Heterozygous NPHS1 or NPHS2 mutations in responsive nephrotic syndrome and the multifactorial origin...

Kidney International, Vol. 66 (2004), pp. 1711–1718

LETTERS TO THE EDITOR

Underuse of Hardy-Weinbergequilibrium

To the Editor: Similar to other analyses of articles pub-lished in several journals [1–3], Kocsis et al [4] recentlyalso reported an evaluation of manuscripts dealing withgenetic polymorphisms and the underuse of calculationsof Hardy-Weinberg equilibrium in Kidney International.

Although this article has some merits, we wish topoint out that the work of our group [5] has been mis-reported and misinterpreted. In Table 1 of the articleby Kocsis et al, our association study of a genetic poly-morphism (MTHFR 677C>T) with total homocysteine(tHcy) plasma levels in kidney transplant patients [5] hasbeen referred to as having included 85 subjects showinga significant deviation of “MTFR-677” genotype distri-bution from the Hardy-Weinberg equilibrium (reference[19] of the article of Kocsis et al).

Instead, we have applied the concept of Mendelian ran-domization in our study, and have included all 63 indi-viduals that were homozygous for the MTHFR 677C>Tfrom a total study population of 636 kidney transplant pa-tients. These 63 homozygous subjects were matched with63 heterozygous individuals (CT genotype), and with 63patients homozygous for the wild-type alleles (CC geno-type) for age, gender, estimated glomerular filtration rate,and body mass index, and were compared for biochemicalmarkers of homocysteine metabolism. This study showedthat MTHFR 677TT patients have significantly highertHcy levels in contrast to patients with the CT or the CCgenotype. Finally, multivariate analyses that controlledfor important confounders confirmed this association.

Thus, Kocsis et al should more carefully read, ana-lyze, and interpret the data reported in the literaturebefore concluding that deviations from Hardy-Weinbergequilibrium need more consideration in gene-associationstudies.

GERE SUNDER-PLASSMANN and MANUELA FODINGER

Vienna, Austria

Correspondence to Gere Sunder-Plassmann, M.D., Division ofNephrology and Dialysis, Department of Medicine III, University of Vi-enna, Wahringer Gurtel 18-20, A-1090 Vienna, Austria.E-mail: [email protected]

C© 2004 by the International Society of Nephrology

REFERENCES

1. BARDOCZY Z, GYORFFY B, KOCSIS I, VASARHELYI B: Re-calculatedHardy-Weinberg values in papers published in Atherosclerosis be-tween 1995 and 2003. Atherosclerosis 173:141–143, 2004

2. GYORFFY B, KOCSIS I, VASARHELYI B: Biallelic genotype distributionsin papers published in Gut between 1998 and 2003: Altered con-clusions after recalculating the Hardy-Weinberg equilibrium. Gut53:614–615, 2004

3. GYORFFY B, KOCSIS I, VASARHELYI B: Missed calculations and newconclusions: Re-calculation of genotype distribution data publishedin Journal of Investigative Dermatology. J Invest Dermatol 122:644–646, 2004

4. KOCSIS I, GYORFFY B, NEMETH E, VASARHELYI B: Examination ofHardy-Weinberg equilibrium in papers of Kidney International: Anunderused tool. Kidney Int 65:1956–1958, 2004

5. FODINGER M, WOLfl G, FISCHER G, et al: Effect of MTHFR 677C>Ton plasma total homocysteine levels in renal graft recipients. KidneyInt 55:1072–1080, 1999

Reply from the Authors

Our aim was not to blame any authors, but to call theattention to the importance of Hardy-Weinberg equilib-rium (HWE) calculation in genetic studies [1].

In an association study like in Fodinger et al [2], thefulfillment of HW criteria is not a must. However, we areconvinced that the significant deviation of the genotypedistribution is also worth to mention in these reports. Intheir report, according to Table 1, the 85 kidney graft re-cipients with polycystic kidney disease (PKD) are out ofHWE in point of MTHFR C677T polymorphism, and thismay be interesting for others working in this field. In ourstudy we suggested that while the results of Fodinger et alare applicable for most graft recipient patients, possiblythis is not the case for PKD patients. The missing HWE inthis subgroup supports an independent correlation withMTHFR gene, and this may have an impact on the in-terpretation of their results. The skewed genotype distri-bution calls the attention that genetic polymorphisms ofPKD patients should be analyzed separately.

