lack of accumulation in the parotids relates to the impossibility of making an adequate parasympathectomy in the parotid and to the low level of sequestered calcium in the secretory granules of this gland. Although the secretory granules of the sublingual gland contain a high level of sequestered calcium, this gland secretes spontaneously in the absence of nervous stimulation, and this spontaneous secretion is sufficient to prevent an accumulation of sialomicroliths after parasympathectomy.
Another model, established by Triantafyllou and colleagues, is the parotid of ferret, which frequently contains sialomicroliths and in which the association between secretory inactivity and sialomicrolithiasis has been confirmed. Secretory inactivity and stagnation and autophagy of calcium-rich secretory material were found to lead to the production of sialomicroliths, which were seen obstructing small intraglandular ducts.
Experimental investigations have also yielded information on obstructive atrophy and recovery. Investigations of ductal ligation have shown that: the parotid is the most susceptible to obstruction, with progressive atrophy; the submandibular gland is more resistant, with variable atrophy; the sublingual gland is the most resistant, with not only variable atrophy but also extravasation of mucus that sometimes forms an extravasation mucocele; the parenchyma of obstructed glands can adapt and survive; and obstructed glands are capable of recovery, which depends on the duration and degree of obstruction. Complete obstruction, however, does not lead to an accumulation of sialomicroliths or produce sialoliths.
Causes and natural history of chronic obstructive sialadenitis and sialolithiasis
Obstruction Primarily Caused by Inflammation
Secretory inactivity in a normal gland leads to an accumulation of sialomicroliths and ascent of the main duct by microbes. Impaction of a sialomicrolith in a small intraglandular duct causes focal obstructive atrophy ( Fig. 9 ). Microbes proliferate in atrophic parenchyma, where they are protected from the flushing and microbicidal activity of saliva and from systemic immunity by the surrounding fibrosis. The diffusion of their waste products and local invasion cause inflammation, the fluid and cellular exudate of which compresses surrounding parenchyma and causes further atrophy ( Fig. 10 ). The process eventually spreads to involve more of the lobules until the inflammatory swelling and fibrosis compress large intraglandular ducts. This causes partial obstruction that leads to ductal dilatation and stagnation of the calcium-rich secretory material. This can precipitate on the phospholipid exposed in degenerating cellular membranes to form a sialolith ( Figs 11–13 ). As the process progresses, the gland becomes increasingly inflamed, atrophic, and fibrosed ( Fig. 14 ). The process may sometimes eventually end as a symptomless sialolith in a very fibrosed duct together with completely obstructed, uninflamed, atrophic remnants of the gland.
Stenosis of the main duct is sometimes found in chronic sialadenitis and is likely secondary to chronic inflammation. The partial obstruction caused by the stenosis is an important factor in the persistence of sialadenitis and formation of sialoliths. Also, the lining of Stensen’s duct in juvenile recurrent parotitis has been seen endoscopically as white and avascular and the duct as stenotic, likely representing fibrosis of chronic inflammation, and could itself cause partial obstruction.
Plugs described as mucous, fibrinous, or fiber-like have been seen endoscopically. They have been found associated with inflammatory stenoses. Microscopic examination shows that they contain desquamated parenchymal cells and inflammatory cells. They also include the albuminous coagulum found when plasma proteins leak into the lumina of inflamed glands. The plugs are secondary to inflammation and are obstructive.
A sphincter has been seen endoscopically in Stensen’s and Wharton’s ducts, and a layer of smooth muscle has been found histologically in the wall of both ducts. Skeletal muscle fibers from the buccinator muscle are inserted into the terminal part of Stensen’s duct, other fibers from the buccinator muscle run parallel to the duct, and a valve-like structure is present in the terminal part. These muscles are likely of importance in the flow of saliva and to function as sphincters. Malfunction could allow ascent by microbes or cause partial obstruction, and microbes have been found histologically in normal Stensen’s ducts.
Foreign bodies that migrate from the orifice of the main duct or penetrate the main duct have occasionally been found and cause inflammation leading to partial obstruction and the consequent formation of a sialolith. Malfunction of ductal muscle is a likely factor in the introduction of foreign bodies via the orifice.
Many cases of chronic sialadenitis are of normal appearance on diagnostic imaging or endoscopy. These cases correspond to the 12% to 21% of cases of a normal histologic appearance.
Irrigation, even by saline alone, is effective in many cases of chronic submandibular and parotid sialadenitis because the irrigation: dilutes and flushes microbes out of atrophic foci into regions where the microbicidal capacity of the saliva is effective; flushes out obstructing plugs; dilates ducts, thus allowing small sialoliths to be passed; and dislodges sialoliths adherent to the walls of ducts.
The concentration of sequestered calcium in the secretory granules is lower in the parotid than in the submandibular gland because the secretory glycoprotein is not acidic, which accounts for the lower incidence of sialomicroliths and sialoliths in the parotid. The parotid, however, is much less resistant to noxious stimuli than the submandibular gland, and diminished or absent secretory activity facilitates ascending infection.
Sialoliths are found in the minor salivary glands, although rarely and somewhat later than in the submandibular and parotid glands. This relates to the ongoing spontaneous secretion of the minor glands, and also the sublingual gland, that occurs in the absence of nervous stimulation. This is in contrast to the submandibular and parotid glands, in which secretion is dependent on nervous stimulation, in the absence of which there is secretory inactivity. Thus, the spontaneous secretion safeguards the minor and sublingual salivary glands, which are histologically and functionally identical, until eventually age-related degenerative changes occur, including acinar atrophy, reduced discharge of secretory granules, organic sialomicroliths in ductal lumina, and an increase of inflammatory cells. This degeneration may lead to stagnation of the calcium-rich secretory material, and ultimately to the formation of sialoliths.
