- Email: [email protected]
Salivary and lacrimal secretion is reduced in patients with ESRD
Salivary and Lacrimal Secretion Is Reduced in Patients With ESRD Maurizio Postorino, MD, Carlo Catalano, MD, Carmela Martorano, MD, Sebastiano Cutrupi, Dr Biol, Carmen Marino, Technician, Pietro Cozzupoli, MD, Paolo Scudo, MD, and Carmine Zoccali, Prof ● Background: A reduction in salivary and lacrimal secretion has been described in several diseases. However, such alterations have not been investigated fully in patients with chronic renal failure. The aim of the present study is to estimate the frequency of alterations in salivary and lacrimal secretion in long-term hemodialysis patients. Methods: Sixty-three hemodialysis patients and 23 healthy control subjects were studied. In all of them, we tested salivary secretion (Saxon’s test), lacrimal secretion (Shirmer’s test), and the presence of xerostomia and xerophthalmia symptoms. In a subgroup of patients, we performed other tests to evaluate evidence of ocular lesions and tissue damage to salivary glands. We also tested the relationship between salivary and lacrimal secretion and autonomic nervous system function. Results: On average, salivary and lacrimal secretion were markedly reduced in uremic patients compared with healthy controls, and alterations in salivary gland function were related strongly to salivary gland fibrosis and atrophy. Xerophthalmia often was asymptomatic, but frequently was associated with corneal lesions. Xerostomia and xerophthalmia were unrelated to autonomic dysfunction and hepatitis C virus infection. Conclusion: A reduction in lacrimal and salivary secretion is frequent in long-term dialysis patients. Such alterations often are asymptomatic and could be the expression of acceleration of an age-dependent decline in glandular function and attendant fibrosis and atrophy. Am J Kidney Dis 42:722-728. © 2003 by the National Kidney Foundation, Inc. INDEX WORDS: Xerostomia; xerophthalmia; dialysis; salivary secretion; lacrimal secretion.
A
REDUCTION IN salivary and lacrimal secretion has been described in association with a number of systemic diseases.1-3 In these diseases, as in Sjo¨gren’s disease, the involvement of lacrimal and salivary glands is attributed to autoimmune phenomena.4 In patients with chronic renal failure, dryness of the oral cavity is not considered a common symptom, even if the first report of xerostomia in patients with end-stage renal disease (ESRD) dates back to the late 1960s.5 There are sparse observations on xerostomia in patients with chronic renal failure.6-15 Conversely, to our knowledge, xerophthalmia in patients with ESRD has not been investigated, and in a MEDLINE search, we failed to find studies dealing with this disturbance in these patients. Because xerostomia and xerophthalmia may be important factors predisposing to oral and
From the Epidemiology and Physiopathology of Renal Diseases and Hypertension Section, Nephrology Unit, Ospedali Riuniti, Reggio Calabria, Italy. Received April 1, 2003; accepted in revised form June 16, 2003. Address reprint requests to Carmine Zoccali, CNR-IBIM and Divisione di Nefrologia, Azienda Ospedaliera BianchiMelacrino-Morelli, Reggio Calabria, Italy. E-mail: [email protected] © 2003 by the National Kidney Foundation, Inc. 0272-6386/03/4204-0016$30.00/0 doi:10.1053/S0272-6386(03)00908-9 722
ocular complications, in this study, we investigate the prevalence of these alterations and their anatomic pathological background in a sizeable group of dialysis patients. PATIENTS AND METHODS Patients The protocol conformed to the ethical guidelines of our institution, and informed consent was obtained from each participant. We studied 63 patients (38 men, 25 women; mean age, 50.2 ⫾ 13.8 [SD] years; range, 13 to 79 years) treated with long-term hemodialysis for 6.5 ⫾ 5.2 years (range, 0.1 to 17 years). Main biochemical results for these patients are listed in Table 1. These patients represent 73% of the dialysis population treated in our hemodialysis unit. We excluded from the study patients hospitalized for intercurrent illnesses, those with dementia or terminal diseases, and those who could not be investigated for logistic reasons or because of unwillingness to take part in the study. We also excluded all patients that were on treatment with drugs known to interfere with lacrimal and salivary secretion, those being treated with ophthalmic medications, and 2 patients affected by systemic diseases affecting salivary and lacrimal secretion (primary amyloidosis and diabetes mellitus, respectively). Approximately 35% of patients were current smokers. Antihypertensive medications (-blockers and calcium antagonists) were being administered to approximately 40% of patients. Causes of ESRD were chronic glomerulonephritis in 18 patients, chronic pyelonephritis or chronic tubulointerstitial nephritis in 15 patients, adult polycystic kidney disease in 5 patients, Alport’s disease in 1 patient, oxalosis in 1 patient, medullary sponge kidney in 1 patient, nephroangiosclerosis in 2 patients, and unknown in 20 patients. No patient was human immunodeficiency virus positive. Hemo-
American Journal of Kidney Diseases, Vol 42, No 4 (October), 2003: pp 722-728
XEROSTOMIA AND XEROPHTALMIA IN ESRR Table 1.
Main Biochemical Data for Study Patients
Blood urea nitrogen (mg/dL) Serum creatinine (mg/dL) Serum calcium (mEq/L) Serum potassium (mEq/L) Serum phosphorous (mg/dL) Serum parathormone (pg/mL)
129.1 ⫾ 21.5 12.0 ⫾ 1.0 4.9 ⫾ 0.7 4.7 ⫾ 0.6 5.4 ⫾ 1.3 174.0 ⫾ 104.5
NOTE. Values expressed as mean ⫾ SD. To convert blood urea nitrogen from mg/dL to mmol/L, multiply by 0.357; serum creatinine from mg/dL to mol/L, multiply by 88.4; serum calcium from mEq/L to mmol/L, multiply by 0.5; serum phosphorous from mg/dL to mmol/L, multiply by 0.323; serum parathormone from pg/mL to pmol/L, multiply by 0.105; for serum potassium, mEq/L is equivalent to mmol/L.
723
ters of wet strip (average results of both eyes) in 5 minutes. In 14 patients, we repeated Shirmer’s test after a local anaesthetic was instilled into the lacrimal bag (Jones’ test2) to exclude that the test result could be related to mechanical stress induced by test strips. Lacrimal film breakup time and studies of corneal lesions. In a group of 19 hemodialysis patients selected on the basis of Shirmer’s test (11 patients, normal results; 8 patients, pathological results), we evaluated lacrimal film breakup time. This test measures the time between eyelid opening and breaking of the precorneal lacrimal film. Breakup time becomes shorter as lacrimal secretion lessens. A breakup time less than 15 minutes is considered pathological.2 Moreover, in these patients, we searched for corneal lesions using a slit-lamp microscope with ultraviolet light after treating the ocular surface with 1% fluorescein.2
Symptoms dialysis patients were studied before the hemodialysis session to ensure adequate hydration. We also studied 23 control healthy subjects similar to study patients for age and sex (age, 46.0 ⫾ 13.2 years; range, 19 to 79 years; 13 men, 10 women).