ISTVAN KOCSIS, BALAZS GYORFFY, and BARNA VASARHELYI

Budapest, Hungary

Correpondence to Istvan Kocsis, First Department of Paediatrics, Sem-melweis University, Budapest, Hungary.E-mail: [email protected]

REFERENCES

1. KOCSIS I, GYORFFY B, NEMETH E, VASARHELYI B: Examination ofHardy-Weinberg equilibrium in papers of Kidney International: Anunderused tool. Kidney Int 65:1956–1958, 2004

2. FODINGER M, WOLFL G, FISCHER G, et al: Effect of MTHFR 677C>Ton plasma total homocysteine levels in renal graft recipients. KidneyInt 55:1072–1080, 1999

1711

1712 Letters to the Editor

Regarding impact of epoetinalfa on clinical end points inpatients with chronic renalfailure: A meta-analysis

To the Editor: We read the recent article by Joneset al [1] with interest. This meta-analysis purports to showthat recombinant human erythropoietin (rHuEPO) ther-apy increases hemoglobin levels and reduces hospital-ization rates. Unfortunately, the design of this study hasseveral potential limitations that were not expressed inthe manuscript.

First, the inclusion criteria were so exclusive that sev-eral relevant trials were not included, which may haveinfluenced the outcome of the meta-analysis. It is note-worthy that these criteria excluded the three largest andmost recent randomized studies. An example of an ex-cluded study that might have altered the pooled effect ofrHuEPO therapy is the work by Besarab et al [2].

Second (and more important), 11 of 16 trials includedwere nonrandomized. Because nonrandomized studiesare more likely to exaggerate estimates of treatment ef-fect [3, 4], and because inclusion of lower quality studies isknown to affect the results of meta-analyses [5], we urgethat the reader be circumspect of results driven by these11 trials alone. Because the excluded trials tended to findno benefit of rHuEPO with respect to hospitalization, wewonder what the meta-analysis would have shown if allavailable randomized trials were considered.

In addition to these considerations about study de-sign, we disagree with the authors’ interpretation of thedata. The authors state that “strong evidence” showsthat rHuEPO reduces hospitalization rates, and by ex-tension, healthcare costs. However, hospitalization costsare determined primarily by the number of hospital days,rather than by the total number of hospitalizations. WhileJones et al state that the length of stay was reducedwith rHuEPO therapy, its “benefit” was not statisticallysignificant.

Although erythropoietic therapy undoubtedly raiseshemoglobin levels and probably improves quality of life,the available data do not convincingly indicate that it re-duces hospitalization rates in people with kidney disease.Rather, we view the information as weak and inconsis-tent.

We have no conflict of interest to declare.

MARCELLO TONELLI, WILLIAM F. OWEN, JR., KAILASH JINDAL,WOLFGANG C. WINKELMAYER, and BRADEN MANNS

Edmonton, Alberta, Canada, Durham, North Carolina, Waukegan,Michigan, Boston, Massachusetts, and Calgary, Alberta, Canada

Correspondence to Marcello Tonelli, Division of Nephrology, Uni-versity of Alberta, Edmonton, Alberta, Canada.E-mail: [email protected]

REFERENCES

1. JONES M, IBELS L, SCHENKEL B, et al: Impact of epoetin alfa on clinicalend points in patients with chronic renal failure: A meta-analysis.Kidney Int 65:757–767, 2004

2. BESARAB A, BOLTON WK, BROWNE JK, et al: The effects of normal ascompared with low hematocrit values in patients with cardiac diseasewho are receiving hemodialysis and epoetin. N Engl J Med 339:584–590, 1998

3. LINDE K, SCHOLZ M, MELCHART D, et al: Should systematic re-views include non-randomized and uncontrolled studies? The caseof acupuncture for chronic headache. J Clin Epidemiol 55:77–85,2002

4. GREEN SB: Patient heterogeneity and the need for randomized clin-ical trials. Control Clin Trials 3:189–198, 1982

5. MOHER D, PHAM B, JONES A, et al: Does quality of reports of ran-domised trials affect estimates of intervention efficacy reported inmeta-analyses? Lancet 352:609–613, 1998

Editor’s note: The study by Jones et al was sponsoredby RW Johnson Pharmaceutical Research and Develop-ment (Raritan, NJ), a subsidiary of Johnson and John-son, which operates companies that manufacture epoetinalfa.


