Spontaneous tendon ruptures in patients on chronic dialysis

Spontaneous Tendon Ruptures in Patients on Chronic Dialysis Niall Jones, MD, and Carl M. Kjellstrand, MD, PhD, FACP, FRCP(C) t Large tendon rupture is a rare catastrophic occurrence in dialysis patients. Pathogenesis of this has been variably thought to be due to malnutrition, insufficient dialysis, amyloidosis, chronic acidosis, or hyperparathyroidism. We investigated contributory causes and timing of this complication in 44 dialysis patients (42 hemodialysis and two peritoneal dialysis patients). Five cases were our own; the other 39 were reported in the literature during the last two decades. Data were compared with a hospital database of 916 patients. The patients who experienced tendon ruptures had been on dialysis longer (mean duration, 7.6 years v 4.0 years; P = 0.001), were younger (mean age, 39.7 years v 48.4 years; P = 0.0001), had much higher parathyroid hormone levels (1,802 pg/mL v 202 pg/mL; P = 0.0001), had a higher phosphate level (6.8 mg/dL v 5.9 mg/dL; P = 0.001), had a slightly higher calcium level (9.2 mg/dL v 8.8 mg/dL; P = 0.038), and had a higher alkaline phosphatase level (649 lU/L v 109 lU/L; P = 0.0001) than control patients. Patients with tendon ruptures had no evidence of malnutrition (albumin 3.7 g/dL v 3.8 g/dL; P = 0.237) and had the same acidosis (bicarbonate 22.2 mEq/L v 22.0 mEq/L; P = 0.180). The time on dialysis to rupture was inversely related to the patient's age (r = 0.47, P = 0.004). Most patients had evidence of years of poorly controlled hyperparathyroidism with high and increasing levels of parathyroid hormone and alkaline phosphatase. Previous steroid use was also much more common in patients with tendon ruptures, as was radiographic evidence of osteitis fibrosa. The disease led to long hospitalization and prolonged morbidity, with mobility limitations in several patients. Spontaneous large tendon rupture in patients is secondary to hyperparathyroidism; is more common in young patients, particularly if exposed to corticosteroids; leads to long-lasting morbidity; and should be preventable by better supervision and treatment of hyperparathyroidism.

© 1996 by the Nationa/Kidney Foundation, Inc. INDEX WORDS: Tendon; hemodialysis; continuous ambulatory peritoneal dialysis; trauma; hyperparathyroidism; corticosteroids.

URVIVAL in dialysis patients is now counted in many decades.t'2 Approximately one in every thousand individuals is now alive on chronic dialysis, and approximately 0.25 million people are now on chronic dialysis in the United StatesJ These patients are subjected to an increased incidence of common diseases, such as cardiovascular complications and musculoskeletal complications, secondary to hyperparathyroidism. 34 However, some diseases or their manifestations are unique to dialysis patients. Among these are /32-microglobulin accumulation and a special form of amyloidosis. 6'7 Even more peculiar and rare is spontaneous tendon rupture. While this is rarely reported in other patients, there has been an increasing number of reports of such complications in dialysis patients. 8-23 We recently encountered two patients who ruptured their patellar tendons, and were struck by the catastrophic event, the long-term morbidity associated with it, and the controversy about its pathogenesis. We therefore reviewed our own experience and that reported in the literature in an effort to ascertain risk factors and etiology with the ultimate goal of trying to prevent this devastating complication.

S

MATERIALS AND METHODS

Patients We reviewed the charts of all patients treated for tendon injuries at our institution between 1985 and 1995. Four of

these patients were also on dialysis, but we excluded one patient as the tendon rupture occurred secondary to severe trauma. Two patients had one episode of rupture, but one patient ruptured large tendons on three occasions: 4.5 years (both quadriceps tendons), 5.5 years (right quadriceps tendon), and 6 years (left quadriceps tendons) after start of dialysis. This patient was entered in our database as three cases, for a total of five cases.

The Literature The literature between the years 1980 and 1995 was searched through the use of Medline, cross-referencing tendon injuries with hemodialysis or peritoneal dialysis. Twelve articles were obtained; four additional articles were then obtained from their bibliographies.

