Trends in the Utilization of Percutaneous and Open Nephrolithotomy in the Treatment of Renal Calculi

Abstract

Purpose:

To evaluate trends in the use of percutaneous nephrolithotomy (PCNL) and nephrolithotomy (NL) in patients with renal pelvis calculi.

Materials and Methods:

An analysis of the 5% Medicare Public Use Files (years 2001, 2004, 2007, and 2010) was performed to assess changes in the use of PCNL and NL over a 10-year period. Patients were identified using the International Classification of Diseases-9 (cm) and Current Procedure Terminology codes. Statistical analyses, including the Fisher and chi-square tests and multivariate regression analyses, were performed using SAS 9.3 (SAS Institute Inc, Cary, NC) and SPSS v20 (IBM Corp., Armonk, NY).

Results:

A total of 26,100 patients underwent either PCNL or NL. Use of PCNL and NL decreased from 3.1% to 2.5% in patients with a diagnosis of stones (P<0.0001). Women (odds ration [OR]=1.19, P=0.003) were more likely to undergo surgery. Patients aged ≥65 years were less likely to be treated (OR=0.65-0.71, P<0.05). Patients treated after 2004 were less likely to undergo surgery (OR=0.77–0.84, P<0.05). The use of PCNL exceeded NL at a stable 10:1 ratio.

Conclusions:

The use of PCNL and NL for treatment of patients with stone disease slightly decreased from 2001 to 2010, although the number of patients with renal calculi increased. The use of PCNL vs NL was unchanged during this period. Multiple inequalities existed in overall surgical treatment rates and were influenced by sociodemographic factors such as age and sex.

Introduction

U pper urinary tract calculi can affect up to 8.8% of patients.¹ Stone disease represents a significant burden for patients and healthcare systems. In 2000, more than 1% of all working age adults underwent some treatment for kidney calculi, with costs of $3494 per patient.² Treatment modalities for renal pelvis calculi currently revolve around stone size and location.³

Percutaneous nephrolithotomy (PCNL) and nephrolithotomy (NL) represent treatment options for patients with a large stone burden.³ PCNL has become favored over NL because of its lower morbidity.⁴ NL is, therefore, rarely used, although it has found utility in patients with extremely large calculi, significant obesity, or those with anatomic abnormalities that would preclude use of PCNL.⁴ Studies assessing trends in the use of PCNL and NL are rare. In this study, we examined the use of PCNL and NL within the United States over a 10-year period.

Materials and Methods

After obtaining Institutional Review Board approval, Medicare claims data from the years 2001, 2004, 2007, and 2010 provided by the Centers for Medicare and Medicaid Services were analyzed. The Public Use Files include multiple datasets encompassing physician and hospital inpatient and outpatient medical claims for a 5% national random sample of beneficiaries. Patients with a diagnosis of kidney calculi were identified on the basis of International Classification of Diseases, 9th Edition (ICD-9). Patients undergoing surgery were identified on the basis of Current Procedure Terminology (CPT-4) and ICD-9 (cm) codes (see Appendix 1). Individual de-identified subjects were tracked using encrypted beneficiary identification numbers to link data across the multiple datasets representing care for both inpatient and outpatient settings. The national estimates of service use were obtained by multiplying counts by a constant weight of 20.⁵ Patients were categorized according to sex, ethnicity, age, and their geographical residence.

Statistical tests used included the chi-square and Fisher tests as well as multivariate logistic regression analyses. Statistical analyses were performed using SAS v9.3 (SAS Institute Inc., Cary, NC) and SPSS v20 (IBM Corp, Armonk, NY).

Results

Overall, a total of 983,840 patients with kidney calculi were identified during the entire study period (Table 1). The number of patients with a diagnosis of renal calculi increased from 167,280 in 2001 to 313,640 in 2010 (+87.5%). While the overall number of treated patients increased correspondingly from 5160 in 2001 to 7760 in 2010 (+50.4%), the percentage of patients with a diagnosis and who then moved on to surgical treatment, however, decreased from 3.1% to 2.5% (P<0.0001).

Table 1.

