Renal Vein Thrombosis

Renal Vein Thrombosis

THE JOURNAL OF UROLOGY Vol. 116, October Copyright© 1976 by The Williams & Wilkins Co. Printed in U.S.A. RENAL VEIN THROMBOSIS M. J. O'DEA, R. S. ...

369KB Sizes 1 Downloads 110 Views

THE JOURNAL OF UROLOGY

Vol. 116, October

Copyright© 1976 by The Williams & Wilkins Co.

Printed in U.S.A.

RENAL VEIN THROMBOSIS M. J. O'DEA, R. S. MALEK,* R. M. TUCKER

AND

R. E. FULTON

From the Department of Urology, Division of Nephrology, and Department of Diagnostic Radiology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota

ABSTRACT

The manifestations, clinical course and treatment of 14 patients with non-malignant renal vein thrombosis are described. Most patients (10 of 14) had generalized vague illness and nephrotic syndrome but 4 were initially seen with acute symptoms of flank pain, hematuria or hypertension. Renal vein thrombosis affected young men 2.5 times more often than women and occurred on the left side 2.6 times more commonly than on the right or both sides. Red blood cell casts in the urinary sediment, heavy proteinuria and hypoalbuminemia were useful indicators of the disease. Excretory urographic signs were suggestive of renal vein thrombosis in all patients and these were corroborated by angiographic studies. Systemic anticoagulation with or without a renal failure program and diuretics, or simply a combination of the last 2 modalities, was used in 9 patients. In 2 of the 9 patients who were unresponsive the adjuvant use of cyclophosphamide and steroids effected a cure. The remaining 5 patients underwent nephrectomy or thrombectomy. All 14 patients were followed for 1 to 7 years (mean 1.6 years). Ten patients were cured or improved, 1 patient was unchanged, and in the remaining 3 patients the condition deteriorated and they subsequently required a renal allograft. The rationale for various forms of treatment is discussed. Renal vein thrombosis was first described by Rayer in 1837. 1 In contrast to an aggressive surgical approach, characterized by thrombectomy, which was proposed earlier by some authors, 2 others have proposed conservative management by systemic anticoagulation. 3 During a 10-year period (1963 through 1972) 24 adult patients with renal vein thrombosis that did not result from renal malignancy were seen at our clinic. In 10 patients renal vein thrombosis was diagnosed only at postmortem examination. These patients had died of various disease states. The findings in these 10 patients included retroperitoneal lymphoma in 3, chronic brain disease in 2, iliofemoral thrombosis in 1, Rathke pouch tumor in 1, renal failure in 1, severe generalized thermal bums in 1 and hepatic carcinoma with invasion of the inferior vena cava and renal veins in 1. Treatment was directed at the correction of the primary disease and is beyond the scope of this communication. Experience gained in the diagnosis and management of the remaining 14 patients with clinically proved renal vein thrombosis is presented herein. CLINICAL MATERIAL

Of the 14 patients in whom the diagnosis of renal vein thrombosis was entertained at the time of clinical presentation 10 were men and 4 were women, with a ratio of 2.5 to 1. Patient ages ranged from 15 to 63 years, with a mean of 36 years. Renal vein thrombosis occurred most commonly in the fifth decade (6 of 14 patients). Symptomatology is summarized in table 1. Most often the presenting symptoms were of a chronic nature and consisted of nausea, apathy, weakness and generalized edema (10 of 14 patients). Less often, renal vein thrombosis presented acutely with sudden flank pain, with or without hematuria or hypertension (4 of 14 patients). A history ofrecent blunt abdominal trauma was obtained in only 1 of the latter group of patients. Association, Coronado, California, February 26, 1976. Read at annual meeting of Western Section, American Urological Association, Coronado, California, February 22-26, 1976. * Requests for reprints: Department of Urology, Mayo Clinic, 200 First St. S. W., Rochester, Minnesota 55901. 410

