Acute and chronic renal impairment resulting from urinary tract obstruction is common among patients with cancer and not solely limited to patients with pelvic tumors. Unlike benign causes of urinary tract obstruction, urinary obstruction resulting from malignancies poses several unique clinical challenges. The rate of recurrence of the obstruction and complications from measures undertaken to relieve the obstruction, tends to be higher in the cancer population. The recovery of renal function is related to the severity and duration of obstruction. Hence it is important to diagnose and treat urinary obstruction in cancer patients promptly.
A 57-year-old woman was evaluated in the onconephrology clinic for elevated creatinine and hypertension. She had a history of stage III ovarian carcinoma treated with debulking surgery, followed by chemotherapy with cisplatin, paclitaxel, and bevacizumab 9 years ago. She was diagnosed with disease progression 7 years after the initial diagnosis and was treated with gemcitabine maintenance therapy. After 2 years of being on gemcitabine therapy, she was found to have schistocytes in the peripheral smear, normal serum creatinine, nonnephrotic range proteinuria, and a new diagnosis of hypertension requiring three antihypertensive agents. Imaging of her abdomen showed incidental finding of new left-sided hydronephrosis and dilatation of the ureter to the level of midpelvis ( Fig. 31.1 ). In addition, peritoneal involvement of malignancy was noted to be stable. She underwent retrograde ureteric stent placement with exchanges every 3 to 4 months. She continued to have difficult to control hypertension and eventually developed acute renal dysfunction for which she was referred to the onconephrologist. Repeat imaging was obtained without contrast and this showed stable peritoneal carcinomatosis and nonobstructed kidneys with left-sided ureteric stent in place. She subsequently underwent a biopsy of the right kidney, which showed features of acute and chronic thrombotic microangiopathy. Gemcitabine was discontinued, and chemotherapy was switched to paclitaxel. Hypertension control and kidney function improved, and she survived another 2 years before succumbing to her cancer.
As illustrated in the aforementioned case vignette, obstructive nephropathy in cancer patients often presents insidiously and in conjunction with other causes of renal dysfunction. This is in stark contrast to benign causes of urinary obstruction, such as nephrolithiasis or benign prostatic hyperplasia, where the clinical presentation is quite straightforward. The clinical team should perform a careful review of the patient’s history, current and previous cancer therapies to manage the renal dysfunction, and obstruction to the urinary tract. A high degree of suspicion and vigilance is warranted in managing obstructive nephropathy in the onconephrology world.
Hydronephrosis is defined as dilation of the renal pelvis and calyxes proximal to the point of obstruction. Obstructive uropathy refers to blockage of urine flow because of a functional or structural derangement, anywhere from the tip of the urethra back to the renal pelvis that increases pressure proximal to the site of obstruction. Obstructive uropathy may or may not cause renal parenchymal damage. Such functional or pathologic parenchymal damage is referred to as obstructive nephropathy . Hydronephrosis and obstructive uropathy are not interchangeable terms—dilation of the renal pelvis and calyces can occur without obstruction and urinary tract obstruction may occur in the absence of hydronephrosis.
Tumors causing urinary tract obstruction
Urinary obstruction may occur anywhere along the urinary tract, but certain anatomic sites are more prone for obstruction. This is due to the physiologic narrowing of the tract, which increases the risk of obstruction at the uretero pelvic junction, the crossing of the ureter over the common iliac vessels at the pelvic brim and the ureterovesical junction. Ureteric diameter varies along its course: the diameter is 2 to 3 mm at the ureteropelvic junction, it widens to 10 mm until it reaches the pelvic brim and narrows again to 4 to 6 mm, and finally it is narrowest at the ureterovesical junction, about 1 to 5 mm. In women, a fourth area of ureteric narrowing exists at the level of pelvic blood vessels and broad ligament. These are sites of physiologic narrowing of the ureteric diameter and hence prone for obstruction from intrinsic causes, such as calculi. In the setting of malignancy, ureteral obstruction may result from direct tumoral invasion of the ureter or extrinsic compression from lymph nodes or encasement of the ureters in tumoral tissue. Pelvic malignancies, such as prostate, bladder, cervical, uterine, ovarian, and colorectal tumors can cause obstruction by direct metastatic involvement or by external compression.