The letter from Tonelli et al [1] concerning our workreveals what we consider to be a number of misunder-standings and inconsistencies, which we would like toaddress.

The work is criticized as being too exclusive (i.e., omit-ting Besarab et al [2]), and at the same time, too inclusivein that it included studies other than randomized con-trolled trials (RCTs). Besarab et al [2] is actually a studyof congestive heart failure patients on dialysis, and notdirectly relevant to our focus on chronic renal failure pa-tients. Within this population, we sought to include allevidence available.

We must, however, strenuously reject the assertionfrom the letter’s first paragraph “. . . potential limitationsthat were not expressed in the manuscript.” The use ofnon-RCTs, as well as the strengths and weaknesses ofthe accumulated data, are all well documented in themanuscript.

We agree the impact of research design is a criti-cally important issue. However, perhaps it was not im-mediately clear that Table 5 reanalyzes the data fromRCTs alone. In contrast to the assertion by Tonelliet al, the estimates of effect from the RCT’s are simi-lar to, or slightly larger than, those from all studies as awhole.

Letters to the Editor 1713

Tonelli et al disagree with our interpretation of thehospitalization data. While hematocrit and quality oflife were the central focus of our work, we also believethat the hospitalization rate data suggest a benefit withepoetin alfa. We look forward to further research to cor-roborate these findings.

Overall, we would not represent the results in our workas anything but what they are, a synthesis of the availableevidence with all its strengths and weaknesses. We believethe criticisms put forth by Tonelli et al stem largely froma misunderstanding of our article.

MICHAEL P. JONES, LLOYD IBELS,BRAD SCHENKEL, and MARTIN ZAGARI

Raritan, New Jersey, and Sydney, Australia

Correspondence to Brad Schenkel, Johnson & Johnson Pharmaceu-tical Services, LLC, Raritan, NJ.E-mail: [email protected]

REFERENCES

1. TONELLI M, OWEN WF, JINDAL KK, et al: Regarding impact of epoetinalfa on clinical end points in patients with chronic renal failure: Ameta-analysis. Kidney Int 66:1712–1713, 2004

2. BESARAB A, BOLTON WK, BROWNE JK, et al: The effects of normal ascompared with low hematocrit values in patients with cardiac diseasewho are receiving hemodialysis and epoetin. N Engl J Med 339:584–590, 1998

Mycophenolate mofetil in IgAnephropathy

To the Editor: We wish to raise several issues regard-ing the publication by Maes et al [1], reporting on theeffectiveness of mycophenolate mofetil (MMF) in IgAnephropathy (IgAN): (1) A sample of 34 patients is toolittle. We have started a randomized trial, ramipril versusramipril plus MMF in IgAN [2]: 57 patients per group arerequired to maintain a power of 80% and a type error of5%. For a dropout rate of 10%, a total sample size of126 patients needs to be enrolled. In the placebo group,2 patients (1 death, 1 adverse event) out of 13 patients(15%), and in the MMF group, 8 patients [2 end-stagerenal disease (ESRD), 2 emigrated, 1 adverse event, 1tuberculosis (TB), 2 gastrointestinal (GI) problems] outof 21 patients (38%) reduced or stopped the treatmentbefore the end of the trial.

(2) Inclusion criteria. “Eligible patients were random-ized (2:1; MMF:placebo).” In a randomized trial all pa-tients should have the same probability of receiving one

or the other of the treatments being compared [3]. Renalfunction for inclusion was taken at the time of diagnosis(ie, 0 to 5 years back from randomization). Were patientswith glomerular filtration rate (GFR) <20 mL/min in-cluded? Options for eligibility were also hypertension,proteinuria, and histologic severity, alone or in combina-tion. Were these risk factors evenly distributed betweengroups?

(3) A wash-out period for patients assuming enalaprilwas not performed, biasing proteinuria at entry.

(4) In a small patient sample, data distribution is notusually normal and a nonparametric approach wouldhave been more appropriate.