Methods
Salivary Function Tests and Histological Appearance of Minor Salivary Glands Saxon’s test. Saxon’s test was used to study salivary secretion.16 This simple noninvasive test measures saliva production in a standardized manner (by chewing a gauze for 2 minutes). In 3 patients, the test could not be completed because of nausea. Reproducibility of Saxon’s test in uremic patients was tested by repeating it (2 to 4 weeks apart) in 16 patients. The effect of hemodialysis on Saxon’s test was evaluated by repeating the test at the end of the dialysis procedure in 6 patients. Histological evaluation of minor salivary glands. Histological evaluation of salivary glands was performed in a subgroup of 17 patients specifically selected on the basis of salivary secretion to cover a wide salivary secretion range from normal to severely impaired (salivary production from 1.05 to 4.17 g/2 min; normal reference range, 2.75 g/2 min).16 Formalin-fixed and paraffin-enclosed biopsy specimens were cut in 2-m thick slices and colored by hematoxylin-eosin using Schiff’s procedure to evaluate fibrosis, atrophy, white blood cell infiltrates, and fat content. To evaluate amyloid content, we used 6-m thick slices stained by Congo red and observed by polarized light. We also looked for 2-microglobulin deposition by using the immunoperoxidase technique (direct method, Dako antiserum; Dako Ltd, Cambridgeshire, UK). Biopsy samples were evaluated by one of the authors (C.M.), who was not aware of Saxon’s test results. Evaluation was performed by using a semiquantitative method (score from ⫹ to ⫹⫹⫹⫹).
Lacrimal Function Tests Shirmer’s test. Lacrimal secretion was measured by means of Shirmer’s test.2 Results are expressed as millime-
All patients were interviewed regarding symptoms related to xerostomia and xerophthalmia by means of an ad hoc questionnaire. We looked for the following ocular symptoms: sensation of a foreign body in one or both eyes, burning sensation, difficulty opening eyes in the morning, reduced lacrimation after emotive stimuli, eye redness, sight difficulties, itching, pain, and photosensitivity. We considered the following to be symptoms of oral cavity dryness: lack of salivation (reported by the patient or observed at clinical examination), a tongue that sticks to the palate and makes speech difficult, difficulty swallowing, dry mouth sensation, cracked lips, cracked tongue, oral cavity ulcers, and dysgeusia.17 All symptoms were categorized as present or absent. We decided against the use of analogical scales to avoid bias related to different subjective perceptions of symptoms.
Autonomic Function Tests Autonomic function was studied in 27 hemodialysis patients (mean age, 47.7 ⫾ 13.5 years; dialysis age, 7.0 ⫾ 4.8 years; 20 men, 7 women) by means of simple tests of sympathetic and parasympathetic function.18 Parasympathetic function was studied by means of Valsalva’s maneuver, the deep-breathing test, and the 30:15 ratio. Parasympathetic function was considered impaired if 2 of the 3 test results were abnormal.18 Sympathetic function was studied by measuring the change in blood pressure on standing.
Statistics All data were stored on a computer database (DBase III Plus; Ashton-Tate, Borland, Scotts Valley, CA) and analyzed using SPSS software (SPSS Inc, Chicago, IL) by means of t-test and linear correlation for normally distributed data. Nonparametric data were analyzed using chi-square or Fisher’s exact test, as appropriate. Relationships between variables were tested by means of Spearman’s rank test. Agreement between categorical tests was analyzed using the K statistic, and test reproducibility was studied by calculating the variation coefficient (VC) using the standard formula: VC ⫽ SD/mean.
724
POSTORINO ET AL
Fig 1. Results of (A) Saxon’s and (B) Shirmer’s tests in dialysis patients and healthy subjects. The line shows minimum control values.
RESULTS Salivary Secretion
Saxon’s Test and Xerostomia Symptoms On average, salivary secretion was markedly reduced in uremic patients in comparison to healthy controls (3.30 ⫾ 1.32 versus 4.09 ⫾ 1.31 g/2 min; P ⬍ 0.02). Nineteen uremic patients (32%) presented with values less than the minimum observed in healthy controls (2.6 g/2 min; Fig 1A). In patients with reduced salivary secretion in whom the test was repeated after 4 months, average results remained unchanged and the test reproducibility VC was satisfactory (VC, 18%). In hemodialysis patients, there was a slight inverse correlation between age and Saxon’s test results (r ⫽ ⫺0.25; P ⫽ 0.05), but not between Saxon’s test results and dialytic age. Symptoms of xerostomia were present in 68% of hemodialysis patients with reduced salivary secretion and 59% of those with normal salivary secretion (chi-square, 11.2; P ⫽ 0.08). None of the healthy controls reported symptoms suggestive of xerostomia. We found no difference in interdialytic weight gain between patients with normal and altered Saxon’s test results (4.5 ⫾ 1.22 and 4.6 ⫾ 1.3; P ⫽ not significant). Similarly, xerostomia symptoms did not correlate with interdialytic weight gain. Test results were similar before and after each dialysis session (2.70 ⫾ 0.71 g/2 min before dialysis, 2.77 ⫾ 0.95 g/2 min after dialysis).
Histological Evaluation of Minor Salivary Glands Patient age correlated with atrophy (Fig 2; Rank-Spearman correlation coefficient [rs] ⫽ 0.45; P ⫽ 0.05), but not fibrosis degree (rs ⫽ 0.34; P ⫽ 0.18) or fat content (rs ⫽ 0.38; P ⫽ 0.14). Atrophy was considered relevant by the pathologist in 41% of cases. Dialytic age did not correlate with any histological parameter. Saxon’s test results correlated inversely with atrophy degree (Spearman’s rs ⫽ ⫺0.68; P ⫽ 0.003; Fig 3A), fibrosis degree (rs ⫽ ⫺0.51; P ⫽
Fig 2. Histological examination of a minor salivary gland showing marked atrophy of glandular tissue. (Hematoxylin and eosin staining; original magnification ⴛ200).
XEROSTOMIA AND XEROPHTALMIA IN ESRR
725
Fig 3. Correlation between (A) atrophy, (B) fibrosis, and (C) fat deposits in minor salivary glands and results of Saxon’s test.