Data Collection The following data were extracted: (1) demographic factors (age, sex, duration of dialysis until tendon rupture, and site of rupture); (2) nutritional factors (albumin, predialysis and postdialysis blood urea nitrogen, and creatinine); (3) chronic acidosis (serum bicarbonate levels); (4) hyperparathyroidism (intact parathyroid hormone [PTH; when PTH was reported

From the Division of Nephrology, Department of Medicine, University of Alberta, Edmonton, Alberta, Canada. Received February 28, 1996; accepted in revised form July 30, 1996. Address reprint requests to Carl M. Kjellstrand, MD, PhD, FACP, FRCP(C), 2E3.31 Walter Mackenzie Centre, University of Alberta, Edmonton, AB, T6G 2B7. © 1996 by the National Kidney Foundation, Inc. 0272-6386/96/2806-001053.00/0

American Journal of Kidney Diseases, Vol 28, No 6 (December), 1996: pp 861-866

861

862

JONES AND KJELLSTRAND

Table 1. Anatomic Location of Ruptured Tendon Ruptured Tendon

No, of Patients

%

Quadriceps Achilles Triceps Patellar Digital extensor Total

41 7 6 5 4 63

65 11 9 8 7 100

as other than the intact level we used the C-terminal data and converted it to a multiple of the upper limit of normal], alkaline phosphatase, calcium, and phosphate levels, and radiologic evidence of osteitis fibrosa); (5) other (when available, /32-microglobulin levels in blood, biopsy results from repaired tendons or surrounding synovium, and co-existing carpal tunnel disease are reported); (6) adequacy of dialysis (estimated by percent urea reduction); and (7) previous steroid use.

Statistical Analysis The data were entered into the Statview IV program (Abacus Concepts Inc; Berkely, CA) and the parameters in patients with tendon rapture were compared by chi-square analysis with the Yates continuity correction, finear regression analysis, and onesample Student's t-test, to a database containing 906 patients, with detailed treatment parameters documented. The data are give as mean values _+ SD.

RESULTS

In our own material, three of 906 (0.3%) dialysis patients at risk experienced tendon rupture over the last 10 years. The mean time on dialysis of our patients was 4.2 years, resulting in one tendon rupture per 760 patient-years. We found 50 cases in the literature. However, nine of these cases were excluded because of insufficient data. 8'1°'1921 Thus, we have data on 44 patients. Table 1 summarizes the anatomic location of the tendon rupture and Table 2 summarizes the relevant demographic and clinical data of our own patients, as well as the cases reported in the literature. Twenty-five (68%) men and 12 (32%) women (patient's sex not being identified in one report) experienced spontaneous tendon rupture. There is no gender difference when compared with all patients in our database. The patients with tendon rupture were significantly younger (mean age, 39.7 _+ 11.6 years v 48.4 years; P = 0.0001, one-sample Student's t-test). The mean duration of dialysis was 7.3 _ 4.6 years, with a minimum of 3 months and a maximum of 16 years; this is much longer than that in our own