Given the anatomy of the urinary tract, differences exist in how urinary obstruction affects male and female patients. The commonest malignancies associated with urinary obstruction are cervical cancer in women and prostate cancer in men. Prostate, cervical, and bladder tumors comprise about three-fourths of the tumors causing urinary tract obstruction. The rest are breast cancer, gastrointestinal malignancies, and lymphomas causing urinary obstruction. About 10% of patients with prostate cancer present with or develop symptomatic urinary obstruction during the course of their illness. A significant proportion of patients with colorectal cancer also develop hydronephrosis with renal dysfunction. In fact, any widely metastatic malignancy causing extensive retroperitoneal lymphadenopathy can cause urinary obstruction by extrinsic compression or by causing peritoneal carcinomatosis. In addition, obstruction may also result from radiation therapy and pelvic lymphoceles.
In terms of tumors causing bilateral ureteral obstruction, cervical cancers are the commonest cause followed by stomach cancer and urologic malignancies. Primary tumors of the ureter and urethra are quite uncommon. Ureteric tumors represent only about 2.5% to 5% of transitional cell carcinomas of the urinary tract. Urethral tumors are 4 times more common in women compared with men but are generally uncommon in onconephrology practice.
Pathophysiology of urinary obstruction
Acquired obstructive nephropathy in humans results from partial urinary obstruction in most cases and tends to be prolonged in its clinical course. But most physiologic studies of renal function in obstruction are based on models of acute complete obstruction for 24 hours. In the case of cancer patients, the physiology can be altered by numerous other factors, including changes in patients’ body weight, nutritional status, and vascular tone. Regardless, the animal models of acute urinary obstruction illustrate several key elements in the underlying pathophysiology.
Urinary obstruction significantly alters renal blood flow, glomerular filtration rate (GFR), and tubular function even before anatomic changes occur in the kidney. , Within the first 2 to 3 hours of obstruction, there is an early vasodilator response; termed the ‘ hyperemic phase. ’ The rise in hydrostatic pressure in the proximal tubule initially results in reduced resistance of the afferent arteriole and increased glomerular hydrostatic pressure to counteract the proximal tubular pressure. Therefore there is afferent arteriolar dilatation. This effect is demonstrably inhibited by nonsteroidal antiinflammatory drugs (NSAIDs) and is also seen in denervated kidneys, indicating that this is an intrarenal autoregulatory response to obstruction. The reduced distal tubular flow contributes to the initial rise in single nephron GFR as part of tubuloglomerular feedback. Thus in the initial phase of obstruction, single nephron GFR is maintained at approximately 80% of the preobstruction values, despite the marked increase in proximal tubular pressure. As obstruction persists in the next 12 to 24 hours, there is a late vasoconstrictor phase, which is characterized by a drop in renal blood flow to about 40% of normal and poor renal perfusion.
Two major vasoconstrictors, angiotensin II and thromboxane A2, play an important role in the markedly reduced renal blood flow and reduction in single nephron GFR in obstruction. After the obstruction is relieved, there is further vasoconstrictor response in the kidney caused by the release of angiotensin II, as the macula densa senses the change in tubular flow. In animal experiments, simultaneous inhibition of thromboxane A2 and angiotensin production normalized GFR in the postobstructed kidney. The administration of atrial natriuretic peptide after release of obstruction in rats also resulted in an increase of GFR, urine flow, and sodium excretion, suggesting a role of atrial natriuretic peptide in the hemodynamic changes of the postobstructed kidney.
Obstruction affects tubular function by reducing the ability of renal tubules to transport sodium (Na+), potassium (K+), and hydrogen (H+), and reduces the ability to concentrate and dilute the urine. , This contributes to postobstructive diuresis. For example, the apical Na-K-2 chloride (Cl) cotransporter and the basolateral Na+K+ adenosine triphosphatase in fresh suspensions of medullary thick ascending loop cells of obstructed kidneys, show reduced transporter activity. Severe downregulation of aquaporin 2 expression contributes to the impaired urinary concentrating ability. The local increase in prostaglandin E2 synthesis in postobstructed kidney is thought to play a role in aquaporin 2 downregulation. Significant downregulation of apical membrane expression of the distal convoluted tubule Na+ Cl- cotransporter also occurs from obstruction. , The defect in H+ and K+ secretion in the distal nephron in obstructive uropathy has been shown to be independent of aldosterone.