INVESTIGATIONS AND FINDINGS

Heavy proteinuria (urinary protein more than 3 gm. per 24 hours) was found in 9 of the 10 patients; the remaining patient had moderate proteinuria (more than 0.4 to less than 3.0 gm. per 24 hours). Red blood cell casts were found in 8 of 14 patients. A reduced glomerular filtration rate as judged by elevation of serum creatinine concentration (1.3 to 9.3 mg. per di.) or diminished clearance of creatinine (75 to 16 ml. per minute per 1.73 m. 2) was found in 12 of 14 patients who had one or the other or both tests; serum creatinine was normal in the remaining 2 patients. Serum protein concentration was measured in 10 patients and all had hypoalbuminemia (less than 3.3 gm. per di.), while 8 had total hypoproteinemia (less than 7.2 gm. per dl.). Excretory urographic (IVP) signs suggestive of renal vein thrombosis were found in all 14 patients. These included increased size of one or both kidneys, especially in the acute form in 14 patients, diminished density of the nephrogram and decreased concentration of contrast medium in the collecting system in 5, stretching and narrowing of the collecting system in 3 (fig. 1), indentations of the renal pelvis and infundibula in 2 (fig. 2), and ureteral notching, especially on the left side from engorged collateral venous channels-gonadal, ureteral, lumbar, adrenal, azygos and renal capsular in 3 (fig. 3). Results of at least 1 form of angiographic study-inferior venocavography, renal venography or selective renal arteriography-were diagnostic in 11 patients who underwent some of these studies. The significant findings in these 11 patients were the presence of thrombus in one or both renal veins in 9 patients, protrusion of thrombus into the inferior vena cava in 3 (fig. 4) and additional filling defects in the inferior vena cava and iliofemoral veins in 3 (fig. 5). Furthermore, extensive perirenal venous collateral circulation was demonstrated in 5 patients. Delayed transit of contrast medium and non-filling of the renal vein during selective arteriography were suggestive of the diagnosis in 3 patients. Thrombosis involved the left renal vein in 8 patients, right renal vein in 3 and both renal veins in the remaining 3. Percutaneous needle renal biopsy was done in 7 patients. Chronic membranous glomerular changes consistent with renal

RENAL VEIN THROMBOSIS

411

vein thrombosis were found in 6 patients (fig. 6); material available in the remaining patient was insufficient for interpretation. TREATMENT AND RESULTS

Nine patients were managed conservatively with oral anticoagulants. In 2 of those who were unresponsive anticoagulation was supplemented by cyclophosphamide and prednisone and in 1 by a renal failure program, whereas in 2 other paTABLE

1. Clinical presentation of renal vein throm basis in 14 patients No. Pts.*

Peripheral edema: Severe, 9 Mild, 1 Flank pain Hematuria Renal failure Hypertension Recent chest pain Recent iliofemoral thrombosis * Thirteen patients had more than 1 symptom.

10

4 3 4 3 2

1

FIG. 2. IVP shows persistent irregularity of upper pole calix and infundibulum, and scalloped appearance of upper ureter produced by collateral circulation in patient with proved and successfully treated left renal vein thrombosis.

FIG. 1. A, IVP shows enlarged left kidney (pole-to-pole measurement 17.7 cm. compared to right, which is 14.3 cm.), with a smooth outline, incomplete distension of collecting system, stretching and attenuation of infundibula and calices and mucosa! irregularity of renal pelvis. B, selective left renal venogram shows thrombus in proximal left renal vein (note second catheter in left renal artery). There is some filling of proximal left inferior phrenic and adrenal veins.

FIG. 3. IVPs. A, persistent irregularity of upper pole infundibulum and renal pelvis (arrows). B, ureteral notching (arrow) produced by collateral circulation. Patient had undergone left renal venous and inferior vena caval thrombectomy 2 years previously.