database (mean, 4.0 _+ 3.5 years; P = 0.001). Younger patients were on dialysis for a longer period of time than older patients before rupture occurred (r = 0.47, P = 0.004; Fig 1). Although we could only obtain the creatinine in eight of the patients, there was not a significant difference with our database (11.9 mg/dL v 9.6 mg/dL; P = 0.16). Predialysis blood urea nitrogen (89.6 mg/dL v 70 mg/dL; P = 0.0001) was significantly elevated in patients who experienced tendon ruptures, but postdialysis blood urea nitrogen was similar. The adequacy of dialysis as assessed by percent urea reduction was 60%, which did not differ significantly from the population mean of 63% (P = 0.440). Intact PTH, serum phosphorous, and calcium levels were significantly elevated in patients with tendon ruptures. The mean intact PTH was 1,802 pg/mL versus 202 pg/mL (P = 0.0001), the mean phosphorus was 6.8 mg/dL versus 5.9 mg/dL (P = 0.001), and the mean calcium was 9.2 mg/dL versus 8.8 mg/dL (P = 0.038). The patients with tendon rupture also demonstrated a higher (649 IU/L v 109 IU/L) and progressively increasing alkaline phosphatase level prior to the time of tendon rupture (Fig 2). Of 23 patients who had been radiographically investigated for bone disease, evidence of osteitis fibrosa was found in 17 (74%). Neither albumin (3.7 g/dL v 3.8 g/dL; P = 0.237) nor bicarbonate (22.2 mEq/L v 22.0 mEq/ L; P = 0.180) were different in patients with tendon ruptures, but these patients had a lower hemoglobin (8.1 g/dL v 10.3 g/dL; P = 0.002). Of the 44 patients reported, 15 had chronic glomerulonephritis, four had chronic pyelonephritis, four had congenital renal disease, one had hypertensive nephrosclerosis, and 20 had other or unknown renal disease. No patient who experienced tendon rupture had diabetes, while this is the diagnosis of 20% of our database patients (P = 0.001), and only one patient had the diagnosis of hypertensive renal disease compared with 12% of our patients (P = 0.051). At least nine (20%) of the 44 patients with tendon ruptures had documented previous steroid use, compared with only 53 (9%) of 586 other patients in our database (P = 0.023). DISCUSSION

Large tendon rupture is a rare, but seemingly increasingly common complication, almost unique

863

TENDON RUPTURES IN CHRONIC DIALYSIS PATIENTS Table 2. C o m p a r i s o n of 44 Patients Who Experienced Tendon Ruptures to Control Patients Tendon Rupture Patients

N

Control Patients

Probability Values

39.7 -- 11.6 32 7.3 _+ 4.6 89.6 _+ 16.8 39.9 _+ 39.2 60 _+ 9 11.9 _+ 3.9 8.1 + 2.3 3.7 _+ 0.4 22 _+ 3 1,802 _+ 823 9.2 _+ 1.2 6.8 _+ 1.6 649 _+ 439 20

44 37 39 19 7 6 8 15 15 12 20 34 36 31 44

48.4 41 4.0 _+ 3.5 70.0 _+ 16.8 25.2 _+ 23.5 63 _+ 8 9.6 +_ 3.0 10.3 +_ 4.7 3.8 + 0.6 22 _+ 3 202 _+ 234 8.8 +_ 0.8 5.9 _+ 1.6 109" 9%

< 0.0001 0.104 0.001 < 0.0001 0.094 0.449 0.156 0.002 0.234 0.180 < 0.0001 0.038 0.001 < 0.0001 0.023

Age (yr) Women (%) Duration of dialysis (yr) Pre-BUN (mg/dL) Post-BUN (mg/dL) PRU (%) Creatinine (mg/dL) Hgb (g/dL) Albumin (g/dL) Bicarbonate (mEq/L) PTH (pg/mL) Calcium (mg/dL) Phosphorus (mg/dL) Alkaline phosphatase (IU/L) Previous prednisone (%)

NOTE. All comparisons to values for 906 patients in our database, except as noted by the asterisk (*), which is the upper normal value. Abbreviations: BUN, blood urea nitrogen; PRU, percent reduction of urea; Hgb, hemoglobin.

to patients on chronic dialysis. This complication mainly affects young patients who have been on dialysis for a long period of time. The actual incidence of tendon rupture is unknown, but has been reported to be as high as 15%, 11 although this may be an aberration and much higher than the 0.6% observed by us. With increasing age the duration of dialysis until rupture is shorter

and tendon ruptures may become more frequent as more older patients start dialysis and have improved survival. Tendon rupture also has been seen in association with gout, systemic lupus erythematosus, steroids, arteriosclerosis, and primary hyperparathyroidism, all of which are also increased in dialysis patients. 1600-

20.

12004 RV

[]

.

.

.

.

.

.

.

.

.

O 13 Cl

,.

[]

13

<

400 ,,.o AA

.<

0

13

0

[]

"*~'-...

0

D°

0

3;

4;

5~0

o

"o'-.