Diagnosis of Kidney Calculi and Surgical Treatment Rates 2001–2010

	Diagnosis of kidney calculi	Patients treated	%	P value
2001	167,280	5160	3.1	<0.0001
2004	236,140	6280	2.7
2007	266,780	6900	2.6
2010	313,640	7760	2.5
Total	983,840	26,100	2.7

Of the 26,100 patients undergoing surgery, 91% underwent PCNL vs 9% for NL (Table 2). The use of PCNL relative to NL remained constant from 2001 to 2010. Most procedures were performed in men (n=15,460; 52.8%). Most patients were white (n=25,320; 86.6%), while 13.4% (n=3920) were minorities. More surgeries were performed in the South (n=10,760; 36.9%) compared with other regions of the country.

Table 2.

Distribution of Surgical Treatments by Year

	PCNL	%	NL	%	P value
2001
Sex
Male	3040	93.8	200	6.2	<0.0001
Female	2540	87.6	360	12.4
Ethnicity
Whites	4700	91.8	420	8.2	<0.0001
Minorities	880	86.3	140	13.7
Age (years)
<65	1460	91.3	140	8.8	<0.0001
65–69	940	88.7	120	11.3
70–74	1260	92.6	100	7.4
75–79	980	90.7	100	9.3
80–84	480	82.8	100	17.2
>84	460	100.0	0	0.0
Geography
Northeast	1320	90.4	140	9.6	<0.0001
Midwest	1520	92.7	120	7.3
South	2280	95.0	120	5.0
West	400	80.0	100	20.0
2004
Sex
Male	3720	93.9	240	6.1	<0.0001
Female	3320	88.3	440	11.7
Ethnicity
Whites	5840	91.0	580	9.0	0.16
Minorities	1180	92.2	100	7.8
Age (years)
<65	1960	88.3	260	11.7	<0.0001
65–69	1520	92.7	120	7.3
70–74	1520	90.5	160	9.5
75–79	1020	92.7	80	7.3
80–84	560	90.3	60	9.7
>84	460	100.0	0	0.0
Geography
Northeast	1520	92.7	120	7.3	<0.0001
Midwest	1980	90.8	200	9.2
South	2560	89.5	300	10.5
West	960	94.1	60	5.9
2007
Sex
Male	3040	93.8	200	6.2	<0.0001
Female	2540	87.6	360	12.4
Ethnicity
Whites	7000	90.9	700	9.1	<0.0001
Minorities	1000	94.3	60	5.7
Age (years)
<65	2400	93.8	160	6.3	<0.0001
65–69	1860	89.4	220	10.6
70–74	1140	87.7	160	12.3
75–79	1120	90.3	120	9.7
80–84	820	91.1	80	8.9
>84	660	97.1	20	2.9
Geography
Northeast	1680	90.3	180	9.7	<0.0001
Midwest	2480	92.5	200	7.5
South	2740	89.0	340	11.0
West	1100	96.5	40	3.5
2010
Sex
Male	4740	88.8	600	11.2	<0.0001
Female	3940	92.5	320	7.5
Ethnicity
Whites	7780	90.9	780	9.1	0.001
Minorities	860	87.8	120	12.2
Age (years)
<65	2460	90.4	260	9.6	<0.0001
65–69	1720	86.0	280	14.0
70–74	1760	91.7	160	8.3
75–79	1160	90.6	120	9.4
80–84	900	93.8	60	6.3
>84	680	94.4	40	5.6
Geography
Northeast	1840	90.2	200	9.8	<0.0001
Midwest	2460	89.1	300	10.9
South	3180	91.4	300	8.6
West	1180	90.8	120	9.2
Total
Sex
Male	15,460	91.5	1,440	8.5	<0.0001
Female	13,840	90.3	1,480	9.7
Ethnicity
Whites	25,320	91.1	2,480	8.9	0.11
Minorities	3920	90.3	420	9.7
Age (years)
<65	8280	91.0	820	9.0	<0.0001
65–69	6040	89.1	740	10.9
70–74	5680	90.7	580	9.3
75–79	4280	91.1	420	8.9
80–84	2760	90.2	300	9.8
>84	2,260	97.4	60	2.6
Geography
Northeast	6360	90.9	640	9.1	0.28
Midwest	8440	91.1	820	8.9
South	10,760	91.0	1,060	9.0
West	3640	91.9	320	8.1
Years
2001	5580	90.9	560	9.1	<0.0001
2004	7040	91.2	680	8.8
2007	8000	90.9	800	9.1
2010	8680	90.4	920	9.6
Total	29,300	90.8	2,960	9.2

The choice of treatment was correlated with multiple demographic factors (Table 2). PCNL was performed slightly more frequently in men, although the difference was marginal (91.5% vs 90.3%, P<0.0001). PCNL was performed more often compared with NL for all age groups. The choice of treatment was not influenced by geography and ethnicity.