0.034; Fig 3B), and fat content (rs ⫽ ⫺0.45; P ⫽ 0.068; Fig 3C). White blood cell infiltrates were not present in any biopsy specimen. To the same extent, the search for amyloid and 2-microglobulin was negative in all biopsy specimens. Lacrimal Secretion
Shirmer’s Test and Xerophthalmia Symptoms On average, lacrimal secretion in hemodialysis patients was significantly less than that in healthy subjects (29.2 ⫾ 6.5 versus 20.8 ⫾ 12.5 mm/5 min; P ⬍ 0.001). Considering 17 mm/5 min as the lower limit of the normal range, 23 of 61 uremic patients (38%) had a subnormal Shirmer’s test result (Fig 1B). In dialysis patients, there was an inverse relationship between Shirmer’s test result and age (r ⫽ ⫺0.39; P ⫽ 0.002), but no relationship between this test and duration of dialysis treatment (r ⫽ ⫺0.21; P ⫽ 0.11). Shirmer’s test had good reproducibility within subjects because the VC between the right and left eye was 8%. Moreover, in patients with impaired lacrimal secretion, the VC was satisfac-
tory (15%). Local instillation of an anesthetic produced a well-defined (P ⬍ 0.02) reduction in lacrimal secretion in comparison to the standard Shirmer’s test. However, the relationship between lacrimal secretion tested with and without anesthetic instillation was very high in both quantitative terms (r ⫽ 0.92; P ⬍ 0.001) and terms of patient classification (K ⫽ 0.84; P ⫽ 0.002). Thirty-four hemodialysis patients (56%) reported 1 or more symptoms related to impaired lacrimal secretion. However, 7 of 17 patients (41%) with reduced lacrimal secretion were asymptomatic, and xerophthalmia-related symptoms were reported even by a number of patients with lacrimal function test results in the normal range. No subject in the control group reported xerophthalmia symptoms. Lacrimal Film Breakup Time and Corneal Lesions Breakup time was normal (28.9 ⫾ 10.4 minutes; range, 15 to 53 minutes) in all patients with Shirmer’s test results in the normal range. In 8
726
POSTORINO ET AL Table 2.
Correlations Between Results of Shirmer’s and Saxon’s Tests and Autonomic Function Tests Parasympathetic Test
Shirmer’s test Normal Pathological Saxon’s test Normal Pathological
Sympathetic Test
Normal
Pathological
Normal
Pathological
13 2 chi-square ⫽ 4.3; P ⫽ 0.09*
6 6 chi-square ⫽ 0.5; P ⫽ 0.9*
16 8
1 0
10 5 chi-square ⫽ 0.1; P ⫽ 0.9*
8 4 chi-square ⫽ 0.6; P ⫽ 0.9*
15 9
1 0
*After Fisher’s correction.
patients with impaired Shirmer’s test results, 5 patients had a markedly reduced breakup time (7.5 ⫾ 3.8 minutes; range, 2.5 to 12 minutes), whereas in 3 patients, breakup time was normal (17, 20, and 23 minutes). Three of 23 patients with subnormal lacrimal secretion (ie, 13%) had evident corneal lesions. Relationship Between Salivary and Lacrimal Tests, Hepatitis C Virus, and Autonomic Test Results
Shirmer’s test results did not correlate with Saxon’s test results analyzed using a parametric (r ⫽ 0.2; P ⫽ 0.8) or categorical (chi-square, 0.13; P ⫽ 0.9) approach. Thirty-four percent of patients were antibody positive to hepatitis C virus (HCV) positive. However, there was no relationship between Saxon’s and Shirmer’s test results and antibody to HCV positivity (chi-square, 0.9 and 0.7, respectively; P ⬎ 0.4). Likewise, no significant relationship was found between results of salivary and lacrimal function and autonomic function tests (Table 2). DISCUSSION
The main findings in this study are that salivary and lacrimal function are subnormal in approximately one third of dialysis patients, and impaired salivary function is associated with histological evidence of salivary gland atrophy and fibrosis. A number of symptoms related to functional impairment of exocrine glands have been described in hemodialysis patients. Alterations of the gastric mucosa with reduced acid secretion associated with amyloid deposition,19 interstitial
and periductal pancreatic fibrosis associated with severely impaired peptic secretion,20,21 and microangiopathic cutaneous xerosis22 have been well documented in these patients. Alterations in parotid salivary secretion were noted in a small series of uremic patients,10 as well as in a study by Kao et al,13 who performed quantitative isotopic studies of salivary gland secretion. Furthermore, accumulation of a fibrillary component in salivary glands23 has been described in another study of patients on conservative treatment. To our knowledge, ours is the first study to systematically investigate both lacrimal and salivary secretion in a reasonably large sample of dialysis patients. Of note, in addition to confirming that salivary function is compromised in patients with ESRD, we found that exocrine gland functional alterations also are extended to lacrimal glands. Lack of saliva increases the risk for oral infections and dental caries.1 Oral and dental infections are common in patients with ESRD, and in addition to being responsible for local damage, they may have systemic effects because they contribute to increased serum Creactive protein levels in these patients.24 An important finding in our study is that hemodialysis patients might not report symptoms related to xerostomia, even if salivary production is impaired. This lack of symptoms probably is related to that even modest saliva production is sufficient to overcome dryness of the oral cavity. Conversely, it should be recognized that changes in salivary composition might explain, at least in part, the lack of symptoms in patients with functional impairment. Furthermore, in future studies, measurement of unstimu-
XEROSTOMIA AND XEROPHTALMIA IN ESRR