~;

1'2

2'4

a6

48

60

72

MONTHS ON DIALYSIS

70

Fig 2. Plot of alkaline phosphatase versus duration of dialysis and time to tendon ruptures in our three

Fig 1. Linear regression analysis of the time from start of dialysis to t e n d o n rupture versus age at start of dialysis in patients. There is a strong linear correlation (r = 0.47, P : 0.003) b e t w e e n these factors. It takes a very long time for t e n d o n rupture to o c c u r in y o u n g e r patients, but a much shorter time as patients get older. Perhaps t e n d o n ruptures will b e c o m e m o r e c o m m o n as older patients are increasingly started on dialysis and their survival time increases.

patients. Patient ED had a ruptured t e n d o n on t w o occasions. 0, Start of dialysis. In all patients, alkaline p h o s p h a t e increased years before t e n d o n rupture occurred. Thus, a relentlessly increasing alkaline phosphatase level strengthens the indication for parathyr o i d e c t o m y in y o u n g e r patients w h o will not shortly receive a renal transplant, particularly if they also have received steroids. The solid horizontal line indicates the upper normal limit of alkaline phosphatase.

10

5;

u

864

A number of pathophysiologic mechanisms have been proposed in spontaneous rupture of tendons in dialysis patients. Chronic acidosis has been previously related to a structural disorder of collagen predisposing to tendon rupture. 24 This series of patients demonstrate a mean serum bicarbonate level that is similar to all dialysis patients and within the normal limits. Kurer et al reported five cases of tendon rupture in association with dialysis-related amyloidosis. 2° None of the patients in our series was reported to have the more common musculoskeletal manifestations of amyloidosis, which are severe joint pain and carpal tunnel syndrome. In addition to the absence of carpal tunnel syndrome and joint pain without radiologic changes, in one case in which we were able to obtain a biopsy specimen of synovial tissue, Congo red staining for amyloid was negative. De Franco et al also reported negative congo red staining in both of their cases, which seems to rule out amyloid as a cause. 23 Kurer et al did demonstrate a pattern of age and duration of dialysis in patients with tendon ruptures similar to that of the development of amyloidosis. All 23 of their patients who had been on dialysis for over 20 years were affected. 2° Kurer et al also noted that symptoms of amyloidosis developed earlier in older patients. Our interpretation of these data is that amyloid and tendon ruptures are unrelated, but both occur in patients who have been on dialysis for many years. In 1972, Preston first proposed hyperparathyroidism as being the etiologic culprit in spontaneous tendon ruptureY Since that time a number of investigators have documented a possible relationship between secondary hyperparathyroidism and spontaneous tendon rupture. 14'15'~8'23 In our study the association with biochemical evidence of hyperparathyroidism was very strong. All five of our cases had secondary hyperparathyroidism marked by elevation of PTH, alkaline phosphatase, calcium, and phosphorus. Each of the patients demonstrated a progressive increase in their PTH and particularly their alkaline phosphatase levels in the 24 to 48 months preceding spontaneous tendon rupture. Radiographic evidence of hyperparathyroidism was also common, being present in 74% of investigated patients. As shown in our results, an analysis of the reports from the literature revealed a significant associa-

JONES AND KJELLSTRAND

tion of spontaneous tendon rupture with biochemical data of hyperparathyroidism. Spontaneous rupture of tendons in a young person is rare; the normal quadriceps tendon can withstand forces of up to 15 to 40 kg/mm. Tendon rupture in this study was found to occur at a significantly younger age than that of the general dialysis population. The explanation for this oddity may lie in the fact that young people retain their muscle mass better than older people, and therefore are more likely to be able to produce the force necessary to cause their tendons to rupture. The older patient may do so because the structure of the tendon was weakened prior to the initiation of dialysis and the development of hyperparathyroidism. As shown in Fig 1, ruptured tendons in older patients occur much earlier after the onset of dialysis than in younger patients. As noted earlier, a similar pattern was demonstrated by Kurer et al in the development of amyloidosis-related musculoskeletal disease. 2° Another finding in our patients was the absence of diabetes, the leading cause of end-stage renal failure in most Western countries. 1 Hyperparathyroid bone disease is rarer in diabetic patients than in other patients. 26"27Also notable was a very low incidence of hypertensive nephrosclerosis; this was documented in only one of our patients, while it causes over 20% of all endstage renal failure in the United States. 1 This finding may perhaps be a reflection of the fact that large tendon rupture is a disease of young dialysis patients. No less than nine of the patients reviewed by us had used steroids previously, which is much higher than usually experienced in a dialysis population. It appears as if previous steroid use is an independent risk factor in dialysis patients. There are several reports of tendon ruptures in patients with normal renal function following