In the first few days after onset of obstruction, there is interstitial edema and an influx of leukocytes, predominantly macrophages, into the kidneys. If the obstruction is persistent and not relieved, glomerular size decreases, tubular cells lose apical microvilli and basolateral interdigitations and have fewer mitochondria. Nephrons atrophy from reduced renal blood flow and inflammatory responses. If obstruction is maintained for a longer period of time, hydronephrosis eventually develops and there is tissue loss with tubular atrophy, interstitial fibrosis, and interstitial inflammation.
In murine experimental models, interstitial fibrosis has been shown to develop within days in the obstructed kidney with increased renal synthesis of extracellular matrix proteins and transforming growth factor-β. The interstitial fibrosis is also mediated by angiotensin II. Proliferation of interstitial fibroblasts with myofibroblast transformation leads to extracellular matrix deposition. Phenotypic transition of renal tubular cells, endothelial cells, and pericytes has been implicated in this process. The compression of medullary and cortical tissue from the renal calyceal distension results in widespread apoptosis and tubular atrophy as early as 3 days postobstruction. The interstitial response to urinary tract obstruction further intensifies the injury, resulting in renal dysfunction. In patients with malignant urinary obstruction, these changes happen over time and may not manifest as clinically evident renal failure for several days to weeks.
In general, patients with benign acute unilateral ureteric obstruction, from a kidney stone for example, present with acute pain typical of a renal colic. Pain from acute ureteric obstruction is caused by pressure or stretch in the lumen. The ureter is supplied by two distinct types of neurons with different activation thresholds. U1 units are low threshold units activated by peristalsis in the absence of mechanical stimulus and U2 units are activated in response to mechanical stimuli, such as increased pressure. U2 units can activate nociceptive afferent input in the nervous system, even in the absence of inflammation. ,
In the setting of malignant obstruction, the clinical presentation depends on the duration and location of the obstruction. The process is more chronic, with extrinsic obstruction developing over a period of days to weeks, unlike in the setting of acute renal colic. This type of insidious urinary obstruction may be unilateral or bilateral. Most patients with malignant urinary obstruction present with vague symptoms of nonspecific lethargy, flank discomfort, or feeling of fullness or being bloated. These symptoms may be accompanied by varying degrees of nausea or anorexia. It has been postulated that distension of the renal pelvis and ureter from obstruction may cause a reflex change in pyloric sphincter pressure contributing to the nausea and vomiting experienced by patients—the so-called renogastric reflex . Urinary tract infection (UTI) may also be a heralding symptom of urinary obstruction. Often, the patient is asymptomatic and the obstruction is identified incidentally on imaging.
Patients with severe renal impairment present with symptoms of renal failure—nausea, vomiting, anorexia, weight loss, edema, and change in mental status. It is important to note that patients with unilateral and even bilateral ureteric obstruction are not necessarily oliguric or anuric. Patients with partial obstruction may continue to have a normal urine output. In some cases polyuria may be noted because of the concentration defect in distal nephron. It is only in cases of complete urinary obstruction that anuria is seen.
Patients with lower urinary obstruction may present with varying degrees of lower urinary tract symptoms including urinary frequency, hesitancy, and sensation of inadequate emptying of the bladder. On physical examination, palpable kidney may be seen only in patients with significant hydronephrosis. Otherwise, examination will show findings of the patients’ primary malignancy, such as the presence of pelvic mass, or ascites from peritoneal carcinomatosis.
Though many patients may have a known diagnosis of cancer before presentation, a careful history and physical examination should be performed. In patients with advanced malignancy, more than one cause for renal dysfunction may be present—prerenal injury in the setting of poor oral intake and postrenal failure from obstruction. Most patients with advanced malignancy have poor oral intake and loss of muscle mass may lead to underestimation of the severity of their renal dysfunction. The postrenal component may only reveal itself after the prerenal component is corrected with volume resuscitation. In some patients, urinary obstruction may be the first presentation of malignancy.
Urine analysis may show microscopic hematuria, pyuria, or mild to moderate proteinuria depending on the duration of obstruction. If leukocyte esterase and nitrites are present, urine cultures should be obtained to rule out UTI.
Early reports of metabolic derangements seen in urinary obstruction detailed the development of hyperkalemia and metabolic acidosis caused by transport defects in the distal nephron, resulting in decreased excretion of K+ and H+.