lated salivary secretion may be useful to better characterize the disturbance in salivary function observed in this study. Dry mouth may be a stimulus to drinking and therefore a potential trigger of fluid accumulation. However, we found that patients with impaired salivary function and those with significant xerostomia did not show increased interdialytic weight gain compared with patients with normal salivary function or no symptoms. Because our observations are based on prevalent patients, we cannot exclude that the lack of high interdialytic weight gains in patients with xerostomia was related to repeated recommendations by staff in patients who, if left untrained, would have presented with fluid overload. Smoking may influence salivary secretion. In this study, we failed to find major differences between smokers and nonsmokers (data not shown), but the number of patients was too small to allow statistical adjustments for factors that may confound this association. Our working hypothesis was that salivary disturbances in uremic patients could be associated with 2-microglobulin deposition. 2-Microglobulin deposition has been described in various organs in uremic patients25 and those on conservative treatment. Amyloid-like fibrils were observed in labial salivary glands.23 Contrary to our expectation, in our patients, amyloid deposits were absent with both specific staining and histochemical analysis. Because we did not perform ultrastructural studies, we cannot exclude the presence of a fibrillary component. However, this component, if any, must be different from 2-microglobulin. In our study, impaired salivary function was associated with glandular atrophy and fibrosis of minor salivary glands and a conspicuous lack of inflammation. The lack of inflammation is in keeping with the parallel observation that hepatitis C, a recognized cause of salivary gland inflammation,26 was completely unrelated to salivary function in dialysis patients. The autonomic system has a major role in the regulation of salivary1 and lacrimal gland27 function, but again, we found no relationship between results of tests of cardiovascular autonomic control and salivary and lacrimal tests. This finding implies that altered reflex control of the cardiovascular system is dissociated from autonomic control of salivary and lacrimal glands in uremic patients, but does
727
not exclude that local (ie, limited to salivary or lacrimal glands) alterations in the autonomic system have a role in xerostomia and xerophthalmia in these patients. More powerful studies using sophisticated autonomic tests are required to clarify the relationship between autonomic neuropathy and alterations in salivary and lacrimal gland function in dialysis patients. The most important and novel observation in our study is that in uremic patients, exocrine impairment also is present in lacrimal glands. Lacrimal secretion was markedly reduced, and 28% of dialysis patients presented with lacrimal function values less than the lower limit of normal range. However, as previously noted for xerostomia, xerophthalmia also was often asymptomatic. As discussed, residual secretion might have been sufficient to prevent symptoms. Conversely, it remains to be seen whether differences in lacrimal composition might account for differences in symptom perception. Notably, approximately one third of patients with impaired Shirmer’s test results presented with corneal alterations, showing the clinical relevance of xerophthalmia. Contrary to our expectation, impairments in lacrimal and salivary function were not associated. This observation might suggest that causes of the 2 phenomena are different or that local factors (in salivary or lacrimal glands) interfere with clinical expression of the illness. Conversely, alterations in salivary and lacrimal function showed the expected age-dependent decline, a phenomenon attributable to a complex interplay of risk factors, including comorbidities, use of drugs, and other factors. Thus, our findings suggest that uremia accelerates the age-dependent decline in salivary and lacrimal function, a concept in keeping with observations made in other organ systems, such as the cardiovascular system. In conclusion, reduction in salivary secretion is frequent in hemodialysis patients. It is associated with atrophy and fibrosis, but often is without symptoms. Lacrimal function impairment also is frequent in hemodialysis patients. However, the 2 dysfunctions are dissociated. Additional studies are needed to assess the link between qualitative salivary and lacrimal alterations and xerostomia and xerophthalmia in patients with renal diseases.
728
POSTORINO ET AL
ACKNOWLEDGMENT The authors thank Pauline Jackson for reviewing the manuscript and Giovanni Tripepi (Epidemiology and Physiopathology of Renal Diseases and Hypertension Section statistical technician) for help given in preparation of the manuscript.
REFERENCES 1. Vissink A, Panders AK, Johannes’s-Gravenmade E, Vemey A: The causes and consequences of hyposalivation. Ear Nose Throat J 67:166-176, 1988 2. Murube-Del-Castillo J, Cortes-Rodrigo MD: Eye parameters for the diagnosis of xerophthalmos. Clin Exp Rheumatol 7:145-150, 1989 3. Frith P: The eye in general medicine, in Oxford Textbook of Medicine (book on CD-ROM). Oxford, UK, Oxford University Press and Electronic Publishing, 1996, chapt 26 4. Venables PJW: Sjo¨ gren syndrome, in Oxford Textbook of Medicine (book on CD-ROM). Oxford, UK, Oxford University Press and Electronic Publishing, 1996, chapt 18, section 11.5 5. Dahlberg WH, Sreebny LM, King B: Studies of parotid saliva and blood in hemodialysis patients. J Appl Physiol 23:100-108, 1967 6. Epstein SR, Mandel I, Scopp IW: Salivary composition and calculus formation in patients undergoing hemodialysis. J Periodontol 51:336-338, 1980 7. Obry F, Belcourt A, Frank RM, Geisert J, Sommermater J, Fischbach M: [Low caries activity and salivary pH in youngsters dialyzed for chronic renal failure]. J Biol Buccale 12:181-186, 1984 8. Kho HS, Lee SW, Chung SC, Kim YK: Oral manifestations and salivary flow rate, pH, and buffer capacity in patients with end-stage renal disease undergoing hemodialysis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 88:316-319, 1999 9. Gavalda C, Bagan J, Scully C, Silvestre F, Milian M, Jimenez Y: Renal hemodialysis patients: Oral, salivary, dental and periodontal findings in 105 adult cases. Oral Dis 5:299-302, 1999 10. Van Der Lichte JL, Mulder AW, Michels LFE: Parotid gland dysfunction in hemodialysis patients. Neth J Med 26:39-43, 1983 11. Kaya M, Cermik TF, Ustun F, Sen S, Berkarda S: Salivary function in patients with chronic renal failure undergoing hemodialysis. Ann Nucl Med 16:117-120, 2002
