Nephrolithiasis and CKD

CKD is a major public health problem and affects 13% of the adult population. CKD can be classified into five categories according to their estimated glomerular filtration rate (eGFR) (in milliliters/minute/1.73 m ² ) with or without renal damage for greater than or equal to 3 months. The categories, or stages, have an eGFR of (1) ≥90, (2) 60–89, (3) 30–59, (4) 15–29, and (5) <15 (ml/min/1.73 m ² ).

With more than $23 billion spent by Medicare in 2009 in the United States for patients on any type of renal replacement therapy or renal transplantation, significant efforts have been made to prevent and treat possible causes of CKD. According to the United States Renal Data System (USRDS), the prevalence of CKD has remained relatively stable over the last decade, despite the use of different formulas to calculate kidney function. In contrast, data extracted from Medicare claims for CKD demonstrate a clear increase from 3.3% in 1998 to 8.5% in 2009. Similarly, the 2005 to 2008 analysis of the National Health and Nutrition Examination Survey (NHANES) reports that the prevalence of CKD has increased to 15.1% and an incidence rate for end stage renal disease (ESRD) of 355 new cases per million population in 2009.

Although nephrolithiasis is rarely identified as a prominent risk factor for either CKD or ESRD, there are many studies that associate stone disease with varying degrees of renal insufficiency. The cause of renal insufficiency in patients with nephrolithiasis is multifactorial and includes renal obstruction, recurrent urinary tract infection (UTI), repeated surgical interventions, and coexisting medical disease. Early recognition of CKD and management of its underlying factors could help prevent several adverse effects and decrease the risk of morbidity and mortality associated with this disease, including long-term renal replacement therapy and transplantation, as well as social and psychological distress. Patients with CKD represent 0.8% to 17.5% of those presenting with urinary stone disease. CKD remains an uncommon event among stone formers and nephrolithiasis-related CKD accounts for only a small percentage (usually lower than 3%) of patients for whom renal replacement therapy is necessary.

It has been assumed that CKD will develop in 3 of 100,000 stone formers annually. In a cohort of 171 subjects with severe idiopathic calcium stone disease, Marangella and colleagues reported that 18% had mild renal insufficiency, with a mean GFR of 67 mL/min/1.73 m ² at referral; but there was no significant decline during a mean follow-up of nearly 3.5 years. Similarly, of the 3266 subjects with nephrolithiasis reported by Worcester and colleagues, CKD did not develop in any of them and there was no significant decline in creatinine clearances on long-term follow-up. On the other hand, the same investigators reported that renal function decreased with age in stone formers at a higher pace than in non–stone formers.

In the general population, lower eGFR has been observed in overweight or obese patients and also in stone formers with hypertension (HTN). Gillen and colleagues used NHANES III of the US population to compare eGFR between the 6% who reported a history of kidney stones and the rest who did not report a history of kidney stones. Among individuals who were overweight or obese, a history of kidney stones was associated with an eGFR that was 3.4 mL/min/1.73 m ² lower after multivariate adjustment. A similar association in normal-weight individuals was not evident. Vupputuri and colleagues demonstrated that subjects who had CKD (identified by diagnostic codes) and elevated serum creatinine (SCr) levels were 1.9 times more likely to report a history of kidney stones on telephone interview when compared with matched community control subjects. Interestingly, the association was strongest in individuals without HTN and in individuals who were identified by CKD diagnostic codes for interstitial nephritis or diabetic nephropathy. Because cases were identified via hospital record review, they may have had more undiagnosed comorbidities than did control subjects. Population-based prospective cohort studies with active follow-up to determine the risk for CKD among stone formers are lacking. In a population-based historical cohort evaluation, Rule and colleagues compared subjects with diagnostic codes and/or elevated SCr levels (or eGFR <60 mL/min/1.73 m ² ) sustained for at least 3 months with a matched control group. With a mean follow-up of 8.6 years, the risk for clinical diagnosis of CKD was 50% to 67% higher, risk for a sustained elevated SCr was 26% to 46% higher, and risk for a sustained reduced eGFR was 22% to 42% higher in the stone group. These increased risks were independent of any comorbidities associated with CKD.