412

o'DEA AND ASSOCIATES

FIG. 4. A, IVP (20 degrees are tomograms) shows enlarged kidneys resulting from bilateral renal vein thrombosis. Renal outlines are smooth. Collecting systems are incompletely filled, distorted and attenuated. There is diminished nephrographic density. B, inferior venacavogram shows irregular radiolucent filling defects (renal vein thrombi) protruding into lumen ofvena cava bilaterally (arrows). C, selective left renal venogram shows abnormal, recanalized left renal vein. Radiolucent streaks and filling defects in left renal, phrenic and adrenal veins represent thrombus. There is filling of lumbar and gonadal veins not normally seen. Inner arrow shows point of complete left renal vein obstruction. Outer 2 arrows show contrast medium in calices.

FIG. 5. Left external iliac and inferior venacavogram in patient with renal vein thrombosis shows radiolucent filling defect representing thrombus just below bifurcation. Note washout effect from right external iliac venous effluent and retrograde filling of collaterals, which arise proximal to point of obstruction.

tients anticoagulation was soon discontinued because of a bleeding disorder and replaced by the renal failure program and diuretics (table 2). Dosages were approximately 2 mg. per kg. per day cyclophosphamide (depending on white blood cell count) and 0.5 to 1.0 mg. per kg. per day prednisone.

A response to this regimen is characterized by a decrease in 24-hour urinary protein excretion of more than 50 per cent or to less than 1 gm. per 24 hours. Dosage may then be adjusted according to individual variation. If there is no response to this therapy after 8 weeks cyclophosphamide is stopped and prednisone is tapered off and discontinued. Surgical intervention in the remaining 5 patients included nephrectomy in 4 and thrombectomy in 1 (table 2). In all 4 patients who underwent nephrectomy the clinical presentation had been acute, with flank pain, hematuria or hypertension. In 3 of these 4 patients renal vein thrombosis was discovered after some form of trauma-8 days after pelviolithotomy in one, 1 year after creation of splenorenal shunt in another and 1 week after blunt abdominal trauma in the third. The fourth patient had had nephrotic syndrome and subsequently experienced left flank pain. Neither in this patient nor in the one who underwent inferior vena caval and renal venous thrombectomy was a history of any form of trauma elicited. The various treatment modalities and their results are outlined in table 2. All 14 patients were followed for 1 to 7 years (mean 1.6 years). Seven patients were cured (urinary protein less than 400 mg. per 24 hours) and 3 were improved. One patient remained unchanged and the condition deteriorated in the remaining 3, who later received renal allografts (table 2). DISCUSSION

Rayer's description of renal vein thrombosis and its association with the nephrotic syndrome 1 was soon followed by at-

RENAL VEIN THROMBOSIS

FIG. 6. Glomerulopathy of renal vein thrombosis. A, renal glomerulus exhibits generalized thickening of capillary wall in absence of cellular proliferation. H & E, reduced from x400. B, thickened renal glomerular basement membrane and subepithelial projections of basement membrane (spikes). Jones methenamine silver, reduced from x 1,000. C, renal glomerular immunofluorescence shows 3 plus diffusely granular pattern similar to that found in chronic membranous glomerulopathy. IgG, reduced from x 220. Similar staining with fibrinogen and beta-1 complement was demonstrated. TABLE

2. Treatment of renal uein thrombosis and results No. Pts.

Followup-Mean (yrs.)

Anticoagulants

4

1 to 4-2.5

Anticoagulants, cyclophosphamide and prednisone Anticoagulants and renal failure program Renal failure program and diuretics (anticoagulants contraindicated)

2 1 2

1.5 and 2.5-2 1-1 1 and 1.5-1.:3

Nephrectomy

4

1 to 7-2

Thrombectomy

1-1

Results Cured*-1 lmproving-1 Unchanged-! Deteriorated (transplant)-! Both cured* Improving lmproving-1 Deteriorated (transplant)-] Cured*-3 Deteriorated (transplanl)-1 Cured*

., 24-hour urinary protein less than 400 mg. (upper limit of normal in our laboratories).