12. Bayraktar G, Kazancioglu R, Bozfakioglu S, Ecder T, Yildiz A, Ark E: Stimulated salivary flow rate in chronic hemodialysis patients. Nephron 91:210-214, 2002 13. Kao CH, Hsieh JF, Tsai SC, Ho YJ, Ghang HR: Decreased salivary function in patients with end-stage renal disease requiring hemodialysis. Am J Kidney Dis 36:110114, 2000 14. Postorino M., Zoccali C, Martorano C, Alati G, Maggiore Q: “Sicca like” syndrome in hemodialysis patients. Kidney Int 37:315A, 1990 (abstr) 15. Zoccali C: Dry mouth in dialysis patients. Semin Dial 7:304, 1994 16. Kohler PF, Winter ME: A quantitative test for xerostomia. Arthritis Rheum 28:1128-1132, 1985 17. Whaley K, Williamson J, Chisholm DM, Webb J, Mason DK, Buchanan WW: Sjo¨ gren’s syndrome. Q J Med XLII:279-304, 1973 18. Zoccali C, Ciccarelli M, Maggiore Q: Defective reflex control of heart in dialysis patients: Evidence for an afferent autonomic lesion. Clin Sci 63:285-292, 1982 19. Takahashi S, Morita T, Koda Y, Murayama H, Hirasawa Y: Gastrointestinal involvement of dialysis-related amyloidosis. Clin Nephol 30:168-171, 1988 20. Sacs EF, Bloch HM, Milne FJ: Pancreatic supplementation in end-stage renal disease. Nephron 37:120-122, 1984 21. Avram MM: High prevalence of pancreatic disease in chronic renal failure. Nephron 18:68-71, 1977 22. Gilchrest BA, Rowe JW, Mihm MC: Clinical and histological skin changes in chronic renal failure: Evidence for a dialysis-resistant, transplant-responsive microangiopathy. Lancet 2:1271-1275, 1980 23. Kinashi M, Ohbayashi S, Ohbayashi M: Detection of amyloid-like fibrils using labial salivary gland biopsy in patients with nondialysis renal insufficiency. Am J Nephrol 9:435-437, 1989 24. Rahmati MA, Craig RG, Homel P, Kaysen GA, Levin NW: Serum markers of periodontal disease status and inflammation in hemodialysis patients. Am J Kidney Dis 40:983989, 2002 25. Danesh F, Ho LT: Dialysis-related amyloidosis: History and clinical manifestation. Semin Dial 14:80-85, 2001 26. McMurray RW: Hepatitis C-associated autoimmune disorders. Rheum Dis Clin North Am 24:353-374, 1998 27. Guyton AC: The autonomic nervous system, in Guyton AC (ed): Textbook of Medical Physiology (ed 6). Philadelphia, PA, Saunders, 1981, pp 710-722
A
REDUCTION IN salivary and lacrimal secretion has been described in association with a number of systemic diseases.1-3 In these diseases, as in Sjo¨gren’s disease, the involvement of lacrimal and salivary glands is attributed to autoimmune phenomena.4 In patients with chronic renal failure, dryness of the oral cavity is not considered a common symptom, even if the first report of xerostomia in patients with end-stage renal disease (ESRD) dates back to the late 1960s.5 There are sparse observations on xerostomia in patients with chronic renal failure.6-15 Conversely, to our knowledge, xerophthalmia in patients with ESRD has not been investigated, and in a MEDLINE search, we failed to find studies dealing with this disturbance in these patients. Because xerostomia and xerophthalmia may be important factors predisposing to oral and
From the Epidemiology and Physiopathology of Renal Diseases and Hypertension Section, Nephrology Unit, Ospedali Riuniti, Reggio Calabria, Italy. Received April 1, 2003; accepted in revised form June 16, 2003. Address reprint requests to Carmine Zoccali, CNR-IBIM and Divisione di Nefrologia, Azienda Ospedaliera BianchiMelacrino-Morelli, Reggio Calabria, Italy. E-mail: [email protected] © 2003 by the National Kidney Foundation, Inc. 0272-6386/03/4204-0016$30.00/0 doi:10.1053/S0272-6386(03)00908-9 722
ocular complications, in this study, we investigate the prevalence of these alterations and their anatomic pathological background in a sizeable group of dialysis patients. PATIENTS AND METHODS Patients The protocol conformed to the ethical guidelines of our institution, and informed consent was obtained from each participant. We studied 63 patients (38 men, 25 women; mean age, 50.2 ⫾ 13.8 [SD] years; range, 13 to 79 years) treated with long-term hemodialysis for 6.5 ⫾ 5.2 years (range, 0.1 to 17 years). Main biochemical results for these patients are listed in Table 1. These patients represent 73% of the dialysis population treated in our hemodialysis unit. We excluded from the study patients hospitalized for intercurrent illnesses, those with dementia or terminal diseases, and those who could not be investigated for logistic reasons or because of unwillingness to take part in the study. We also excluded all patients that were on treatment with drugs known to interfere with lacrimal and salivary secretion, those being treated with ophthalmic medications, and 2 patients affected by systemic diseases affecting salivary and lacrimal secretion (primary amyloidosis and diabetes mellitus, respectively). Approximately 35% of patients were current smokers. Antihypertensive medications (-blockers and calcium antagonists) were being administered to approximately 40% of patients. Causes of ESRD were chronic glomerulonephritis in 18 patients, chronic pyelonephritis or chronic tubulointerstitial nephritis in 15 patients, adult polycystic kidney disease in 5 patients, Alport’s disease in 1 patient, oxalosis in 1 patient, medullary sponge kidney in 1 patient, nephroangiosclerosis in 2 patients, and unknown in 20 patients. No patient was human immunodeficiency virus positive. Hemo-
American Journal of Kidney Diseases, Vol 42, No 4 (October), 2003: pp 722-728
XEROSTOMIA AND XEROPHTALMIA IN ESRR Table 1.
Main Biochemical Data for Study Patients
Blood urea nitrogen (mg/dL) Serum creatinine (mg/dL) Serum calcium (mEq/L) Serum potassium (mEq/L) Serum phosphorous (mg/dL) Serum parathormone (pg/mL)
129.1 ⫾ 21.5 12.0 ⫾ 1.0 4.9 ⫾ 0.7 4.7 ⫾ 0.6 5.4 ⫾ 1.3 174.0 ⫾ 104.5
NOTE. Values expressed as mean ⫾ SD. To convert blood urea nitrogen from mg/dL to mmol/L, multiply by 0.357; serum creatinine from mg/dL to mol/L, multiply by 88.4; serum calcium from mEq/L to mmol/L, multiply by 0.5; serum phosphorous from mg/dL to mmol/L, multiply by 0.323; serum parathormone from pg/mL to pmol/L, multiply by 0.105; for serum potassium, mEq/L is equivalent to mmol/L.
723
ters of wet strip (average results of both eyes) in 5 minutes. In 14 patients, we repeated Shirmer’s test after a local anaesthetic was instilled into the lacrimal bag (Jones’ test2) to exclude that the test result could be related to mechanical stress induced by test strips. Lacrimal film breakup time and studies of corneal lesions. In a group of 19 hemodialysis patients selected on the basis of Shirmer’s test (11 patients, normal results; 8 patients, pathological results), we evaluated lacrimal film breakup time. This test measures the time between eyelid opening and breaking of the precorneal lacrimal film. Breakup time becomes shorter as lacrimal secretion lessens. A breakup time less than 15 minutes is considered pathological.2 Moreover, in these patients, we searched for corneal lesions using a slit-lamp microscope with ultraviolet light after treating the ocular surface with 1% fluorescein.2
Symptoms dialysis patients were studied before the hemodialysis session to ensure adequate hydration. We also studied 23 control healthy subjects similar to study patients for age and sex (age, 46.0 ⫾ 13.2 years; range, 19 to 79 years; 13 men, 10 women).
