British Journal of Urology (1998), 82, 12–15

The incidence of lower-pole nephrolithiasis — increasing ornot?A.S. CASS, W.B. GRINE, J .McK. JENKINS, W.R. JORDAN, T.B. MOBLEY and D.A. MYERSMidwest Urologic Stone Unit, Minneapolis, Minnesota and Lithotripters, Inc., Fayetteville, North Carolina, USA

Objective To determine if the incidence of lower-pole in both samples was compared with that reportedearlier.nephrolithiasis is increasing.

Methods A previously published meta-analysis of trends Results The meta-analysis for stone location trends fromthe previously published article suggested that thein the location of stones in the kidney, using data

from 1984 to 1992, determined the percentage of percentage of kidney stones in the lower pole at ESWLincreased erratically from 1984 to 1989 but was thenlower pole stones in 26722 kidney stones treated by

extracorporeal shockwave lithotripsy (ESWL). We per- stable for 3 years. The Lithotripters Inc. sample showedan essentially constant incidence from 28% in 1990formed prospective studies on all patients treated by

ESWL for a single renal stone (not manipulated from to 30% in 1995, and the Midwest Urologic Stone Unitsample showed an essentially constant incidence fromthe ureter) in two organizations; at Lithotripters Inc.,

47 303 stones were treated with ESWL by 1000 35% in 1988 to 36% in 1995.Conclusion The incidence of lower pole nephrolithiasisurologists in private practice from 1989 to 1995.

At the Midwest Urologic Stone Unit, 9357 stones has remained stable from 1990.Keywords Kidney stone, location, distribution, trendswere treated with ESWL by 200 urologists in private

practice from 1987 to 1995. The distribution of stones

the confidence-profile method, which allows an analysisIntroductionof data from studies that are not randomized controlledstudies. The meta-analysis for stone location trends inIn comparing the results of ESWL and percutaneous

nephrolithotomy (PCN) for lower-pole renal stones, patients treated by ESWL determined the percentage ofkidney stones in the lower pole from 26722 renal stonesLingeman et al. [1] reported that a meta-analysis for stone

location trends, using the combined data set from the treated from 1984 to 1992. The authors did not state ifthe 26722 renal stones were single or multiple.abstract forms used by the AUA Nephrolithiasis Guidelines

Panel, showed the percentage of kidney stones in the In the present study, two groups of urologists inprivate practice (Lithotripters, Inc., comprising 1000lower pole at ESWL had increased steadily from 2% in

1984 to 48% in 1992 [1]. The percentage of kidney urologists, and Midwest Urologic Stone Unit, 200 urol-ogists) entered all patients treated by ESWL into astones in the lower pole at ESWL was evaluated for two

groups of urologists in private practice (Lithotripters, Inc. prospective study, with the data recorded at the time ofESWL. In this evaluation, only single renal stones notand Midwest Urologic Stone Unit) to determine if the

incidence of lower pole stones was increasing. manipulated from the ureter were assessed. Lithotripters,Inc. started in November 1988 and determined thepercentage of kidney stones in the lower pole at ESWLMethodsfrom 1989 to 1995, with 47303 stones treated byESWL. The Midwest Urologic Stone Unit started inThe AUA Nephrolithiasis Clinical Guidelines Panel

searched the literature using Medline and retrieved all September 1986 and determined the percentage ofkidney stones in the lower pole at ESWL from 1987 toarticles related to renal calculi from peer-reviewed jour-

nals in the English literature; the search also included 1995, with 9357 stones treated by ESWL.other important journals not listed on Medline (e.g. TheJournal of Endourology). From the 1250 articles retrieved Results[2], Lingeman et al. [1] performed a meta-analysis using

The meta-analysis of published articles [1] showed anincidence of stones in the various locations in the kidneyAccepted for publication 27 March 1998

12 © 1998 British Journal of Urology

INCIDENCE OF LOWER-POLE NEPHROLITHIASIS 13

as given in Table 1. From 1984, the increase in the Discussionincidence of lower-pole stones was erratic. Statisticalevaluation using 95% CIs showed a highly significant Lingeman et al. [1] stated that from a meta-analysis of

published results, the increasing percentage of stones indecrease in 1987, maintained in 1988, and followed bya significant increase in 1989. The incidence was stable the lower pole being treated by ESWL reflected new

trends in stone management since the advent of less-for 3 years, apart from a significant increase in 1992.The results from Lithotripters, Inc. for the incidence of invasive techniques such as ESWL. This finding diCers

from the present results, which show an essentiallystones in various locations in the kidney is shown inTable 1; the percentage of stones in the lower pole at constant incidence of kidney stones in the lower pole at

ESWL. This diCerence could be explained: (i) if the stoneESWL showed an essentially constant incidence of 28%in 1990 to 30% in 1995. The results from the Midwest distribution or the method of management of stones