tempts at producing this condition in experimental animals. •-s The success of the experiments varied but enlargement of the involved kidney and proteinuria were noted. However, ligation of a traumatized left renal vein in a young man was later reported to be unaccompanied by any urographic changes or varicocele formation. 7 The rather benign adult form of renal vein thrombosis was differentiated from the more serious and often fatal infantile variety almost 100 years after recognition of the disease. 8 Reviews indicate that until just more than 10 years ago most cases of renal vein thrombosis, such as those seen earlier in our series, were diagnosed at postmortem examination. In most of these patients (80 per cent) the nephrotic syndrome developed 2 months to 2 years before death. 9 • 10 More recent experience here and in other centers shows that clinical recognition of renal vein thrombosis is becoming more frequent. 11 It is important to remember that in contrast to most who are seen initially with chronic symptoms such as nausea, and ( 10 of 14 patieP.ts in our series) a fev1 rnay have acute with flank hematuria and cur series) aJ:-td may thus i

infrequently, a history of surgical or accidental trauma may be elicited. Careful examination of the urinary sediment and estimation of 24-hour urinary protein content are important diagnostic steps. Microscopic hematuria, although present in all of our patients, is a non-specific indicator. Red blood cell casts, which are found in a majority of patients (8 of 14 in this series), are a more reliable index of renal parenchymal disease. Heavy proteinuria (more than 3 gm. per 24 hours) is also present in most cases of renal vein thrombosis (9 of 14 in our series). It is a sensitive indicator of the severity of the disease and of the response to therapy. Improvement or deterioration in over-all renal function, measured by concentration of serum creatinine or its clearance, parallels changes in 24-hour urinary protein excretion. Total hypoproteinemia is another helpful but nonspecific indicator in most patients (8 of 10 in our series). However, hypoalbuminemia is a more specific and constant finding (10 of 10 patients in our series). Radiographic studies are important in the vein thrombosis. An IVP is sei~ies). Increase in size and of ~~e inf 1r.-dibt:la and dirninished coEcentration of 1

414

O'DEA AND ASSOCIATES

contrast medium are usual findings, with evidence also of augmented collateral circulation such as ureteropelvic notching. Inferior venacavography or selective renal venography may disclose characteristic presence of thrombus and associated filling of perirenal collateral venous channels. These findings are essential for the diagnosis of renal vein thrombosis. Indeed, renal venography is reported to have demonstrated renal vein thrombosis in a third of a group of unselected patients with the nephrotic syndrome. 12 In contrast to the experience of Miller and associates 3 thromboembolism was not a complication of venography in our series, even though 3 of our patients had had pulmonary embolism before venography. We do not recommend placing a selective catheter in the renal vein if an obvious thrombus has been diagnosed. Selective renal arteriography should be performed initially in these patients. Delayed transit of contrast medium in the renal vessels, a poor nephrogram, non-filling of the main veins and evidence of perirenal venous collaterals may establish the diagnosis without an absolute need for renal venography. Arteriography may offer additional useful information about unsuspected pathology. In our experience renal vein thrombosis on the left side is 2.6 times more common than bilateral disease or disease on the right side. The glomerular lesions of renal vein thrombosis are similar, if not identical, to those observed in chronic membranous glomerulonephritis. Diffuse and generalized glomerular basement membrane thickening, interstitial edema, leukocyte stasis in the glomerular capillaries and deposits in subepithelial locations have been observed on electron microscopy. 18 The pathogenesis of the glomerular and tubular lesions is a matter of speculation. It has been postulated that gradual or slow occlusion of the veins would result in increased capillary pressure, especially in the peritubular regions, with resultant damage to the tubules and glomeruli. 11 • 13 Systemic anticoagulation with coumadin-like drugs is the most popular form of treatment for renal vein thrombosis. 3 , 11 • 12• 14 To this regimen a renal failure program and diuretics may be added (table 2). A combination of the last 2 modalities may be used to replace anticoagulation if bleeding becomes troublesome. On the other hand, massive proteinuria may persist despite anticoagulant therapy since it does not affect the glomerulonephritis that is associated with renal vein thrombosis. 14 To this end, and for the first time in our knowledge, cyclophosphamide and prednisone were used in the treatment of 2 unresponsive nephrotic patients with renal vein thrombosis. Both patients were cured (table 2). Surgical intervention is occasionally indicated and is usually undertaken in those patients in whom the disorder presents acutely. Traditionally, and especially for bilateral disease, thrombectomy appears to be the surgical treatment of choice. 2 • 15 However, in the acute form of the disease, with thrombosis limited to the renal vein, nephrectomy seems to be equally effective in achieving a cure. Deterioration in the condition of one of our patients (who had long-standing