Methods
Salivary Function Tests and Histological Appearance of Minor Salivary Glands Saxon’s test. Saxon’s test was used to study salivary secretion.16 This simple noninvasive test measures saliva production in a standardized manner (by chewing a gauze for 2 minutes). In 3 patients, the test could not be completed because of nausea. Reproducibility of Saxon’s test in uremic patients was tested by repeating it (2 to 4 weeks apart) in 16 patients. The effect of hemodialysis on Saxon’s test was evaluated by repeating the test at the end of the dialysis procedure in 6 patients. Histological evaluation of minor salivary glands. Histological evaluation of salivary glands was performed in a subgroup of 17 patients specifically selected on the basis of salivary secretion to cover a wide salivary secretion range from normal to severely impaired (salivary production from 1.05 to 4.17 g/2 min; normal reference range, 2.75 g/2 min).16 Formalin-fixed and paraffin-enclosed biopsy specimens were cut in 2-m thick slices and colored by hematoxylin-eosin using Schiff’s procedure to evaluate fibrosis, atrophy, white blood cell infiltrates, and fat content. To evaluate amyloid content, we used 6-m thick slices stained by Congo red and observed by polarized light. We also looked for 2-microglobulin deposition by using the immunoperoxidase technique (direct method, Dako antiserum; Dako Ltd, Cambridgeshire, UK). Biopsy samples were evaluated by one of the authors (C.M.), who was not aware of Saxon’s test results. Evaluation was performed by using a semiquantitative method (score from ⫹ to ⫹⫹⫹⫹).
Lacrimal Function Tests Shirmer’s test. Lacrimal secretion was measured by means of Shirmer’s test.2 Results are expressed as millime-
All patients were interviewed regarding symptoms related to xerostomia and xerophthalmia by means of an ad hoc questionnaire. We looked for the following ocular symptoms: sensation of a foreign body in one or both eyes, burning sensation, difficulty opening eyes in the morning, reduced lacrimation after emotive stimuli, eye redness, sight difficulties, itching, pain, and photosensitivity. We considered the following to be symptoms of oral cavity dryness: lack of salivation (reported by the patient or observed at clinical examination), a tongue that sticks to the palate and makes speech difficult, difficulty swallowing, dry mouth sensation, cracked lips, cracked tongue, oral cavity ulcers, and dysgeusia.17 All symptoms were categorized as present or absent. We decided against the use of analogical scales to avoid bias related to different subjective perceptions of symptoms.
Autonomic Function Tests Autonomic function was studied in 27 hemodialysis patients (mean age, 47.7 ⫾ 13.5 years; dialysis age, 7.0 ⫾ 4.8 years; 20 men, 7 women) by means of simple tests of sympathetic and parasympathetic function.18 Parasympathetic function was studied by means of Valsalva’s maneuver, the deep-breathing test, and the 30:15 ratio. Parasympathetic function was considered impaired if 2 of the 3 test results were abnormal.18 Sympathetic function was studied by measuring the change in blood pressure on standing.
Statistics All data were stored on a computer database (DBase III Plus; Ashton-Tate, Borland, Scotts Valley, CA) and analyzed using SPSS software (SPSS Inc, Chicago, IL) by means of t-test and linear correlation for normally distributed data. Nonparametric data were analyzed using chi-square or Fisher’s exact test, as appropriate. Relationships between variables were tested by means of Spearman’s rank test. Agreement between categorical tests was analyzed using the K statistic, and test reproducibility was studied by calculating the variation coefficient (VC) using the standard formula: VC ⫽ SD/mean.
724
POSTORINO ET AL
Fig 1. Results of (A) Saxon’s and (B) Shirmer’s tests in dialysis patients and healthy subjects. The line shows minimum control values.
RESULTS Salivary Secretion
Saxon’s Test and Xerostomia Symptoms On average, salivary secretion was markedly reduced in uremic patients in comparison to healthy controls (3.30 ⫾ 1.32 versus 4.09 ⫾ 1.31 g/2 min; P ⬍ 0.02). Nineteen uremic patients (32%) presented with values less than the minimum observed in healthy controls (2.6 g/2 min; Fig 1A). In patients with reduced salivary secretion in whom the test was repeated after 4 months, average results remained unchanged and the test reproducibility VC was satisfactory (VC, 18%). In hemodialysis patients, there was a slight inverse correlation between age and Saxon’s test results (r ⫽ ⫺0.25; P ⫽ 0.05), but not between Saxon’s test results and dialytic age. Symptoms of xerostomia were present in 68% of hemodialysis patients with reduced salivary secretion and 59% of those with normal salivary secretion (chi-square, 11.2; P ⫽ 0.08). None of the healthy controls reported symptoms suggestive of xerostomia. We found no difference in interdialytic weight gain between patients with normal and altered Saxon’s test results (4.5 ⫾ 1.22 and 4.6 ⫾ 1.3; P ⫽ not significant). Similarly, xerostomia symptoms did not correlate with interdialytic weight gain. Test results were similar before and after each dialysis session (2.70 ⫾ 0.71 g/2 min before dialysis, 2.77 ⫾ 0.95 g/2 min after dialysis).
Histological Evaluation of Minor Salivary Glands Patient age correlated with atrophy (Fig 2; Rank-Spearman correlation coefficient [rs] ⫽ 0.45; P ⫽ 0.05), but not fibrosis degree (rs ⫽ 0.34; P ⫽ 0.18) or fat content (rs ⫽ 0.38; P ⫽ 0.14). Atrophy was considered relevant by the pathologist in 41% of cases. Dialytic age did not correlate with any histological parameter. Saxon’s test results correlated inversely with atrophy degree (Spearman’s rs ⫽ ⫺0.68; P ⫽ 0.003; Fig 3A), fibrosis degree (rs ⫽ ⫺0.51; P ⫽
Fig 2. Histological examination of a minor salivary gland showing marked atrophy of glandular tissue. (Hematoxylin and eosin staining; original magnification ⴛ200).
XEROSTOMIA AND XEROPHTALMIA IN ESRR
725
Fig 3. Correlation between (A) atrophy, (B) fibrosis, and (C) fat deposits in minor salivary glands and results of Saxon’s test.