Obstruction not Primarily Caused by Inflammation
Noninflammatory factors of obstructive etiologic importance are kinks of Wharton’s duct, ductal polyps, ductal invaginations, and pelvis-like abnormalities of the duct at the submandibular hilum.
An obstructive etiologic factor in juvenile recurrent parotitis arising from occlusal disharmony that leads to increased tone of the masseter muscle, which then obstructs Stensen’s duct, was cured by orthodontic therapy. Support for the notion that ductal obstruction can be caused by adjacent skeletal muscle is given by the findings that fibers of the buccinator muscle are inserted into the anterior part of Stensen’s duct, fibers also run parallel to the duct, and atrophy in von Ebner’s glands is related to adjacent skeletal muscle.
Autoimmunity
The term Küttner’s tumor only became familiar to the English-speaking medical profession after 1991, when it was included in the second edition of the WHO Histologic Typing of Salivary Gland Tumours . This led to a crop of case reports purporting to be of a rare, exotic condition, namely Küttner’s tumor, without the realization that it was the mundane chronic obstructive sialadenitis without sialolithiasis. Then, recently, a rare, IgG4-related, autoimmune form of chronic sialadenitis, was named Küttner’s tumor, or Mikulicz’s disease when the lacrimal glands are involved. Unfortunately, the use of these eponyms is fraught with confusion. Küttner had no doubt that the disease that he described was chronic obstructive sialadenitis that may or may not lead to sialolithiasis. Furthermore, examination of the original detailed illustration of the histology in Mikulicz’s 1892 article revealed that the disease he described is MALT lymphoma. This confusion indicates that these eponyms should no longer be used.
Nevertheless, it is established that the rare, autoimmune disease, best named IgG4-related sclerosing disease , can involve many organs, especially the pancreas, bile duct, retroperitoneum, and salivary glands. Although the term IgG4-related sialadenitis is cumbersome compared with Küttner’s tumor , it is accurate and should not lead to confusion and possible mismanagement.
Relation to Other Diseases
Patients with sialolithiasis suffer from nephrolithiasis more often than the general population. Patients with hyperparathyroidism exhibit an increased incidence of sialolithiasis, and those with hyperparathyroidism and sialolithiasis exhibit a greater incidence of nephrolithiasis than those without sialolithiasis. This indicates that hypercalcemia can be a factor in the development of sialolithiasis, as demonstrated in animal experiments.
Survival and Recovery of Glands
Salivary glands affected by chronic obstructive sialadenitis adapt to the altered environment by cell death through apoptosis and necrosis and by autophagy in which redundant secretory granules and organelles associated with synthesis of secretory material are digested by lysosomal enzymes. Ihrler and colleagues found immunohistochemical evidence of a profound inducible capacity for regeneration in the salivary glands, which is the biologic basis for the good functional results from conservative therapies that improve the internal environment of the glands.
The function of submandibular glands affected by sialolithiasis has been investigated scintigraphically after surgical removal of the sialoliths, and improvement was inversely related to age. This is similar to the finding that a favorable clinical outcome after extracorporeal lithotripsy of the submandibular gland and parotid is inversely related to age. There is a decreased resistance and potential for recovery of the glands with increasing age, which is also shown by the greater proportion of cases of chronic submandibular sialadenitis with widespread atrophy in older patients. Additionally, a good functional recovery or favorable clinical outcome that was inversely related to the size of the sialolith probably relates to a more complete obstruction by a larger sialolith that results in greater atrophy of the gland. Support for this explanation is given by scintigraphy in which the preoperative glandular function was poor in the two cases that failed to recover.
Structure and Composition of Sialomicroliths and Sialoliths
Investigations on the structure and composition of sialomicroliths and sialoliths have yielded information about the causes and natural history of chronic obstructive sialadenitis and sialolithiasis; they are reviewed elsewhere and summarized as follows:
A sialomicrolith is defined as a concretion in a salivary gland that can only be seen microscopically and is most often calcified.
Sialomicroliths range from consisting of hydroxyapatite crystals to condensed degenerate secretory material (see Figs. 4, 5, 7 and 8 ). Sialomicroliths grow and fuse by accretion, and the presence or absence of crystals relates to the local concentration of calcium when a particular part is forming.
A sialolith is defined as a concretion in a salivary gland or main duct that can be seen with the naked eye and is most often calcified.
Sialoliths share many features with sialomicroliths, including great variation in structure and in the content of mineral and organic matrix. The organic matrix contains glycoprotein and lipids derived from secretory material and cellular membranes, which indicates a mechanism of formation similar to that of sialomicroliths.
Sialoliths contain cores, which are single or multiple, vary from purely organic to heavily calcified, and have surrounding calcified lamellae that alternate with less calcified or purely organic lamellae. The core is considered the initial sialolith, yet it sometimes exhibits a substructure, which indicates a formation by accretion and fusion of smaller sialoliths, and this process possibly also includes sialomicroliths.
The absence of bacteria in sialomicroliths shows that bacteria are unnecessary for calcification, and they are absent from the cores of sialoliths. The partial obstruction caused by a sialolith enables bacteria to ascend more easily from the mouth to reach and colonize the surface of the sialolith and to become incorporated.
Hydroxyapatite is the most widespread mineral in sialoliths, but other minerals are variably present, such as whitlockite and octacalcium phosphate. The microenvironment and its variations with time determine the type of mineral formed and the degree of calcification. Thus, the saliva of patients with calcified sialoliths was found to contain more calcium and less phytate than that of a healthy group and of patients with purely organic sialoliths. Phytate is obtained in the diet from the seeds of plants and particularly cereals and is a potent inhibitor of hydroxyapatite crystallization.
Causes and natural history of chronic obstructive sialadenitis and sialolithiasis are outlined in Fig. 15 .