corticosteroids.28'29 In cases for which we had details, it also was clear that it takes a prolonged period of time for secondary hyperparathyroidism to cause weakness of the tendon to the point of rupture. The average patient had been on dialysis for 7 years 3 months, and in our three patients there was clear evidence of hyperparathyroidism with rising alkaline phosphatase level, a sharply increasing parathyroid level, and inadequate treatment of hyperphosphatemia. Apparently these warning

TENDON RUPTURES IN CHRONIC DIALYSIS PATIENTS

signs were either overlooked or it was thought that treating them w o u l d be more dangerous than the possible side effects. Our data indicate that secondary hyperparathyroidism, besides its more w e l l - k n o w n effects on bone, is also dangerous for tendon integrity. The m e c h a n i s m s that have been postulated for this are either a w e a k e n i n g of the insertion site secondary to cortical b o n e resorption la or a direct effect of P T H on the tendon matrix by a depolymerization of glycoproteins. 25 Whatever the mechanism, we believe that the evidence is strong e n o u g h to r e c o m m e n d more intensive treatment of dialysis patients with evidence of secondary hyperparathyroidism. The treatment is well established in the literature: the strict control of hyperphosphatemia, the early use of vitamin D analogues, and in recalcitrant cases, parathyroidectomy with or without autotransplantation of part of a parathyroid gland. 5 The indications for parathyroidectomy in the patient who has radiographic evidence of osteitis fibrosa are much increased if the patient also has been exposed to steroids, and particularly if he or she faces a long time on dialysis. In summary, our analysis indicates that the primary pathogenesis of tendon rupture in dialysis patients is hyperparathyroidism. The risk of rupture is increased if the dialysis patient has been on steroids, and therefore the failed transplant patient, who often faces a very long time on dialysis, would be most at risk. Our analysis also reveals that the y o u n g e r dialysis population is more prone to this complication, although this epidemiologic feature m a y change with the changing demographics of the dialysis population, in that older patients m a y rupture their tendons after a shorter duration of dialysis. REFERENCES

l. US Renal Data System: USRDS 1995 Annual Report. Bethesda, MD, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 1995 2. Charra B, Calemard E, Ruffet M, Chazot C, Terrat J-C, Vanel T, Laurent G: Survival as an index of adequacy of dialysis. Kidney Int 41:1286-1291, 1992 3. Bloembergen WE, Port FK, Mauger EA, Wolfe RA: Causes of death in dialysis patients: Racial and gender differences. J Am Soc Nephrol 5:1231-1242, 1994 4. Iseki K, Kawazoe N, Osawa A, Fukiyama K: Survival analysis of dialysis patients in Okinawa, Japan (1971-1990). Kidney Int 43:404-409, 1993 5. Torres A, Lorenzo V, Herfiandez D, Roddguez J,