0.034; Fig 3B), and fat content (rs ⫽ ⫺0.45; P ⫽ 0.068; Fig 3C). White blood cell infiltrates were not present in any biopsy specimen. To the same extent, the search for amyloid and 2-microglobulin was negative in all biopsy specimens. Lacrimal Secretion
Shirmer’s Test and Xerophthalmia Symptoms On average, lacrimal secretion in hemodialysis patients was significantly less than that in healthy subjects (29.2 ⫾ 6.5 versus 20.8 ⫾ 12.5 mm/5 min; P ⬍ 0.001). Considering 17 mm/5 min as the lower limit of the normal range, 23 of 61 uremic patients (38%) had a subnormal Shirmer’s test result (Fig 1B). In dialysis patients, there was an inverse relationship between Shirmer’s test result and age (r ⫽ ⫺0.39; P ⫽ 0.002), but no relationship between this test and duration of dialysis treatment (r ⫽ ⫺0.21; P ⫽ 0.11). Shirmer’s test had good reproducibility within subjects because the VC between the right and left eye was 8%. Moreover, in patients with impaired lacrimal secretion, the VC was satisfac-
tory (15%). Local instillation of an anesthetic produced a well-defined (P ⬍ 0.02) reduction in lacrimal secretion in comparison to the standard Shirmer’s test. However, the relationship between lacrimal secretion tested with and without anesthetic instillation was very high in both quantitative terms (r ⫽ 0.92; P ⬍ 0.001) and terms of patient classification (K ⫽ 0.84; P ⫽ 0.002). Thirty-four hemodialysis patients (56%) reported 1 or more symptoms related to impaired lacrimal secretion. However, 7 of 17 patients (41%) with reduced lacrimal secretion were asymptomatic, and xerophthalmia-related symptoms were reported even by a number of patients with lacrimal function test results in the normal range. No subject in the control group reported xerophthalmia symptoms. Lacrimal Film Breakup Time and Corneal Lesions Breakup time was normal (28.9 ⫾ 10.4 minutes; range, 15 to 53 minutes) in all patients with Shirmer’s test results in the normal range. In 8
726
POSTORINO ET AL Table 2.
Correlations Between Results of Shirmer’s and Saxon’s Tests and Autonomic Function Tests Parasympathetic Test
Shirmer’s test Normal Pathological Saxon’s test Normal Pathological
Sympathetic Test
Normal
Pathological
Normal
Pathological
13 2 chi-square ⫽ 4.3; P ⫽ 0.09*
6 6 chi-square ⫽ 0.5; P ⫽ 0.9*
16 8
1 0
10 5 chi-square ⫽ 0.1; P ⫽ 0.9*
8 4 chi-square ⫽ 0.6; P ⫽ 0.9*
15 9
1 0
*After Fisher’s correction.
patients with impaired Shirmer’s test results, 5 patients had a markedly reduced breakup time (7.5 ⫾ 3.8 minutes; range, 2.5 to 12 minutes), whereas in 3 patients, breakup time was normal (17, 20, and 23 minutes). Three of 23 patients with subnormal lacrimal secretion (ie, 13%) had evident corneal lesions. Relationship Between Salivary and Lacrimal Tests, Hepatitis C Virus, and Autonomic Test Results
Shirmer’s test results did not correlate with Saxon’s test results analyzed using a parametric (r ⫽ 0.2; P ⫽ 0.8) or categorical (chi-square, 0.13; P ⫽ 0.9) approach. Thirty-four percent of patients were antibody positive to hepatitis C virus (HCV) positive. However, there was no relationship between Saxon’s and Shirmer’s test results and antibody to HCV positivity (chi-square, 0.9 and 0.7, respectively; P ⬎ 0.4). Likewise, no significant relationship was found between results of salivary and lacrimal function and autonomic function tests (Table 2). DISCUSSION
The main findings in this study are that salivary and lacrimal function are subnormal in approximately one third of dialysis patients, and impaired salivary function is associated with histological evidence of salivary gland atrophy and fibrosis. A number of symptoms related to functional impairment of exocrine glands have been described in hemodialysis patients. Alterations of the gastric mucosa with reduced acid secretion associated with amyloid deposition,19 interstitial
and periductal pancreatic fibrosis associated with severely impaired peptic secretion,20,21 and microangiopathic cutaneous xerosis22 have been well documented in these patients. Alterations in parotid salivary secretion were noted in a small series of uremic patients,10 as well as in a study by Kao et al,13 who performed quantitative isotopic studies of salivary gland secretion. Furthermore, accumulation of a fibrillary component in salivary glands23 has been described in another study of patients on conservative treatment. To our knowledge, ours is the first study to systematically investigate both lacrimal and salivary secretion in a reasonably large sample of dialysis patients. Of note, in addition to confirming that salivary function is compromised in patients with ESRD, we found that exocrine gland functional alterations also are extended to lacrimal glands. Lack of saliva increases the risk for oral infections and dental caries.1 Oral and dental infections are common in patients with ESRD, and in addition to being responsible for local damage, they may have systemic effects because they contribute to increased serum Creactive protein levels in these patients.24 An important finding in our study is that hemodialysis patients might not report symptoms related to xerostomia, even if salivary production is impaired. This lack of symptoms probably is related to that even modest saliva production is sufficient to overcome dryness of the oral cavity. Conversely, it should be recognized that changes in salivary composition might explain, at least in part, the lack of symptoms in patients with functional impairment. Furthermore, in future studies, measurement of unstimu-
XEROSTOMIA AND XEROPHTALMIA IN ESRR
lated salivary secretion may be useful to better characterize the disturbance in salivary function observed in this study. Dry mouth may be a stimulus to drinking and therefore a potential trigger of fluid accumulation. However, we found that patients with impaired salivary function and those with significant xerostomia did not show increased interdialytic weight gain compared with patients with normal salivary function or no symptoms. Because our observations are based on prevalent patients, we cannot exclude that the lack of high interdialytic weight gains in patients with xerostomia was related to repeated recommendations by staff in patients who, if left untrained, would have presented with fluid overload. Smoking may influence salivary secretion. In this study, we failed to find major differences between smokers and nonsmokers (data not shown), but the number of patients was too small to allow statistical adjustments for factors that may confound this association. Our working hypothesis was that salivary disturbances in uremic patients could be associated with 2-microglobulin deposition. 2-Microglobulin deposition has been described in various organs in uremic patients25 and those on conservative treatment. Amyloid-like fibrils were observed in labial salivary glands.23 Contrary to our expectation, in our patients, amyloid deposits were absent with both specific staining and histochemical analysis. Because we did not perform ultrastructural studies, we cannot exclude the presence of a fibrillary component. However, this component, if any, must be different from 2-microglobulin. In our study, impaired salivary function was associated with glandular atrophy and fibrosis of minor salivary glands and a conspicuous lack of inflammation. The lack of inflammation is in keeping with the parallel observation that hepatitis C, a recognized cause of salivary gland inflammation,26 was completely unrelated to salivary function in dialysis patients. The autonomic system has a major role in the regulation of salivary1 and lacrimal gland27 function, but again, we found no relationship between results of tests of cardiovascular autonomic control and salivary and lacrimal tests. This finding implies that altered reflex control of the cardiovascular system is dissociated from autonomic control of salivary and lacrimal glands in uremic patients, but does
727
not exclude that local (ie, limited to salivary or lacrimal glands) alterations in the autonomic system have a role in xerostomia and xerophthalmia in these patients. More powerful studies using sophisticated autonomic tests are required to clarify the relationship between autonomic neuropathy and alterations in salivary and lacrimal gland function in dialysis patients. The most important and novel observation in our study is that in uremic patients, exocrine impairment also is present in lacrimal glands. Lacrimal secretion was markedly reduced, and 28% of dialysis patients presented with lacrimal function values less than the lower limit of normal range. However, as previously noted for xerostomia, xerophthalmia also was often asymptomatic. As discussed, residual secretion might have been sufficient to prevent symptoms. Conversely, it remains to be seen whether differences in lacrimal composition might account for differences in symptom perception. Notably, approximately one third of patients with impaired Shirmer’s test results presented with corneal alterations, showing the clinical relevance of xerophthalmia. Contrary to our expectation, impairments in lacrimal and salivary function were not associated. This observation might suggest that causes of the 2 phenomena are different or that local factors (in salivary or lacrimal glands) interfere with clinical expression of the illness. Conversely, alterations in salivary and lacrimal function showed the expected age-dependent decline, a phenomenon attributable to a complex interplay of risk factors, including comorbidities, use of drugs, and other factors. Thus, our findings suggest that uremia accelerates the age-dependent decline in salivary and lacrimal function, a concept in keeping with observations made in other organ systems, such as the cardiovascular system. In conclusion, reduction in salivary secretion is frequent in hemodialysis patients. It is associated with atrophy and fibrosis, but often is without symptoms. Lacrimal function impairment also is frequent in hemodialysis patients. However, the 2 dysfunctions are dissociated. Additional studies are needed to assess the link between qualitative salivary and lacrimal alterations and xerostomia and xerophthalmia in patients with renal diseases.