VINCENZO SEPE, NATALIA ROSSI, GABRIELLA ADAMO, andANTONIO DAL CANTON

Pavia, Italy

Correspondence to Vincenzo Sepe, Unita Operativa di Nefrologia,Dialisi Trapianto, IRCCS Policlinico, San MatteoP.le Golgi, 2 – 27100Pavia, Italy.E-mail: [email protected]

REFERENCES

1. MAES BD, OYEN R, CLAES K, et al: Mycophenolate mofetil in IgAnephropathy: Results of a 3-year prospective placebo-controlled ran-domised study. Kidney Int 65:1842–1849, 2004

2. THE ANGIOTENSIN INHIBITION MYCOPHENOLATE MOFETIL IGA

NEPHROPATHY STUDY INVESTIGATORS: One-year angiotensin-converting enzyme inhibition plus mycophenolate mofetil immuno-suppression in the course of early IgA nephropathy: A multicentre,randomised, controlled study. J Nephrol submitted

3. SACKETT DL, HAYNES RB, GUYATT GH, TUGWELL P (editors): Clini-cal Epidemiology. Philadelphia, PA, Lippincott Williams & Wilkins,1991, pp 196


We agree that, from a statistical point of view, the studyis not powered to prove differences or equivalence be-tween the two groups. Therefore, large multicenter trialsare warranted. Even a study of 126 patients is small to un-equivocally prove superiority of mycophenolate mofetil(MMF) treatment over standard renoprotective treat-ment (including salt restriction, angiotensin II suppres-sion) plus frequent follow-up (compliance). Using thedata from our study population, the number of patientsper group after dropout would have to be 83 to maintaina power of 80% and an a-value of 0.05 [1, 2]. Because thatnumber could not be reached, a 2:1 randomization wasperformed in order to maximize the number of patientsexposed to MMF. We clearly stated this in the abstractand discussion section of the manuscript.

As stated, patients with GFR <20 mL/min were ex-cluded, and risk factors did not differ significantly be-tween groups alone or in combination.


There was indeed no washout period of enalapril; how-ever, only 27% of patients were on enalapril at entry [29%in the MMF group and 23% in the placebo group; P =0.7867 (chi-square)]. The lack of impact of MMF on renalfunction noted in our study is difficult to explain by thisphenomenon.

Because of small sample size, the data were ana-lyzed using nonparametric statistics (Wilcoxon rank sumtest, chi-square test, linear mixed models), applicable toboth normal and non-normal distributed sample data[2, 3].

BART D. MAES and YVES F. CH. VANRENTERGHEM

Correspondence to B. Maes, MD, PhD, Nephrology, University Hos-pital Gasthuisberg, B-3000 Leuven, Belgium.E-mail: [email protected]

REFERENCES

1. MAES BD, OYEN R, CLAES K, et al: Mycophenolate mofetil in IgAnephropathy: Results of a 3-year prospective placebo-controlled ran-domised study. Kidney Int 65:1842–1849, 2004

2. SAS/STAT User’s Guide, Release 8.2, SAS Institute, Inc., Cary, NC,2001.

3. VERBEKE G: Linear mixed models for longitudinal data, in Springerseries in statistics, edited by Verbeke G, Molenberghs G, New York,Springer-Verlag, 2000, pp 93–119

Hepatic iron in hemodialysispatients

To the Editors: Canavese et al [1] reported that hepaticiron overload is common in hemodialysis patients, andsuggested a reevaluation of acceptable iron parameters.This was a well-designed study, and the work is an impor-tant contribution to our knowledge on iron storage in thispatient population. It should be noted, however, that itmay not be reasonable to extrapolate these results to thegeneral hemodialysis population. To properly answer thequestion of how prevalent iron overload is in hemodial-ysis patients, an unselected group of patients should bestudied. This was not true of Canavese’s cohort. Thirtyout of 40 subjects (75%) had to discontinue 15 monthsof continuous intravenous iron therapy due to serum fer-ritin >500 ng/mL. Therefore, there was a strong selectionbias towards an iron-overloaded subpopulation.

It should be noted that the major finding of this study,that many patients on hemodialysis had mild to moderatehepatic iron “overload,” is not a new finding. More than20 years ago, Ali et al [2] and Gokal et al [3] performed au-topsy studies and found excess hepatic iron in hemodialy-sis patients. In contrast to Canavese’s study, these authorshad access to hepatic tissue, allowing them to determine if

the excess iron was found in association with tissue dam-age. Ali found significant iron excess in the livers of 48%of subjects. Importantly, no liver pathology or damagewas found, even in cases of severe iron overload. Indeed,anecdotally, there does not appear to be any excess preva-lence of cirrhosis among hemodialysis patients. The factthat hepatic tissue damage does not seem to coexist withhepatic iron overload suggests that hepatic iron excess,as found by Ali, Gokal, and Canavese, may simply be areflection of shifted iron pools. Indeed, an elevated serumferritin (iron storage marker) concurrent with low or nor-mal transferrin saturation (iron circulation marker) is afrequent finding in hemodialysis patients, consistent witha shift of iron pools away from the circulation and intostorage tissues. Inflammation may be the key central linkand driver of this process. Recently, a strong associationwas found between measures of inflammation and serumferritin [4, 5]. The phenomenon may best be understoodas inflammation leading to a state of reticuloendothelialblockade that causes hemodialysis patients to have in-creased hepatic storage of poorly mobile iron.