Although kidney stones may be associated with future CKD, paradoxically, there is reason to believe that CKD is protective against formation of kidney stones. As GFR declines, it is followed by a fall in urine calcium excretion, an important risk factor for stone formation. Indeed, evidence suggests that stone recurrence rates may be lower in stone formers with a reduced GFR. Ironically, if kidney stones lead to CKD, the risk for stone recurrence may decrease, leading to an underrecognition of the contribution of kidney stones to the development of ESRD. It is important to distinguish stone formers in whom stones are the primary cause of ESRD from those in whom they are only a contributing risk factor. Jungers and colleagues specifically investigated ESRD cases that had been attributed to kidney stones by reviewing the case histories of 1391 consecutive subjects with ESRD in France. Forty-five (3.2%) had ESRD attributed primarily or exclusively to kidney stones, with a significant proportion being attributed to struvite stones (42%). Tosetto and colleagues identified a history of kidney stones in only 3.2% of 1901 subjects who had ESRD and were on hemodialysis, two-thirds of whom (2.1%) had ESRD attributed to the kidney stones. The USRDS reports kidney stones to be the primary cause of ESRD in only 0.2% (908 or 546,878) of all-incident subjects from 2004 to 2008. Nevertheless, their findings took in consideration the ESRD Medical Evidence form (CMS 2728) and not a comprehensive chart review.

In a case-control study, Stankus and colleagues surveyed 300 black subjects with hemodialysis for a history of kidney stones and compared findings with the 5341 black individuals who participated in NHANES III. The likelihood of self-reported past kidney stones was higher for patients with ESRD than for the population control subjects (8% vs 3%, respectively). Of the 25 subjects with ESRD and a history of kidney stones, only five had a stone episode within 5 years of starting dialysis and only two had ESRD that was primarily attributed to the stone disease. Hippisley-Cox and Coupland reported an increased risk for ESRD with women (hazard ratio of 2.1), but not men, stone formers in a cohort study. In contrast, though the Olmsted County study demonstrated significant links between CKD and stone disease, there was no statistical impact on the development of ESRD or mortality.

Nephrolithiasis and CKD

CKD is a major public health problem and affects 13% of the adult population. CKD can be classified into five categories according to their estimated glomerular filtration rate (eGFR) (in milliliters/minute/1.73 m ² ) with or without renal damage for greater than or equal to 3 months. The categories, or stages, have an eGFR of (1) ≥90, (2) 60–89, (3) 30–59, (4) 15–29, and (5) <15 (ml/min/1.73 m ² ).

With more than $23 billion spent by Medicare in 2009 in the United States for patients on any type of renal replacement therapy or renal transplantation, significant efforts have been made to prevent and treat possible causes of CKD. According to the United States Renal Data System (USRDS), the prevalence of CKD has remained relatively stable over the last decade, despite the use of different formulas to calculate kidney function. In contrast, data extracted from Medicare claims for CKD demonstrate a clear increase from 3.3% in 1998 to 8.5% in 2009. Similarly, the 2005 to 2008 analysis of the National Health and Nutrition Examination Survey (NHANES) reports that the prevalence of CKD has increased to 15.1% and an incidence rate for end stage renal disease (ESRD) of 355 new cases per million population in 2009.

Although nephrolithiasis is rarely identified as a prominent risk factor for either CKD or ESRD, there are many studies that associate stone disease with varying degrees of renal insufficiency. The cause of renal insufficiency in patients with nephrolithiasis is multifactorial and includes renal obstruction, recurrent urinary tract infection (UTI), repeated surgical interventions, and coexisting medical disease. Early recognition of CKD and management of its underlying factors could help prevent several adverse effects and decrease the risk of morbidity and mortality associated with this disease, including long-term renal replacement therapy and transplantation, as well as social and psychological distress. Patients with CKD represent 0.8% to 17.5% of those presenting with urinary stone disease. CKD remains an uncommon event among stone formers and nephrolithiasis-related CKD accounts for only a small percentage (usually lower than 3%) of patients for whom renal replacement therapy is necessary.

It has been assumed that CKD will develop in 3 of 100,000 stone formers annually. In a cohort of 171 subjects with severe idiopathic calcium stone disease, Marangella and colleagues reported that 18% had mild renal insufficiency, with a mean GFR of 67 mL/min/1.73 m ² at referral; but there was no significant decline during a mean follow-up of nearly 3.5 years. Similarly, of the 3266 subjects with nephrolithiasis reported by Worcester and colleagues, CKD did not develop in any of them and there was no significant decline in creatinine clearances on long-term follow-up. On the other hand, the same investigators reported that renal function decreased with age in stone formers at a higher pace than in non–stone formers.