728
POSTORINO ET AL
ACKNOWLEDGMENT The authors thank Pauline Jackson for reviewing the manuscript and Giovanni Tripepi (Epidemiology and Physiopathology of Renal Diseases and Hypertension Section statistical technician) for help given in preparation of the manuscript.
REFERENCES 1. Vissink A, Panders AK, Johannes’s-Gravenmade E, Vemey A: The causes and consequences of hyposalivation. Ear Nose Throat J 67:166-176, 1988 2. Murube-Del-Castillo J, Cortes-Rodrigo MD: Eye parameters for the diagnosis of xerophthalmos. Clin Exp Rheumatol 7:145-150, 1989 3. Frith P: The eye in general medicine, in Oxford Textbook of Medicine (book on CD-ROM). Oxford, UK, Oxford University Press and Electronic Publishing, 1996, chapt 26 4. Venables PJW: Sjo¨ gren syndrome, in Oxford Textbook of Medicine (book on CD-ROM). Oxford, UK, Oxford University Press and Electronic Publishing, 1996, chapt 18, section 11.5 5. Dahlberg WH, Sreebny LM, King B: Studies of parotid saliva and blood in hemodialysis patients. J Appl Physiol 23:100-108, 1967 6. Epstein SR, Mandel I, Scopp IW: Salivary composition and calculus formation in patients undergoing hemodialysis. J Periodontol 51:336-338, 1980 7. Obry F, Belcourt A, Frank RM, Geisert J, Sommermater J, Fischbach M: [Low caries activity and salivary pH in youngsters dialyzed for chronic renal failure]. J Biol Buccale 12:181-186, 1984 8. Kho HS, Lee SW, Chung SC, Kim YK: Oral manifestations and salivary flow rate, pH, and buffer capacity in patients with end-stage renal disease undergoing hemodialysis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 88:316-319, 1999 9. Gavalda C, Bagan J, Scully C, Silvestre F, Milian M, Jimenez Y: Renal hemodialysis patients: Oral, salivary, dental and periodontal findings in 105 adult cases. Oral Dis 5:299-302, 1999 10. Van Der Lichte JL, Mulder AW, Michels LFE: Parotid gland dysfunction in hemodialysis patients. Neth J Med 26:39-43, 1983 11. Kaya M, Cermik TF, Ustun F, Sen S, Berkarda S: Salivary function in patients with chronic renal failure undergoing hemodialysis. Ann Nucl Med 16:117-120, 2002
12. Bayraktar G, Kazancioglu R, Bozfakioglu S, Ecder T, Yildiz A, Ark E: Stimulated salivary flow rate in chronic hemodialysis patients. Nephron 91:210-214, 2002 13. Kao CH, Hsieh JF, Tsai SC, Ho YJ, Ghang HR: Decreased salivary function in patients with end-stage renal disease requiring hemodialysis. Am J Kidney Dis 36:110114, 2000 14. Postorino M., Zoccali C, Martorano C, Alati G, Maggiore Q: “Sicca like” syndrome in hemodialysis patients. Kidney Int 37:315A, 1990 (abstr) 15. Zoccali C: Dry mouth in dialysis patients. Semin Dial 7:304, 1994 16. Kohler PF, Winter ME: A quantitative test for xerostomia. Arthritis Rheum 28:1128-1132, 1985 17. Whaley K, Williamson J, Chisholm DM, Webb J, Mason DK, Buchanan WW: Sjo¨ gren’s syndrome. Q J Med XLII:279-304, 1973 18. Zoccali C, Ciccarelli M, Maggiore Q: Defective reflex control of heart in dialysis patients: Evidence for an afferent autonomic lesion. Clin Sci 63:285-292, 1982 19. Takahashi S, Morita T, Koda Y, Murayama H, Hirasawa Y: Gastrointestinal involvement of dialysis-related amyloidosis. Clin Nephol 30:168-171, 1988 20. Sacs EF, Bloch HM, Milne FJ: Pancreatic supplementation in end-stage renal disease. Nephron 37:120-122, 1984 21. Avram MM: High prevalence of pancreatic disease in chronic renal failure. Nephron 18:68-71, 1977 22. Gilchrest BA, Rowe JW, Mihm MC: Clinical and histological skin changes in chronic renal failure: Evidence for a dialysis-resistant, transplant-responsive microangiopathy. Lancet 2:1271-1275, 1980 23. Kinashi M, Ohbayashi S, Ohbayashi M: Detection of amyloid-like fibrils using labial salivary gland biopsy in patients with nondialysis renal insufficiency. Am J Nephrol 9:435-437, 1989 24. Rahmati MA, Craig RG, Homel P, Kaysen GA, Levin NW: Serum markers of periodontal disease status and inflammation in hemodialysis patients. Am J Kidney Dis 40:983989, 2002 25. Danesh F, Ho LT: Dialysis-related amyloidosis: History and clinical manifestation. Semin Dial 14:80-85, 2001 26. McMurray RW: Hepatitis C-associated autoimmune disorders. Rheum Dis Clin North Am 24:353-374, 1998 27. Guyton AC: The autonomic nervous system, in Guyton AC (ed): Textbook of Medical Physiology (ed 6). Philadelphia, PA, Saunders, 1981, pp 710-722