Varying degrees of renal dysfunction may be seen depending on the duration and severity of urinary obstruction, from asymptomatic urinary obstruction to severe renal failure with life-threatening metabolic derangements. The history, physical examination, and laboratory evaluation will guide further evaluation with imaging.
Imaging modalities for diagnosis of urinary obstruction continue to evolve. Previously, intravenous pyelourography (IVP) was widely used as the gold standard for the diagnosis of urinary obstruction. Functional and anatomic details were provided by this study, but it required both radiation and contrast use. Hence IVP has now been replaced by ultrasonography or computed tomography (CT) in most institutions.
Ultrasonography is the primary screening modality of choice for patients suspected of urinary obstruction. The hallmark of urinary obstruction is dilation of the collecting system of the kidney, which is easily identified on ultrasound as hypoechoic fluid displaces the echogenic parenchyma. In patients with suspected nephrolithiasis, ultrasonography, when performed by a radiologist, resulted in no need for additional CT imaging in most patients in a multicenter comparative effectiveness trial. There were no significant differences between ultrasonography and CT in risk for subsequent serious adverse events, hospitalizations, and return visits to the emergency department.
In the case of patients with suspected malignant obstruction, there are no head to head studies comparing the two modalities. Ultrasonography has several advantages—it is readily available in most institutions, there is no radiation exposure risk, it is noninvasive, no contrast is required, emergency medical professionals can be trained to perform a bedside assessment and, hence, it can be used at point of care. , In addition to detecting hydronephrosis and pyelectasis in patients with urinary obstruction, ultrasonography is also useful in assessing the kidney size, architecture, and changes in corticomedullary differentiation suggestive of chronic kidney disease. The presence of perinephric abscess and pyonephrosis can also be determined on ultrasound. Because patients with malignant obstruction often present with abnormal renal function, the use of intravenous contrast may be contraindicated. Hence ultrasonography offers several advantages as an initial screening test, given its high sensitivity and specificity for detecting hydronephrosis.
Hydronephrosis may be graded on imaging as: grade I—slight blunting of calyceal fornices; grade II/mild—obvious blunting of calyceal fornices and enlargement of calices, but intruding shadows of papillae easily seen; grade III/moderate—rounding of calices with obliteration of papillae; and grade IV/severe—extreme calyceal ballooning. , It is important to note that the grading criteria do not correlate with the degree and acuity of obstruction. A patient with grade I hydronephrosis on imaging may have severe obstruction and renal dysfunction, whereas a patient with grade III hydronephrosis may have no clinical signs or symptoms.
There are also disadvantages to the use of ultrasonography alone in the diagnosis of malignant urinary obstruction. To the inexperienced eye, the presence of extrarenal pelvis or renal cysts may simulate hydronephrosis. , Moreover, in patients with urinary diversion with ileal conduits, hydronephrosis may be a normal finding and not necessarily a sign of urinary obstruction. And importantly, renal ultrasonography may fail to detect early acute urinary obstruction, especially within the first few hours. , , , This is clinically relevant for patients with malignant obstruction, because many of them present with coexisting prerenal injury from dehydration and volume depletion. In some cases, retroperitoneal fibrosis causing early acute urinary obstruction may be missed on ultrasonography. The resistive index (RI) from Doppler assessment has been assessed in the evaluation of urinary obstruction because of the hemodynamic changes that occur in the kidney during obstruction. Although studies have shown a significant difference in the RI values of acutely obstructed versus nonobstructed kidneys, this is not a clinically reliable test because factors other than obstruction may affect the RI.
Noncontrast CT was initially used in patients in whom ultrasonography or IVP did not reveal the cause of urinary obstruction. Now CT is widely used for identifying the cause of urinary obstruction and has replaced IVP for evaluating the upper urinary tract. Older studies have shown CT to be superior to IVP for identifying the cause of obstruction, specifically in patients with cervical cancer. The higher resolution and visualization of the dilated collecting system eliminates the need for contrast. CT can also be performed relatively quickly and can identify all stones, except indinavir stones, which are not visible on CT. In comparison to ultrasonography, CT has the advantage of determining the level of obstruction and the underlying etiology, not just the presence or absence of obstruction.