Logistic regression analysis was performed to evaluate the influence of demographic factors and time on whether patients underwent surgical treatment for stone disease (Table 3). Women had 1.19 greater odds of undergoing surgery compared with men (P=0.003). Older patients were less likely to undergo surgery (odds ratio [OR]=0.65–0.71, P<0.05). Patients treated after 2004 were less likely to undergo surgery (OR=0.77–0.84, P<0.05). No differences were noted for ethnicity (P=0.23) and geography (P=0.83).

Table 3.

Multivariate Logistic Regression Analysis Examining Likelihood of Surgical Treatment for Renal Stone Disease with Percutaneous Nephrolithotomy or Open Nephrolithotomy

	Treatment vs no treatment
Categories	OR	CI	P value
Sex
Female	1.19	1.06–1.34	0.003
Male	1.00
Ethnicity
Whites	1.00
Minorities	1.11	0.94–1.32	0.23
Geography
Overall	1.01	0.95–1.06	0.83
Northeast	1.13	0.97–1.32	0.11
Midwest	1.13	0.98–1.30	0.10
West	1.00	0.83–1.21	0.97
South	1.00
Age
Overall	0.94	0.91–0.98	0.02
<65	1.00
65–69	0.69	0.58–0.81	<0.0001
70–74	0.71	0.60–0.84	<0.0001
75–79	0.69	0.58–0.83	<0.0001
80–84	0.65	0.52–0.81	<0.0001
>84	0.80	0.64–1.01	0.06
Year
Overall	0.93	0.88–0.98	0.004
2001	1.00
2004	0.86	0.73–1.03	0.10
2007	0.84	0.71–0.99	0.04
2010	0.77	0.65–0.91	0.002

OR=odds ratio; CI=confidence interval.

Logistic regression analysis was also performed to evaluate the influence of demographic factors and time on the choice of stone procedure if patients were treated surgically (Table 4). Choice of stone procedure was independent of sex, ethnicity, age, geography, or year of treatment.

Table 4.

Multivariate Logistic Regression Analysis Examining Choice of Surgical Treatment for Renal Stone Disease with Percutaneous Nephrolithotomy or Open Nephrolithotomy

	PCNL vs NL
	OR	CI	P value
Sex	1.19	0.84–1.69	0.32
Ethnicity	1.09	0.65–1.85	0.74
Geography	0.97	0.80–1.16	0.70
Age	0.90	0.81–1.01	0.08
Year	1.06	0.91–1.25	0.46

PCNL=percutaneous nephrolithotomy; NL=open nephrolithotomy; OR=odds ratio; CI=confidence interval.

Discussion

Large or complex stone burden was traditionally addressed with NL before the advent of less invasive techniques such as PCNL.³ By 1999, for instance, Hollowell and associates⁶ noted that U.S. urologists recommended PCNL in 93.1% of patients and NL in only 6.9% of patients with staghorn calculi. By 2006, Bandi and colleagues⁷ reported that up to 81% of urologists recommended use of PCNL for renal calculi >2 cm, whereas only 1% of urologists would recommend NL.

In this study, we sought to further clarify trends in use of procedures for large stone burden. While the number of patients with a diagnosis of renal calculi increased by 87.5%, the percentage of patients surgically treated with PCNL or NL actually diminished slightly from 3.1 to 2.5% (P<0.0001). Stone patients receiving the diagnosis in the later years of the study—i.e., after 2004—also had lower odds of being treated with PCNL or NL (OR=0.77–0.84, p<0.05).

This is consistent with the study from Kim and coworkers⁸ which reported a declining use of both PCNL and open stone surgery by 48% and 51% of U.S. urologists, respectively, from 1996 to 2000. These data also support trends in the United Kingdom as reported by Turney and associates⁹ in which open surgery for upper tract stones decreased by 83% from 2000 to 2010⁹ as well as an earlier U.S. study by Kerbl and colleagues¹⁰ who reported an fourfold decrease in open stone surgery from 1988 to 2000.