nephrotic syndrome) after removal of a painful kidney would seem to suggest that conservative treatment or perhaps thrombectomy is preferable to nephrectomy in such patients. A high mortality rate (70 to 80 per cent) has been reported for patients with renal vein thrombosis, some of whom have been followed for up to 12 ½ years. 10 • 11 Our 14 patients have been followed for a much shorter period (1 to 7 years, mean 1.6 years). Most patients (10 of 14) have been cured or improved. None has died but 3 have required renal allograft and 1 has remained unchanged.

REFERENCES

1. Rayer, P. F. 0.: Traite des Maladies des Reins. Paris: J. B. Bailliere, 1837-1841. 2. Abeshouse, B. S.: Thrombosis and thrombophlebitis of the renal veins. Urol. & Cutan. Rev., 49: 661, 1945. 3. Miller, R. A., Tremann, J. A. and Ansell, J. S.: The conservative management of renal vein thrombosis. J. Urol., 111: 568, 1974. 4. Robinson, G.: Researches into the connection existing between an unnatural degree of compression of the blood contained in the renal vessels, and the presence of certain abnormal matters in the urine. Trans. Roy. Med. Chir. Soc., 26: 51, 1843. 5. Buchwald, A. and Litten, M.: Ueber die Structurveranderungen der Niere nach Unterbindung ihrer Vene. Virchows Arch. Path. Anat., 66: 145, 1876. 6. Omae, T., Masson, G. M. C. and Corcoran, A. C.: Experimental production of nephrotic syndrome following renal vein constriction in rats. Proc. Soc. Exp. Biol. Med., 97: 821, 1958. 7. Jennings, E. R. and Glucksman, M.A.: Renal vein ligation (letter to the editor). J.A.M.A., 213: 1905, 1970. 8. Derow, H. A., Schlesinger, M. J. and Savitz, H. A.: Chronic progressive occlusion of the inferior vena cava and the renal and portal veins, with the clinical picture of the nephrotic syndrome: report of a case, with a review of the literature. Arch. Intern. Med., 63: 626, 1939. 9. McCarthy, L. J., Titus, J. L. and Daugherty, G. W.: Bilateral renal-vein thrombosis and the nephrotic syndrome in adults. Ann. Intern. Med., 58: 837, 1963. 10. Kowal, J., Figur, A. and Hitzig, W. M.: Renal vein thrombosis and the nephrotic syndrome with complete remission. J. Mt. Sinai Hosp., 30: 47, 1963. 11. Pollak, V. E., Pirani, C. L., Seskind, C. and Griffel, B.: Bilateral renal vein thrombosis. Clinical and electron microscopic studies of a case with complete recovery after anticoagulant therapy. Ann. Intern. Med., 65: 1056, 1966. 12. Llach, F., Arieff, A. I. and Massry, S. G.: Renal vein thrombosis and nephrotic syndrome. A prospective study of 36 adult patients. Ann. Intern. Med., 83: 8, 1975. 13. Rosenmann, E., Pollak, V. E. and Pirani, C. L.: Renal vein thrombosis in the adult: a clinical and pathologic study based on renal biopsies. Medicine, 47: 269, 1968. 14.•Arruda, J. A. L., Gutierrez, L. F., Jonasson, 0., Pillay, V. K. G. and Kurtzman, N. A.: Renal-vein thrombosis in kidney allografts. Lancet, 2: 585, 1973. 15. Cohn, L. H., Lee, J., Hopper, J. and Najarian, J. S.: The treatment of bilateral renal vein thrombosis and nephrotic syndrome. Surgery, 64: 387, 1968.