(PCN vs ESWL) in patients treated by urologists inUrologic Stone Unit are also shown in Table 1; thepercentage of stones in the lower pole at ESWL showed private practice diCered in the early years from that in

University centres which produced most of the publishedan essentially constant incidence of 35% in 1988 to36% in 1995. A similar statistical evaluation of the last results; (ii) if all 1250 published articles were used

in the meta-analysis, then the 26722 renal calculitwo datasets showed a significant increase in the initialperiod and then a stable incidence in both samples. The reviewed would have averaged 21.4 renal calculi per

article, which is very low, i.e. there may have been sometrends in the incidence of lower pole stones at ESWL inthese three studies are shown in Fig. 1. degree of selection of the published results in the

Table 1 Changes in stone distribution1984–92 from Lingeman et al. [1], fromLithotripter, Inc and from Midwest UrologicStone Unit. In the last two samples, stoneswere single renal stones not manipulatedfrom the ureter

Number (%) stones in

Year Upper pole Mid pole Lower pole Pelvis Total

[1]1984 27 (10) 2 (1) 4 (2) 227 (87) 2601985 187 (9) 142 (7) 334 (16) 1395 (68) 20581986 520 (11) 500 (10) 1215 (24) 2727 (55) 49621987 99 (6) 70 (5) 248 (16) 1110 (73) 15271988 256 (5) 219 (4) 797 (14) 4373 (77) 56451989 699 (10) 724 (11) 2261 (34) 3015 (45) 66991990 141 (12) 85 (7) 387 (31) 623 (50) 12361991 429 (13) 292 (9) 1015 (31) 1548 (47) 32841992 110 (10) 169 (16) 501 (48) 271 (26) 1051Total 2468 (9) 2203 (8) 6762 (25) 15289 (57) 26722Lithotripter Inc1989 55 (11) 234 (48) 105 (22) 90 (19) 4841990 462 (13) 1778 (48) 1039 (28) 413 (11) 36921991 942 (13) 3000 (43) 2210 (31) 937 (13) 70891992 1155 (14) 3356 (40) 2549 (31) 1286 (15) 83461993 895 (10) 3633 (43) 2691 (32) 1273 (15) 84921994 1192 (12) 3655 (39) 2899 (31) 1708 (18) 94541995 1116 (11) 3797 (40) 2960 (30) 1873 (19) 9746Total 5817 (12) 19 453 (41) 14 453 (31) 7580 (16) 47303Midwest Stone1987 44 (9) 29 (6) 142 (30) 262 (55) 4771988 65 (9) 61 (8) 252 (35) 348 (48) 7261989 54 (7) 79 (10) 255 (32) 404 (51) 7921990 72 (8) 91 (9.5) 338 (35.5) 450 (47) 9511991 69 (7) 103 (10) 354 (34) 503 (49) 10291992 62 (5) 112 (10) 387 (34) 580 (51) 11411993 82 (6) 97 (7) 448 (35) 667 (52) 12941994 97 (7) 128 (9) 493 (36) 659 (48) 13771995 107 (7) 148 (9) 559 (36) 756 (48) 1570Total 652 (7) 848 (9) 3288 (35) 4629 (49) 9357

© 1998 British Journal of Urology 82, 12–15

14 A.S. CASS et al.

Fig. 1. Incidence of lower-pole stonestreated by ESWL; the lines represent the95% CI for each dataset, where (dark green)is from Lingeman et al. [1], (light green)data from Lithotripter Inc and (light red)

1984

Year

Inci

den

ce o

f lo

wer

-po

le s

ton

es (

%)

19860.0

10.0

1988 1990 1992 1994 1996

20.0

30.0

40.0

50.0

60.0

data from Midwest Stone Management.

meta-analysis; (iii) the erratic increase in the data pre- direction of treatment eCect and in statistical significance,as the larger trial.sented by Lingeman et al. suggests that there are other

factors determining the (published) incidence of lower- Egger and Smith [7] compared the result of meta-analysis showing a fall in mortality with magnesiumpole stones.

Meta-analysis in clinical research involves systemati- treatment in myocardial infarction with the negativeresults of the Fourth International Study of Infarctcally seeking and quantitatively combining the results

of all studies that have addressed a similar research Survival. LeLorier et al. [8] compared the results of large,randomized, controlled trials (involving>1000 patients)question to give single estimates of treatment eCect.