In the general population, lower eGFR has been observed in overweight or obese patients and also in stone formers with hypertension (HTN). Gillen and colleagues used NHANES III of the US population to compare eGFR between the 6% who reported a history of kidney stones and the rest who did not report a history of kidney stones. Among individuals who were overweight or obese, a history of kidney stones was associated with an eGFR that was 3.4 mL/min/1.73 m ² lower after multivariate adjustment. A similar association in normal-weight individuals was not evident. Vupputuri and colleagues demonstrated that subjects who had CKD (identified by diagnostic codes) and elevated serum creatinine (SCr) levels were 1.9 times more likely to report a history of kidney stones on telephone interview when compared with matched community control subjects. Interestingly, the association was strongest in individuals without HTN and in individuals who were identified by CKD diagnostic codes for interstitial nephritis or diabetic nephropathy. Because cases were identified via hospital record review, they may have had more undiagnosed comorbidities than did control subjects. Population-based prospective cohort studies with active follow-up to determine the risk for CKD among stone formers are lacking. In a population-based historical cohort evaluation, Rule and colleagues compared subjects with diagnostic codes and/or elevated SCr levels (or eGFR <60 mL/min/1.73 m ² ) sustained for at least 3 months with a matched control group. With a mean follow-up of 8.6 years, the risk for clinical diagnosis of CKD was 50% to 67% higher, risk for a sustained elevated SCr was 26% to 46% higher, and risk for a sustained reduced eGFR was 22% to 42% higher in the stone group. These increased risks were independent of any comorbidities associated with CKD.

Although kidney stones may be associated with future CKD, paradoxically, there is reason to believe that CKD is protective against formation of kidney stones. As GFR declines, it is followed by a fall in urine calcium excretion, an important risk factor for stone formation. Indeed, evidence suggests that stone recurrence rates may be lower in stone formers with a reduced GFR. Ironically, if kidney stones lead to CKD, the risk for stone recurrence may decrease, leading to an underrecognition of the contribution of kidney stones to the development of ESRD. It is important to distinguish stone formers in whom stones are the primary cause of ESRD from those in whom they are only a contributing risk factor. Jungers and colleagues specifically investigated ESRD cases that had been attributed to kidney stones by reviewing the case histories of 1391 consecutive subjects with ESRD in France. Forty-five (3.2%) had ESRD attributed primarily or exclusively to kidney stones, with a significant proportion being attributed to struvite stones (42%). Tosetto and colleagues identified a history of kidney stones in only 3.2% of 1901 subjects who had ESRD and were on hemodialysis, two-thirds of whom (2.1%) had ESRD attributed to the kidney stones. The USRDS reports kidney stones to be the primary cause of ESRD in only 0.2% (908 or 546,878) of all-incident subjects from 2004 to 2008. Nevertheless, their findings took in consideration the ESRD Medical Evidence form (CMS 2728) and not a comprehensive chart review.

In a case-control study, Stankus and colleagues surveyed 300 black subjects with hemodialysis for a history of kidney stones and compared findings with the 5341 black individuals who participated in NHANES III. The likelihood of self-reported past kidney stones was higher for patients with ESRD than for the population control subjects (8% vs 3%, respectively). Of the 25 subjects with ESRD and a history of kidney stones, only five had a stone episode within 5 years of starting dialysis and only two had ESRD that was primarily attributed to the stone disease. Hippisley-Cox and Coupland reported an increased risk for ESRD with women (hazard ratio of 2.1), but not men, stone formers in a cohort study. In contrast, though the Olmsted County study demonstrated significant links between CKD and stone disease, there was no statistical impact on the development of ESRD or mortality.

SWL

SWL is widely used to treat proximal ureteral calculi less than 1 cm and renal stones less than 2 cm in diameter. In animal models, SWL has been shown to induce parenchymal injury, which increases with the number of shocks, level of energy, and also with smaller kidneys. Furthermore, SWL is associated with an acute reduction in GFR and renal blood flow due to vasoconstriction. Evaluation of “shocked” kidneys with magnetic resonance imaging, contrast urogram, and nuclear scintigraphy revealed that 74% of patients had abnormal findings after SWL consistent with a renal contusion. However, a long-term effect of SWL on kidney function has not been definitively demonstrated. Eassa and colleagues found no change in eGFR or in the relative differential renal function of the treated kidney using nuclear scintigraphy at approximately 4 years of follow-up. Even longer follow-up studies suggest that the risk for an elevated SCr level is not increased with SWL compared with PCNL or conservative management. HTN has been described as a possible long-term complication of SWL in some but not all trials. A recent population-based study did not find an increased risk, suggesting that this complication rarely affects stone formers following SWL. Pretreatment with low-energy shockwaves followed by a ramp-up protocol in energy seems to be protective against renal injury. Recently, el-Assmy and colleagues retrospectively evaluated the long-term effects of SWL in subjects with solitary kidneys with a mean follow-up of 3.8 years and found no significant difference in SCr, systolic and diastolic blood pressure, new onset HTN, calculated GFR, and kidney morphology before and after treatment.