STEVEN FISHBANE, NOBUYUKI MIYAWAKI, and NAVEED MASANI

Mineola, New York

Correspondence to Steven Fishbane, Department of Nephrology,Winthrop-University Hospital, Mineola, NY.E-mail: [email protected]

REFERENCES

1. CANAVESE C, BERGAMO D, CICCONE G, et al: Validation of serumferritin values by magnetic susceptometry in predicting iron overloadin dialysis patients. Kidney Int 65:1091–1098, 2004

2. ALI M, FAYEMI AO, RIGOLOSI R, et al: Hemosiderosis in hemodialysispatients. An autopsy study of 50 cases. JAMA 244:343–345, 1980

3. GOKAL R, MILLARD PR, WEATHERALL DJ, et al: Iron metabolism inhaemodialysis patients. A study of the management of iron therapyand overload. QJM 48:369–391, 1979

4. KALANTAR-ZADEH K, RODRIGUEZ RA, HUMPHREYS MH: Associationbetween serum ferritin and measures of inflammation, nutrition andiron in haemodialysis patients. Nephrol Dial Transplant 19:141–149,2004

5. GUNNELL J, YEUN JY, DEPNER TA, KAYSEN GA: Acute-phase re-sponse predicts erythropoietin resistance in hemodialysis and peri-toneal dialysis patients. Am J Kidney Dis 33:63–72, 1999

The impact of serum uric acidon cardiovascular outcomes inthe LIFE study

To the Editor: The analysis of the role of serum uricacid (SUA) levels in the Losartan In tervention for End-point reduction in hypertension study (LIFE) study is of


considerable interest [1]. Briefly, 29% of the superiorityof losartan (compared with atenolol) on the primary com-posite end point was attributed to a fall in SUA.

In the Greek Atorvastatin and Coronary-Heart dis-ease Evaluation (GREACE) study, using atorvastatin toachieve the low-density lipoprotein cholesterol (LDL-C)goal for high-risk patients (100 mg/dL; 2.6 mmol/L) wasassociated with a significant (P < 0.0001) fall in SUA lev-els [2]. In contrast, a significant (P < 0.0001) increase inSUA occurred in the ‘usual care’ group, where the ma-jority did not achieve the LDL-C target. Every 1 mg/dLfall in SUA resulted in a decreased hazard ratio (0.76;95% CI 0.62–0.89; P = 0.001) for vascular events. The fallin SUA was attributed to a parallel reduction in serumcreatinine (SCr).

A decrease in SUA and SCr was seen in 103 periph-eral arterial disease patients taking simvastatin [3]. In theHeart Protection Study, the simvastatin group had a sig-nificantly smaller (P < 0.0001) increase in SCr than theplacebo group [4].

In the LIFE study, the fall in SUA levels may be at-tributed to the specific uricosuric action of losartan be-cause the final SCr levels were similar in the losartan andatenolol groups. Therefore, statins and losartan probablylower SUA levels via different mechanisms, thus raisingthe possibility of an additive effect.

The use of drugs that lower SUA levels may signifi-cantly lower the risk of vascular events [5].

STELLA S. DASKALOPOULOU, VASILIOS G. ATHYROS,MOSES ELISAF, and DIMITRI MIKHAILIDIS

London, United Kingdom; Thessaloniki and Ioannina, Greece

Correspondence to DP Mikhailidis, MD, FASA, FFPM, FRCP, FRC-Path, Academic Head of Department, Department of Clinical Biochem-istry (Vascular Disease Prevention Clinics), Royal Free Hospital campus,Royal Free and University College Medical School, Pond Street, LondonNW3 2QG, UK.E-mail: [email protected]

REFERENCES

1. HOIEGGEN A, ALDERMAN MH, KJELDSEN SE, et al: LIFE Study Group:The impact of serum uric acid on cardiovascular outcomes in theLIFE study. Kidney Int 65:1041–1049, 2004