865

Concepclaaon MT, Roddguez AP, Herfifindez A, de Bonis E, Darias E, Gonzfilez-Posada G, Losada M, Rufino M, Felsenfeld AJ, Roddguez M: Bone disease in predialysis, hemodialysis, and CAPD patients: Evidence of a better bone response to PTH. Kidney Int 47:1434-1442, 1995 6. Stein G, Schneider A, Thoss K, Ritz KT, Linke RP, Schaefer K, Sperschneider H, Abendroth K, FUnfst~ickR: Beta-2-microglobulin-derivedamyloidosis: Onset, distribution and clinical features in 13 hemodialysedpatients. Nephron 60:274-280, 1992 7. Campistol JM, Skinner M: B2-microglobulin amyloidosis: An overview. Semin Dial 6:117-126, 1993 8. Brotherton BJ, Ball J: Bilateral simultaneousrupture of the quadriceps tendons. Br J Surg 62:918-920, 1975 9. MoreinG, GoldschmidtZ, Pauker M, SeelenfreundM, Rosenfeld JB, Filed A: Spontaneous tendon ruptures in patients treated by chronic hemodialysis. Clin Orthop 124:209213, 1977 10. Siwek KW, Rao JP: Bilateral simultaneousrupture of the quadriceps tendons. Clin Orthop 131:252-254, 1978 11. Lotem M, Bemheim J, Conforty B: Spontaneousrupture of tendons: A complication of hemodialyzed patients treated for renal failure. Nephron 21:201-208, 1978 12. Meneghello A, Bertoli M: Tendon disease and adjacent bone erosion in dialysis patients. Br J Radiol 56:915920, 1983 t3. Bhole R, Flynn JC, Marbury TC: Quadriceps tendon ruptures in uremia. Clin Orthop 195:200-206, 1985 14. Shoji R, Kawaguchi Y, Oda Y, Kimura Y, Kasai K, Ogura Y, Miyahara T, Nakao T, Miyaji T, Akaba K: Tendon rupture in patients on chronic dialysis. Nippon Jinzo Gakkai Shi 28:455-460, 1986 15. Anderson WE, Habermann ET: Spontaneous bilateral quadriceps tendon rupture in a patient on hemodialysis. Orthop Rev 4:411-414, 1988 16. SpencerJD: Spontaneousrupture oftendonsindialysis and renal transplant patients. Injury 19:86-88, 1988 17. Clark RF, Popky LM, Evans TC: Spontaneous fourextremity extensor tendon rupture in a renal dialysis patient. Ann Emerg Med 18:783-784, 1989 18. Ryuzaki M, Konishi K, Kasuga A, KumagaiH, Suzuki H, Abe S, Saruta T, Takami H, Tashiro M: Spontaneous rupture of the quadriceps tendon in patients on maintenance hemodialysis--Report of three cases with clinicopathological observations. Clin Nephrol 32:144-148, 1989 19. Btihren V, Boblet R: Spontane and simultane doppelseitige Quadricepssehnenrupturenbei Hyperparathyreoidismus. Chirurg 60:788-790, 1989 20. Kurer MH, Baillod RA, Madgwick CA: Musculoskeletal manifestations of amyloidosis. J Bone Joint Surg 73B:271-276, 199l 21. Zouari M, Douik M, Karray S, Litaiem T, Slimane N: Rupture spontaneebilaterale de l'appareil extenseur du genou chez l'insuffisantrenal chronique.Tunis Med 71:343-345, 1993 22. Vellani G, Dallari D, Fatone F, Martella D, Bonomini V, Gualtiere G: La sindrome del tunnel carpale nei pazienti in trattamento emodialitico periodico. Chir Organi Mov 48:15-18, 1993 23. De Franco P, Varghese J, Brown WW, Bastani B: Secondary hyperparathyroidism and not /32-microglobulin amyloid, as a cause of spontaneous tendon rupture in pa-

866

tients on chronic hemodialysis. Am J Kidney Dis 24:951955, 1994 24. Finlayson GR, Smith JG, Moore MJ: Effects of chronic acidosis on connective tissue. JAMA 187:659-662, 1964 25. Preston ET: Avulsion of both quadriceps tendons in hypcrparathyroidism. JAMA 221:406-407, 1972 26. Pei Y, Hercz G, Greenwood C, Segre G, Manuel A, Saiphoo C, Fenton S, Sherrard D: Renal osteodystrophy in diabetic patients. Kidney Int 44:159-164, 1993 27. Aubia J, Bosch J, Lloveras J, Marinoso L, Masramon

JONES AND KJELLSTRAND

J, Serraon S, Cuevas X, Orfelia A, Llorach I, Llorach M: Low incidence of hyperparathyroidism in diabetic renal failure. Proc Eur Dial Transplant Assoc Eur Ren Assoc 21:902908, 1985 28. Burchhardt H, Krebs U: Simultaneous and subsequent bilateral spontaneous Achilles tendon ruptures after steroid therapy and in diabetes mellitus. Chirurg 62:830831, 1991 29. Stefanich RJ: Intra-articular corticosteroids in treatment of osteoarthritis. Orthop Rev 15:65-71, 1986