The decrease in the use of PCNL and NL over this period might be related to earlier treatment of stone disease that would ostensibly be smaller or less complex and thereby be more amenable to less invasive approaches such as shockwave lithotripsy (SWL) or ureteroscopy (URS). In fact, Turney and coworkers⁹ stated that the use of SWL increased by 55% and URS by 127% from 2000 to 2010 while the rates of stone surgery were concomitantly declining by 83%.⁹

The use of PCNL vs NL appears to have stabilized at a 10:1 ratio in the Medicare cohort; the choice of surgical approach appears to be independent of patient sociodemographic factors. Interestingly, our study would also suggest that NL is used in up to 10% of patients with large or complex renal calculi in the Medicare cohort, despite the procedure being recommended by only 1% to 6.9% of urologists for large stone burden.^6,7

Multiple factors influenced whether patients underwent surgical treatment. Female patients, for example, had 1.19 greater odds of undergoing surgery (P=0.003). Although relatively more men (n=15,460; 52.8%) received a diagnosis of renal stone disease in this cohort, women, in contrast, were more likely to undergo surgical treatment. These discrepancies in care may reflect differences in availability of medical services, financial barriers such as health insurance coverage, differences in patient perception, as well as differing provider beliefs and behaviors in relation to sex.^11,12

Age also influenced whether patients underwent surgical treatment. Older patients were less likely to undergo surgery (OR=0.65-0.71, P<0.0001), likely because of a greater number of comorbidities as they aged. The impact of age and comorbidities in patients undergoing PCNL, for example, was reported recently by Unsal and colleagues¹³ in a series of 1406 patients. Charlson Comorbidity Indices tended to increase with patient age (P=0.001). Patients with higher Charlson Comorbidity Indices needed longer operative times (70.6 vs 63.8 min, P=0.008), needed more blood transfusions (16.5% vs 8.9%, P=0.049), and experienced more postoperative complications (21.6% vs 2.9%, P<0.001).¹³

Ethnicity and geography had no influence regarding surgical treatment. It was unexpected, however, to observe that while more stone disease was treated in areas like the so-called stone belt in the South, which is considered to have a higher incidence and prevalence of urolithiasis and where larger stone burden would likely exist, the likelihood of being treated with PCNL and NL in our study was not greater after adjustment for age, sex, ethnicity, and year of treatment.¹⁴ Indeed, the contribution of age, sex, and year of treatment seemed to affect the likelihood of treatment more than geography.

Others have reported on the practice variation seen with the use of PCNL. Morris and associates,¹⁵ for example, reported on temporal trends in the use of PCNL from the National Inpatient Sample. The annual use of PCNL was noted to increase from 1988 to 2002. In addition, the “quality” of procedure, defined by hospital mortality, was linked to hospital PCNL and discharge volume (P<0.01). Bandi and associates⁷ reported on practice patterns in the management of upper urinary tract calculi in the North Central Section of the American Urological Association.⁷ PCNL represented the most common treatment for renal calculi >2 cm (72%–92%) and was more likely to be used by fellowship-trained endourologists, academic urologists, and urologists in practice for less than 5 years. In contrast, Bird and coworkers¹⁶ reported in a survey of the same section that younger urologists and those trained to perform PCNL in residency were more comfortable performing PCNL.¹⁶ There are no studies on NL that report on long-term trends in use.

Our study possessed several limitations. Our data did not contain the provider-level data as reported by Bandi and colleagues⁷ and Bird and associates¹⁶ in which the use of PCNL is clearly correlated with surgeon-specific characteristics such as age, fellowship training, and surgical volume. Another critical limitation of this study is that stone size, location, and composition were unavailable because the Medicare Public Use Files represent only a billing-oriented claims database. We assumed that treatment with PCNL and NL in the kidney would be confined to only those patients with large stone burden or complex renal anatomy.