CT scan can accurately distinguish an intraluminal stone from neoplasm, blood clot, or fungus ball, and characterize secondary signs of obstruction, such as perinephric fat stranding and changes in the renal cortical thickness. , In patients with malignant obstruction, CT has the advantage of also demonstrating extent of metastases, lymphadenopathy, retroperitoneal fibrosis, and other abdominal pathologies, such as bowel obstruction, diverticulitis, presence of ascites, presence of rectovaginal or colovesical fistulae that may affect patient management. If contrast is given, a delayed nephrogram can be appreciated on the obstructed side as well, similar to IVP, but it is not a requirement.
When compared with standard kidney, ureter, and bladder x-ray with ultrasonography and magnetic resonance urogram (MRU), noncontrast CT was superior to the other two modalities in identifying stone-related obstruction, with a sensitivity of 100% and specificity of 98.2%. However, for other causes of obstruction, MRU had higher sensitivity (89.4%) and specificity (95.9%) compared with CT. T2 weighted sequences can visualize a dilated urinary tract without contrast material and selectively depict urine in the dilated renal collecting system and the ureter. Thus MRU enables accurate diagnosis of the level of obstruction and grade of obstruction, but has limited application in the evaluation of stones as MRI cannot detect calcification. MRU may also miss subtle urothelial abnormalities, such as small malignancies caused by relatively poor spatial resolution and visualization of the anatomic details of renal calyces and infundibula. It is not yet widely available, is expensive, and more time consuming compared with the other modalities. In addition, there is a risk of nephrogenic systemic fibrosis when gadolinium is used in patients with advanced renal impairment. Glomerular filtration rate may be overestimated in patients with cancer due to loss of muscle mass. Because of these reasons, MRU is not the first-line imaging tool in the evaluation of patients with suspected malignant urinary obstruction.
Radionuclide studies can assess the differential function of each kidney. In acute obstruction, the use of technetium 99m pentetate or MAG-3 allows the evaluation of uptake, transit, and elimination of the radionuclide from the kidney. In the setting of unilateral obstruction, parenchymal activity rises at a slower rate and persists for a longer time than the nonobstructed kidney. Disadvantages of radionuclide imaging include lack of precise anatomic detail and delineation of the obstruction and inability to identify the cause. Radionuclide imaging can differentiate a dilated nonobstructed system from a partially obstructed system, with the use of diuretic renogram, by using a dose of furosemide after the radionuclide study. In the former, the collecting system activity will washout within 10 minutes after furosemide, whereas in the latter, there is no response or a slower response to furosemide. ,
The choice of imaging modality will ultimately depend on the availability at a given institution and cost. Ultrasonography can be used as the first screening imaging modality in patients with acute kidney injury and suspected obstructive nephropathy with or without a cancer diagnosis. CT without contrast can be used as the initial imaging modality in patients with a known diagnosis of cancer who present with flank pain or other abdominal symptoms, and if the extent of metastases will affect their clinical course. In most patients with malignant urinary obstruction, CT scan will help establish the extent of their disease and presence of lymphadenopathy, pelvic disease burden, and peritoneal carcinomatosis. This will help the clinician in managing the urinary obstruction and choosing the right intervention for the patient.
The initial management of a patient with malignant urinary obstruction should include correction of hypovolemia. A Foley catheter should be placed, and accurate charting of patient’s intake and urine output should be maintained. Consultation with Urology is mandatory. Ideally the Onconephrology team should use a multidisciplinary approach with expertise from Nephrology, Urology, and Interventional Radiology. UTI should be identified and treated. Pain from obstruction should be managed, preferably without the use of NSAIDs, because of their effect on renal perfusion. Patients with advanced malignancies are especially prone to renal failure because of high risk of hypovolemia, and renal vasoconstriction from NSAIDs may tip them over the edge. The use of corticosteroids has been proposed to reduce the risk of edema around obstructing tumors and purported antitumor effect. , But there is currently no strong evidence for the use of corticosteroids for urinary obstruction in cancer patients.
The decision to proceed to a surgical intervention for a patient with obstructive nephropathy from cancer is a complex one. The median survival of adult patients with malignant obstruction from the time of intervention is reported to be around 96 days (2–1283 days) with 1-, 6-, and 12-month survival rates at 78%, 30%, and 12%, respectively. The procedural risks are low compared with the benefit of avoiding dialysis, even in patients with advanced malignancy. Despite this, there may be a high price to be paid for urinary diversion procedures performed in patients with advanced malignancies. There is a high failure rate associated with ureteric stenting and concern for increased risk of infections and readmissions with percutaneous nephrostomy (PCN) placement. , Hence the risks and benefits must be considered carefully before recommending urinary diversion for patients with limited treatment options for their primary malignancy.