2. ATHYROS VG, ELISAF M, PAPAGEORGIOU AA, et al: GREACE StudyCollaborative Group: Effect of statins versus untreated dyslipidemiaon serum uric acid levels in patients with coronary heart disease. Asubgroup analysis of the GREek Atorvastatin and Coronary-heart-disease Evaluation (GREACE) study. Am J Kidney Dis 43:589–599,2004

3. YOUSSEF F, GUPTA P, SEIFALIAN AM, et al: The effect of short-termtreatment with simvastatin on renal function in patients with periph-eral arterial disease. Angiology 55:53–62, 2003

4. HEART PROTECTION STUDY COLLABORATIVE GROUP: MRC/BHF HeartProtection Study of cholesterol-lowering with simvastatin in 5963people with diabetes: A randomised placebo-controlled trial. Lancet361:2005–2016, 2003

5. DASKALOPOULOU SS, MIKHAILIDIS DP, ATHYROS VG, et al: Fenofibrateand losartan. Ann Rheum Dis 63:469–470, 2004

Heterozygous NPHS1 orNPHS2 mutations inresponsive nephrotic syndromeand the multifactorial origin ofproteinuria

To the Editor: Lahdenkari et al [1] described heterozy-gous NPHS1 mutations in 2 out of 25 patients with min-imal change nephropathy. Compound heterozygous mu-tations were present in 2 adults with the same condition.Similar clinical features were already reported in fewnephrotic patients with heterozygous NPHS2 mutations,and in 1 child with heterozygous NPHS1 mutation asso-ciated with the NPHS2 R229Q variant [2]. The body ofevidence on the topics is growing, and other confirmatorypapers are now appearing [3]. Therefore, while homozy-gous and/or compound heterozygous mutations of these2 genes are associated with strict steroid/cyclosporine re-sistance, patients with a single mutation may respond totherapy and have good long-term outcome. Because in-herited conditions associated with NPHS1 or NPHS2 fol-low a recessive trait, two issues remain unresolved. Oneis why heterozygous carriers develop proteinuria, and thesecond is the response to drugs.

As authors correctly discussed, mutations in othergenes could explain proteinuria as a part of a complexinheritance (point 1). However, a complex inheritancedoes not explain sensitivity to drugs that should betterbe explained on the basis of a multifactorial mechanism(point 2). Data on permeability activity in patients withfocal glomerulosclerosis and mutations of NPHS2 sup-ports this possibility [4]. In spite of some unresolved prob-lems, it seems worth mentioning the clinical impact ofthese observations. They also suggest the implication ofnongenetic factors and open up to a multifactorial genesisof proteinuria.

GIANLUCA CARIDI, ROBERTA BERTELLI, FRANCESCO PERFUMO,and GIAN MARCO GHIGGERI

Genova, Italy

Correspondence to Gian Marco Ghiggeri, Laboratorio di Fisiopa-tologia dell’Uremia and Divisione di Nefrologia, Dialisi e Taapianto,Istituto Giannina Gaslini, Largo G. Gaslini 5, I-16147 Genova, Italy.E-mail: [email protected]

REFERENCES

1. LAHDENKARI AT, KESTILA M, HOLMBERG C, et al: Nephrin gene(NPHS1) in patients with minimal change nephrotic syndrome(MCNS). Kidney Int 65:1856–1863, 2004


2. CARIDI G, BERTELLI R, DI DUCA M, et al: Broadening the spectrumof diseases related to podocin mutations. J Am Soc Nephrol 14:1278–1286, 2003

3. RUF RG, LICHTENBERGER A, KARLE SM, et al: Patients with mutationsin NPHS2 (podocin) do not respond to standard steroid treatmentof nephrotic syndrome. J Am Soc Nephrol 15:722–732, 2004

4. CARRARO M, CARIDI G, BRUSCHI M, et al: Serum glomerular per-meability activity in patients with podocin mutations (NPHS2) andsteroid-resistant nephrotic syndrome. J Am Soc Nephrol 13:1946–1952, 2002

Different meanings of“glomerular tip lesion”

To the Editor: The term “glomerular tip lesion” [1] hasbeen used with three meanings:

(1) Our original description was in nephrotic patientswith structural changes at the tubular origin in glomerulithat were otherwise normal. The clinical course resem-bled that of minimal change nephropathy [2].