SWL success is dependent of several factors. Stone-free rate (SFR) is influenced by stone size, location, symptom duration previous to the procedure, presence of ureteral stent, gender, stone density (Hounsfield Units), skin-to-stone distance, and stone composition. The influence of renal function on success rates after SWL, however, is debatable. Lee and colleagues reported a significantly lower SFR of 57% for subjects with preoperative SCr between 2.0 and 2.9 mg/dL compared with 66% if SCr was less than 2.0 mg/dL ( P <.05). Hung and colleagues reported that subjects with proximal ureteral calculi had SFR of only 50% if they had underlying CKD, compared with 93% for subjects with an eGFR greater than 60 mL/min/1.73 m ² . On multivariate logistic regression analysis, factors negatively affecting SFR were gender (females), eGFR less than 60 mL/min/1.73 m ² , and stone width greater than 7 mm. Conversely, other studies have failed to find an association between SWL outcomes and preoperative impaired renal function.

PCNL

There have been some attempts to identify predictors of prognosis and outcomes for patients with CKD who have upper urinary tract calculi. Paryani and Ather retrospectively reviewed 500 subjects with urolithiasis (40% with complete or partial staghorn calculi) and reported that in the 12% with baseline renal insufficiency, most had improvements in function independent of relief of obstruction. Singh and colleagues prospectively followed 70 subjects with CKD (mean preoperative SCr of 4.76 mg/dL) who underwent PCNL for staghorn or calyceal stones. They reported an average decrease in SCr of 1.53 mg/dL (32%) and an average functional improvement by renal dynamic scans of 20% 9 months postoperative.

Agrawal and colleagues published their experience with PCNL in 78 subjects with calculous nephropathy and advanced renal failure. An improvement in renal function was seen in most (82%) subjects, whereas in 11 subjects it remained unaltered or deteriorated. Nevertheless, in their study baseline renal function evaluation was performed at presentation, instead of after placing a percutaneous nephrostomy tube or treating any UTI to rule out an overlying component of acute renal failure. They reported parenchymal thickness greater than or equal to 7 mm, clear urine in the collecting system, absence of urosepsis, and recent onset azotemia were predictors of renal recovery. Kukreja and colleagues reviewed the impact of PCNL on renal function in 84 CKD subjects with renal stones. Subjects with acute rise in creatinine due to obstruction or infection were excluded from the study, but SCr, instead of eGFR, was used to evaluate renal function. They reported an overall improvement in renal function in 33 subjects (39%), stable function in 24 (29%), and decreased function in 27 (32%) with a mean follow-up of 2.2 years. Factors predicting deterioration in renal function were proteinuria greater than 300 mg/d, atrophic cortex less than 5 mm, recurrent UTI, stone burden greater than 1500 mm ² , time elapsed after PCNL surgery less than 15 years, and finally pediatric age group. In a retrospective study by Canes and colleagues, the impact of PCNL on renal function was evaluated in 81 subjects with a solitary kidney. Mean eGFR increased from 44.9 mL/min/1.73 m ² preoperatively to 51.5 mL/min/1.73 m ² 1 year after intervention. In another large series, Bilen and colleagues evaluated 185 subjects with eGFR less than 60 mL/min/1.73 m ² undergoing PCNL and found the mean preoperative eGFR significantly increased from 42.4 to 48.4 mL/min/1.73 m ² at three months follow-up. None of the subjects required dialysis during that relatively shorter follow-up. They also found that nearly all subjects with stage 5 CKD had some benefit from surgery, whereas half of stage 4 subjects and only 25% of stage 3 subjects improved. Renal function improvement was the greatest in stage 5 and the least in stage 2 subjects. They hypothesized that the impact of the calculi itself in a severely compromised kidney is greater than the impact of PCNL and the opposite is probably true for moderately affected kidneys in which the harm of PCNL may be more significant, particularly if associated with UTI.