Once the decision is made, taking into consideration the patient’s quality of life, procedural risks, and expected life span, the next step is to choose the best modality to achieve this. Although there have been several retrospective studies published that compare different modalities, few studies have compared modalities in a head to head fashion. Prospective studies and randomized controlled trials are also lacking as most patients’ projected life span is short.
Urinary diversion or decompression procedures are generally defined by the approach into the urinary tract—retrograde or antegrade. Retrograde decompression of urinary obstruction is performed by way of cystoscopic ureteric stent placement. The antegrade approach is via PCN placement and insertion of an antegrade ureteric stent at a later stage. Retrograde stenting is usually done under anesthesia, but can be safely performed in patients with deranged coagulation and thrombocytopenia. Nephrostomy placement can be done under local anesthetic but is a high-risk procedure in patients with coagulopathy and thrombocytopenia.
Retrograde ureteric stenting
Fluoroscopic guidance is usually used to confirm the stent placement by retrograde approach. The first material used for ureteric stents was silicone in the 1970s; other materials, such as polyethylene, polyurethane, and mesh have been used subsequently. , There is a high failure rate for urinary decompression with retrograde ureteric stenting, because the lower urinary tract is anatomically challenging to maneuver, given the tortuosity of the obstructed dilated system and obscured ureteric orifices. , This is especially true in the setting of malignancy; the incidence of stent failure is higher in cases of extrinsic compression compared with intrinsic obstruction, such as stone disease and ureteric strictures. Ureters obstructed by vesical/prostatic/cervical cancers are especially difficult to access. Even when access is obtained into the lower ureter, it may be difficult to advance the guide wire successfully and negotiate the site of rigid lower ureteric obstruction. This imparts a mechanical disadvantage from the lower end, which is much less of a problem during antegrade stenting because of the ‘funnel’ effect of the dilated proximal ureter. Inability to cannulate the ureteric orifices resulting from trigonal distortion, failure to negotiate the lower ureteric segment, and nonvisualization of the ureteric orifice caused by postoperative scarring, are reported as reasons for retrograde stent failure.
In the context of malignant ureteric obstruction, a wide range of retrograde stent failure rates have been reported from 16% to 58%. A stent failure rate of 35.7% was reported in a study of 157 patients with extrinsic malignant compression. Here, the commonest cancer diagnosis was ovarian, lymphoma, and cervical cancer. In this study, the type of cancer did not predict stent failure and need for PCN. Significant need for PCN was noted among patients who had rectal cancer (seven of 12, 58.3%), sarcoma (five of nine, 55.6%), colon cancer (six of 13, 46.2%), and cervical cancer (six of 16, 37.5%). Invasion into the bladder (55.9% patients), when noted on cystoscopy, significantly predicted the need for PCN ( p =.008). In another retrospective study of more than 100 patients, the reported stent failure rate was 41%. Here, cancer diagnosis, baseline creatinine over 1.3 mg/dL, and post-stent systemic therapy for metastatic cancer were identified as predictors of stent failure. The diagnosis of cancer, not specific cancer type, was again identified as a strong predictor for stent failure. There was no correlation between the size of the stent and failure rate.
More recent studies continue to show similar stent failure rates, but a trend toward cancer type also influencing stent success rate—in a retrospective study of 53 patients, the stent failure rate was 34%; primary gastrointestinal cancer, poor performance status and severe preoperative hydronephrosis were independent predictors of stent failure. Another notable difference in recent reports is systemic cancer therapy being shown to be a favorable prognostic factor for survival, after retrograde stenting for malignant ureteric obstruction. In a recent study analyzing prognostic factors for overall survival and stent failure in patients treated with retrograde stenting for malignant ureteric obstruction, serum creatinine before stent placement exceeding 1.2 mg/dL and lack of cancer therapy after stent placement were significant unfavorable prognostic factors, whereas gynecologic cancer was a significant favorable predictor of stent-failure free survival. Median survival after stenting in this study was 228 days. These differences may reflect the advances made in cancer therapy in recent times and differences in patient selection and techniques for surgical management of urinary obstruction.