Only PCNL and NL were evaluated in this study; procedure codes for more novel approaches, such laparoscopic or robot-assisted nephrolithotomy, do not currently exist in the Public Use Files and thus could not be considered. In addition, alternative approaches such as URS, whose feasibility in large and complex renal calculi has been recently reported, were not be taken into consideration becsuse these were not deemed to be the current standard of care.^4,17,18

The study was retrospective and nonrandomized in nature, although sample size was large. The role of patient preferences regarding surgery could not be assessed.

Conclusions

The use of PCNL and NL for management of stone disease slightly decreased from 2001 to 2010, although the number of patients with renal calculi increased. The use of PCNL vs NL was unchanged during this period. Multiple inequalities existed in overall surgical treatment rates and were influenced by sociodemographic factors such as age and sex.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

Appendix

Appendix 1.

International Classification of Diseases-9 and Current Procedure Terminology Codes for Stone Disease Diagnosis and Treatment

Diagnosis codes for stone disease:

ICD-9 diagnosis codes: calculus of kidney 592.0; 274.11;

Procedure codes for stone disease:

ICD-9 cm procedure codes:

Percutaneous nephrolithotomy 55.04, 55.92, 59.95, 55.03; open nephrolithotomy 55.01

CPT-codes:

Percutaneous nephrolithotomy 50080, 50081; open nephrolithotomy 50130, 50060, 50070, 50075

ICD=International Classification of Diseases; CPT=Current Procedure Terminology.

References

Scales CD

, Smith

, Hanley

, Saigal

. Urologic Diseases in America Project. Prevalence of kidney stones in the United States. Eur Urol, 2012; 62:160–165.

Saigal

, Joyce

, Timilsina

. Urologic Diseases in America Project. Direct and indirect costs of nephrolithiasis in an employed population: Opportunity for disease management? Kidney Int, 2005; 68:1808–1814.

Türk

, Knoll

, Petrik

et al. Guidelines on Urolithiasis. Arnhem, The Netherlands: Eauropean Association of Urology, 2010Update 2012.

Preminger

, Assimos

, Lingeman

et al. Chapter 1: AUA guideline on management of staghorn calculi: Diagnosis and treatment recommendations. J Urol, 2005; 173:1991–2000.

Anger

, Rodriguez

, Wang

et al. Racial disparities in the surgical management of stress incontinence among female Medicare beneficiaries. J Urol, 2007; 177:1846–1850.

Hollowell

, Patel

, Bales

, Gerber

. Internet and postal survey of endourologic practice patterns among American urologists. J Urol, 2000; 163:1779–1782.

Bandi

, Best

, Nakada

. Current practice patterns in the management of upper urinary tract calculi in the north central United States. J Endourol, 2008; 22:631–636.

Kim

, Hollowell

, Patel

et al. Use of new technology in endourology and laparoscopy by american urologists: Internet and postal survey. Urology, 2000; 56:760–765.

Turney

, Reynard

, Noble

, Keoghane

. Trends in urological stone disease. BJU Int, 2012; 109:1082–1087.

10.

Kerbl

, Rehman

, Landman

et al.

Current management of urolithiasis: Progress or regress?

J Endourol, 2002; 16:281–288.

11.

Millman

. Access to Health Care in America. Washington DC: The National Academies Press, 1993.

12.

National Healthcare Disparities Report. Quality AfHRa: Washington, DC, 2004.

13.

Unsal

, Resorlu

, Atmaca

et al. Prediction of morbidity and mortality after percutaneous nephrolithotomy by using the Charlson Comorbidity Index. Urology, 2012; 79:55–60.

14.

Brikowski

, Lotan

, Pearle

. Climate-related increase in the prevalence of urolithiasis in the United States. Proc Natl Acad Sci U S A, 2008; 105:9841–9846.

15.

Morris

, Wei

, Taub

et al. Temporal trends in the use of percutaneous nephrolithotomy. J Urol, 2006; 175:1731–1736.

16.

Bird

, Fallon

, Winfield

. Practice patterns in the treatment of large renal stones. J Endourol, 2003; 17:355–363.

17.

Cohen

, Cohen

, Grasso

. Ureteropyeloscopic treatment of large, complex intrarenal and proximal ureteral calculi. BJU Int, 2013; 111:E127–E131.

18.

Takazawa

, Kitayama

, Tsujii

. Successful outcome of flexible ureteroscopy with holmium laser lithotripsy for renal stones 2 cm or greater. Int J Urol, 2012; 19:264–267.