Kurien and colleagues studied 91 adult subjects with SCr greater than 1.5 mg/dL who underwent PCNL. Most subjects had stage 3 or 4 CKD and most showed improvement or stabilization in renal function after PCNL. Postoperative complications and peak eGFR were the main factors predicting deterioration of kidney function during follow-up. Similarly, Akman and colleagues followed 177 subjects who underwent PCNL and had preoperative GFR less than 60 mL/min/1.73 m ² for at least 1 year and found a significant improvement in eGFR after a mean 43-month follow-up.

Another important factor that has been proposed to affect renal function in patients with CKD is the number of tracts. Animal studies have shown that PCNL tracts were associated with minimal scar tissue and no significant morphologic or functional alterations. Handa and colleagues also evaluated the impact of multiple-tract PCNL in pigs and found no significant difference in renal function loss compared with single-tract PCNL. Human studies revealed that renal damage from nephrostomy tracts is minimal based on nuclear renography and has no effect on systemic renal function. Akman and colleagues series compared 142 subjects treated with one access to 35 subjects with multiple-tract PCNL and found similar renal function among them, even when considering CKD subjects.

SFR after PCNL for patients with CKD depend on imaging modality to define it and timing of the examination itself. Bilen and colleagues reported complete stone-free status on postoperative radiograph of 81.1% in subjects with CKD undergoing PCNL. Clinically significant residual stones (>5 mm) were present in 21 renal units (10.7%) and insignificant residual fragments were seen in the remaining 16 (8.1%). Overall, 5 renal units underwent adjuvant SWL, 10 had second-look PCNL, and 5 underwent ureteroscopy. Kurien and colleagues defined complete clearance as nonvisualization of residual fragments on radiograph and ultrasonography at 1 month after PCNL. Stone-free status was achieved in 98 renal units (83.7%) with a mean of 1.3 stages per renal unit. SWL as an auxiliary procedure was required for clearance of residual stones in 2.5%. Akman and colleagues reported SFR at postoperative month 3 of 80.2% (142 out of 177). Definition was based on radiograph, ultrasound, and/or CT. Stone recurred during long-term follow-up in 25.3% of these subjects.

PCNL remains the primary modality for treating complex stones in patients with CKD. However, it is not without complications and the most common are hypothermia, bleeding, metabolic acidosis, serum electrolytes disturbances, urosepsis, and rarely death. Overall PCNL complication rate is 13% to 35% and bleeding remains the leading problem. Anemia and underlying platelet dysfunction in patients with CKD may play an important role in the high rate of transfusion. Bilen and colleagues reported that 38% of CKD subjects had anemia preoperatively and that blood transfusion rates were as high as 36%. Kurien and colleagues and Akman and colleagues had a lower, but still significant, transfusion rate of 20.5% and 9.6%, respectively. Efforts to reduce hemorrhagic complications include ultrasound-guided access, balloon-tract dilatation, optimization of operating time, and staging procedures in cases of large stone burden.

Despite prophylactic antibiotics, urosepsis remains a major concern, even in patients with CKD who have sterile preoperative urine cultures. Because of the inhibition of cell-mediated immunity and humoral defense mechanisms, sepsis may easily develop in patients with CKD. Agrawal and colleagues reported septic complications in eight of the 78 subjects, out of whom three died (3.8%) despite intensive care. Bilen and colleagues reported septic shock in three subjects, regardless of appropriate antibiotic treatment. Only one subject survived and the overall mortality rate was 1.1%. Kurien and colleagues reported UTI, defined as culture proven or prolonged febrile episodes, in nine subjects (7.7%) of their series. There were two postoperative deaths and one related septic shock. In the series of Akman and colleagues, urosepsis was detected in five subjects (2.8%) who were successfully treated with intravenous broad-spectrum antibiotics. No deaths occurred. Steps used in attempts to avoid postoperative infection include adequate preoperative collecting system drainage and preemptive administration of broad-spectrum antibiotics. During the procedure, the use of a large caliber Amplatz sheath to maintain low intrarenal pressure may decrease the incidence of bacteremia in patients with infected stones. Importantly, complete stone clearance is essential to remove all foci of infection. Furthermore, adequate postoperative drainage through a nephrostomy tube or a ureteral stent may reduce bacterial load in the pelvicalyceal system, decreasing the chance of bacteremia. Efforts should also be spent to reduce hospital-acquired infections.