One of the reasons for drainage failure of ureteric stents in advanced pelvic malignancy is postulated to be the lack of long-segment ureteric peristalsis. Stent encrustation with urine and urothelium also develops over time, limiting the duration of stent use, necessitating stent exchanges every 6 to 12 months. Even though overt encrustation may not be present, the lumen of the stents are found to be blocked at the time of routine replacement, hence requiring routine exchange. Complication rates reported after ureteric stenting may be as high as 53%. Early stent-related complications include iatrogenic injury, stent migration, and patient discomfort, whereas late complications include infection, difficulties with stent exchange, hardware malfunction, infection, and stent encrustation. Stent fragmentation and migration were noted in 8% to 10% of patients. Fever was reported in 31%, and 16% of patients reported flank pain on the stented side on voiding, almost half of those requiring stent removal because of severe pain on voiding. The cause of flank pain during voiding may be secondary to vesicorenal reflux through the stent during periods of increased intravesical pressure. In prospective quality of life questionnaire studies, the patient reported outcomes are worse. Eighty percent of patients experienced bothersome urinary symptoms and stent associated pain. Urinary incontinence and sexual dysfunction also affect patient quality of life.
PCNs are tubes placed under sedation and ultrasound or CT guidance by entering the pelvic calyx and using Seldinger technique to dilate the tract. Patients are generally in prone position during procedure. Open nephrostomy tube placements are not favored anymore because of high rate of major complications. PCNs have a high procedural success rate of 90% to 92%. This is offset by the poor quality of life of patients after nephrostomy placement and high rate of complications. Complications following PCN placement include bleeding, tube dislodgement (10%–14% rate), and UTI. , The UTI rate ranges from 25% to 65%. In a retrospective review of 200 patients with cancer, 19% had PCN-related pyelonephritis and 7.5% had asymptomatic bacteriuria. Eighty-nine percent of infections were with the primary nephrostomy tube, whereas the rest happened after nephrostomy tube exchange. Pyelonephritis developed within the first month in 10% of patients and within 3 months in 20% of patients. Prior UTI and neutropenia were significant risk factors for pyelonephritis. History of diabetes mellitus, chronic kidney disease, and presence of ureteral stents was not associated with increased risk of pyelonephritis. Of note, 99% of patients in this study did receive prophylactic antibiotics before PCN placement. Gram-negative organisms, such as Escherichia coli , Klebsiella, and Pseudomonas caused 40% of UTIs. Enterococcus was found in 30%, coagulase-negative Staphylococci in 17%, and Staphylococcus aureus in 9% of cases. Fungal infections were 12% of the total. Other reported complications from PCN include skin excoriation, urine leak, perirenal abscess formation, perirenal hemorrhage, and high readmission rate related to these events.
A recent study examined the long-term management of malignant urinary obstruction with PCN, including estimation of exchange frequency and the financial impact. The simulation model used suggests that the optimal routine exchange interval for PCN exchanges in patients with malignant urinary obstruction is approximately 60 days.
Comparison of retrograde and antegrade stenting
Based on head to head comparisons of both procedures, albeit retrospectively, some experts recommend a two-staged approach in the context of pelvic malignancies. In a study of 65 patients with predominantly urologic cancer, 24 of 65 patients had an initial attempt at retrograde ureteric stenting as a first-line method of decompression. PCN followed by antegrade ureteric stenting at a later stage within a week of nephrostomy insertion was performed in 41 of 65 patients. If the initial nephrostomy was placed in the lower calyx, a mid-calyceal puncture was performed to facilitate antegrade insertion of stent. Success rate was 21% for endoscopic retrograde stenting, whereas two-stage antegrade ureteric stenting had a success rate of 98% with reported minimal morbidity. Failure of antegrade stenting in one case was caused by an impassable stricture of the lower ureter caused by prostate carcinoma. The high failure rate of retrograde stenting was attributed to inability to cannulate the ureteric orifices, caused by trigonal distortion and failure to negotiate the lower ureteric segment in most patients. Thus PCN, first followed by antegrade stenting approach, allows management to be individualized for the patient and highlights the need for a multidisciplinary approach to managing urinary obstruction in the cancer patient.