However, it can generate misleading results through published in four journals (the New England Journal ofMedicine, The Lancet, the Annals of Internal Medicine, andpublication bias, duplicate data and ignoring unpub-

lished trials and data. Easterbrook et al. [3] analysed the Journal of the American Medical Association) with theresults of meta-analyses published earlier on the same285 studies and found that studies with statistically

significant results were more likely to be published than topics. These authors found the outcomes of the 12 largerandomized controlled trials were not predicted accu-those finding no diCerence between the study groups,

and to lead to more publications. Tramer et al. [4] rately 35% of the time by the meta-analyses publishedpreviously on the same topics. Egger et al. [9] usedstudied published reports of randomized controlled trials

investigating the eCect of ondansetron on postoperative funnel plots (plots of eCect estimates against sample size)to detect bias in meta-analyses that were later contra-emesis and found that duplicated data was included in

a meta-analysis which led to a 23% overestimation of dicted by large trials in a study of eight pairs of meta-analyses and large trials found by Medline search (fivethe anti-emetic eBcacy of ondansetron. Simes [5] used

the International Cancer Research Data Bank registry of from cardiovascular medicine, one from diabetic medi-cine, one from geriatric medicine, one from perinatalcancer clinical trials to evaluate the eCect of chemo-

therapy on survival in advanced ovarian cancer and medicine). The authors found four concordant and fourdiscordant pairs (concordance of results was assumed iffound the conclusions based on a pooled analysis of

registered trials had important diCerences from a more eCects were in the same direction and the meta-analyticestimate was within 30% of the trial) and concludedtraditional review of the published trials.

Studies have compared the results of meta-analysis that meta-analysis based on a few small trials should betreated with considerable caution.from several smaller trials to the results of a large trial.

Villar et al. [6] compared 30 meta-analyses of diCerent An editorial in the British Medical Journal [10] con-cluded ‘meta-analysis has made and continues to makeinterventions in perinatal medicine, covering 185 ran-

domized controlled trials, with the results of a large trial major contributions to medical research, clinical decisionmaking and standards of research reportage. However,(total sample size of>1000) and found 24meta-analyses

correctly predicted the direction of the treatment eCect, it is no panacea. Readers need to examine any meta-analysis critically to see whether researchers havebut only 18 of the 30 were the same, both in the

© 1998 British Journal of Urology 82, 12–15

INCIDENCE OF LOWER-POLE NEPHROLITHIASIS 15

3 Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR.overlooked important sources of clinical heterogeneityPublication bias in clinical research. Lancet 1991; 337:among the included trials. They should demand evidence867–72that the authors undertook a comprehensive search,

4 Tramer MR, Reynolds DJM, Moore RA, McQuay HJ. Impactavoiding covert duplicate data and unearthing unpub-of covert duplication on meta-analysis: a case study. Brlished trials and data. Lastly, readers and researchersMed J 1997; 315: 635

alike need to appreciate that not every systematic review5 Simes RJ. Confronting publication bias: a cohort design for

should lead to an actual meta-analysis of data with meta-analysis. Stat Med 1987; 6: 11–29aggregate eCect size estimates. If the process of pooling 6 Villar J, Carrol G, Belizan JM. Predictive ability of meta-data inadvertently drowns clinically important evidence analyses of randomised controlled trials. Lancet 1995;from individual studies, then a meta-analysis can do 345: 772

7 Egger M, Smith GD. Misleading meta-analysis. Br Med Jmore harm than good.’1995; 310: 752In conclusion, the incidence of lower-pole nephrolithi-

8 LeLorier J, Gregoire G, Benhaddad A, Lapierre J, Derderianasis has not increased significantly in patients treated byI. Discrepancies between meta-analyses and subsequenturologists in private practice, at least since 1990; thelarge, randomised, controlled studies. New Eng J Med 1997;previously reported erratically increasing incidence from337: 536the meta-analysis for stone location trends should be

9 Egger M, Smith GD, Schneider M, Minder C. Bias in meta-viewed with caution.

analysis detected by a simple, graphical test. Br Med J1997; 315: 629

10 Naylor CD. Meta-analysis and the meta-epidemiology ofAcknowledgementclinical research. Br Med J 1997; 315: 617

We thank the urologists in Lithotripters, Inc. andMidwest Urologic Stone Unit who cooperated by includ-ing their patients in this study.

AuthorsA.S. Cass, MBBS, FRCS, Medical Director, Midwest Urological

References Stone Unit.W.B. Grine, MD, Urologist, Lithotripters Inc.1 Lingeman JE, Siegel YI, Steele B, Nyhuis AW, Woods John

R. Management of lower pole nephrolithiasis: A critical J. McK. Jenkins, MD, Urologist, Lithotripters Inc.W.R. Jordan, MD, Urologist, Lithotripters Inc.analysis. J Urol 1994; 151: 663–7

2 Segura JW, Preminger GM, Assimos DGL et al. T.B. Mobley, MD, Urologist, Lithotripters Inc.D.A. Myers, MD, Urologist, Lithotripters Inc.Nephrolithiasis clinical guidelines panel summary report

on the management of staghorn calculi. J Urol 1994; Correspondence: Dr A.S. Cass, Midwest Stone Management,1885 County Road C West, Roseville, MN 55113, USA.151: 1648

© 1998 British Journal of Urology 82, 12–15