Some investigators have attempted to predict survival after palliative urinary diversion for malignant ureteric obstruction. In a study of 140 patients undergoing PCN for palliative urinary diversion, in the setting of advanced malignant disease deemed to be incurable, median overall survival was 96 days (2–1283 days) with 1-, 6-, and 12-month survival rates at 78%, 30%, and 12%, respectively. On multivariate analysis, events related to malignant dissemination, degree of hydronephrosis (grade I or II), and serum albumin before nephrostomy (3 g/dL or less) were associated with shorter survival. Patients were divided into three risk groups based on the aforementioned three factors: favorable –0 risk factors, intermediate one risk factor, and poor–two or three risk factors. There were significant differences in survival between the risk groups. The 6-month survival rates for the favorable, intermediate, and poor risk groups were 69%, 24%, and 2%, respectively ( p < .0001). Complications after PCN, in this study, included infection (13%), dislodgement (19%), major hemorrhage (2%), and minor hemorrhage (5%). Malignant diagnoses in this study were as follows: gastric, 21%; colorectal, 24%; uterine/cervical, 21%; ovarian, 4%; urothelial, 9%; and breast, 6%. Given the high mortality rate in the first year postprocedure, initial PCN placement with conversion to internal stenting is favored in some patients. Retrograde ureteral stent placement may be attempted first in patients who do not have advanced malignancy, with high tumor burden in the abdomen and pelvis. It is important to identify the patients who would benefit the most from upfront PCN.
Few studies have examined long-term outcomes in patients presenting with bilateral ureteral obstruction. In a retrospective study of 87 patients who underwent bilateral ureteral stenting, most common cancer origin was cervical followed by stomach cancer. The mean creatinine at the time of bilateral stenting was 3.3 mg/dL, which did fall to 1.6 mg/dL at 6 months postprocedure, but at 3 years poststenting, there was further deterioration in renal function, with mean creatinine at 2.3 mg/dL. A percentage of patients (18.3%) developed chronic kidney disease stage 4 or 5 at 6 months and 57.2% at 3 years. Despite the development of chronic kidney disease, ureteral stenting in bilateral malignant ureteral obstruction may be worth pursuing, because dialysis may be avoided for several months to years. About 14% of patients underwent conversion to PCN. Among these patients ( n =12), eight patients had early conversion to PCN to the dominant kidney or both kidneys less than 1 year after the initial stenting—they showed a trend towards better renal function outcomes compared with patients who had late or no conversion to PCN ( p =.06). Cancer status was an independent prognostic factor for stent failure. Age older than 55 years, diabetes, and estimated GFR < 60 ml/min before obstructive symptoms or signs developed were significant predictive factors for chronic kidney disease stage 4 or more.
In addition to retrograde ureteric stenting and PCN placement, other surgical approaches have been studied in patients who did not have cancer. These include combined antegrade and retrograde approaches (so-called rendez-vous procedure ), the use of two parallel stents and other methods of ureteric diversion including nephrovesical subcutaneous stent and cutaneous ureterostomy placement. , , But these methods have not been studied in patients with cancer and hence cannot be recommended at present.
In the future, antibiotic coated nephrostomy tubes may be developed to reduce the risk of PCN related pyelonephritis. In addition, future management may include urinary proteomes, which are being studied to help identify patients at risk for renal dysfunction from mild obstruction. These innovative approaches may be more relevant in the pediatric population with congenital causes of obstruction rather than adult patients with malignant obstruction.
Palliative urinary diversion specifically in the setting of advanced malignancy is a major clinical challenge in onconephrology. The risk and complications from urinary diversion procedures must be balanced against potential benefits for renal function and patient survival. Shekarriz et al. reported outcomes of palliative urinary diversion in the treatment of advanced malignancies in 103 patients treated with either stent or PCN; 92 of 103 patients had bilateral and 11 had unilateral obstruction. Modified Karnofsky performance scale was used for physical performance, range 0 to 4. The median survival reported in this series was 112 days; 15% patients never left the hospital. Fifty-one percent required secondary PCN after initial retrograde stenting, 68.4% had complications, 86% had cancer-related symptoms, despite the urinary diversion. Average survival was 5 months, 50% of which was spent in hospital. Thus it is important to include the patient’s quality of life and goals of care in the decision making, before proceeding to urinary diversion for palliation. Pain management and palliation of other symptoms such as nausea should be prioritized while undertaking measures to relieve the obstruction.
In conclusion, the approach to obstructive nephropathy in cancer patients should be individualized taking into consideration patient wishes and predicted survival from the underlying malignancy. Overall survival from the underlying malignancy.