(2) We later realized that such changes were a com-mon finding in many disorders, such as membranousnephropathy [3]. These tip changes were not themselves adisease, but could only be interpreted by consideration ofthe rest of the glomerulus. Others have applied the term“glomerular tip lesion” to these changes, irrespective ofthe associated condition.

(3) We also reported tip changes in glomeruli show-ing mesangial hypercellularity in the nephrotic syndrome,sometimes with clinical progression [3, 4]. The ones whodid badly developed segmental sclerosis, correspondingto many descriptions of “focal segmental glomeruloscle-rosis.” We called the early stage early classical focal seg-mental glomerulosclerosis [4].

‘Glomerular tip lesion,’ as defined by D’Agati et al [1],also called “the tip variant of focal segmental glomeru-losclerosis,” clearly includes our original definition, andexcludes tip changes in conditions such as membranousnephropathy. Their definition allows mesangial hypercel-lularity, and some of their patients may correspond toearly classical focal segmental glomerulosclerosis. Differ-entiation between normal mesangium and mild mesan-gial hypercellularity is arbitrary, but at the moment thereis no other satisfactory test to identify those who mayprogress.

We agree with D’Agati et al that most patients with the“glomerular tip lesion” by their definition, have steroidresponsive nephrotic syndrome and a good prognosis.

A.J. HOWIE and D. ADU

Birmingham, United Kingdom

Correspondence to A.J. Howie, Department of Pathology, Universityof Birmingham, Birmingham B15 2TT, United Kingdom.E-mail: [email protected]

REFERENCES

1. STOKES MB, MARKOWITZ GS, LIN A, et al: Glomerular tip lesion:A distinct entity within the minimal change disease/focal segmentalglomerulosclerosis spectrum. Kidney Int 65:1690–1702, 2004

2. HOWIE AJ, BREWER DB: The glomerular tip lesion: A previouslyundescribed type of segmental glomerular abnormality. J Pathol142:205–220, 1984

3. HOWIE AJ: Changes at the glomerular tip: A feature of membra-nous nephropathy and other disorders associated with proteinuria. JPathol 150:13–20, 1986

4. HOWIE AJ, LEE SJ, GREEN NJ, et al: Different clinicopathologicaltypes of segmental sclerosing glomerular lesions in adults. NephrolDial Transplant 8:590–599, 1993


We agree with Dr. Howie that glomerular tip lesion(GTL) may occur as a primary form in patients with idio-pathic nephrotic syndrome, or may develop secondarilyin association with heavy proteinuria in diverse glomeru-lar diseases [1]. We limited our study to primary GTL [2].In our series, mesangial hypercellularity did not predictoutcome, defined as remission status at last follow-up.None of our cases of primary GTL showed more thanmild mesangial hypercellularity, which was detected in47% of GTL biopsies. Notably, the single GTL case thatprogressed to end-stage renal disease lacked mesangialhypercellularity.

Dr. Howie has observed that those cases with “struc-tural changes at the tubular origin in glomeruli that wereotherwise normal” had a “clinical course resembling thatof minimal change disease.” Our data have shown, forthe first time, that routinely processed renal biopsies withGTL frequently contained glomeruli with segmental le-sions at other sites (peripheral or indeterminate, but notperihilar), and most of these cases similarly followed a be-nign course. Importantly, the segmental lesions were pre-dominantly cellular (81%), rather than sclerosing. Therewas no significant difference in remission status whencases with GTL alone (26% of cases) were compared withcases of GTL with segmental lesions at other glomerularlocations. Remission rate for GTL was better than foridiopathic focal segmental sclerosis controls, but not asgood as that reported for adult minimal change disease.For these reasons, we envision GTL as occupying an in-termediate position, morphologically and clinically, in theminimal change disease/focal segmental glomeruloscle-rosis spectrum [3].

MICHAEL B. STOKES, GLEN S. MARKOWITZ,and VIVETTE D. D’AGATI

New York, NY


Correspondence to Michael B. Stokes, Department of Pathology,Columbia-Presbyterian Medical Center, New York, NY.E-mail: [email protected]

REFERENCES

1. HOWIE A: Different meanings of ‘glomerular tip lesion.’ Kidney Int66:1716–1717, 2004

2. STOKES MB, MARKOWITZ GS, LIN A, et al: Glomerular tip le-sion: A distinct entity within the minimal change disease/focalsegmental glomerulosclerosis spectrum. Kidney Int 65:1690–1702,2004

3. D’AGATI VD, FOGO AB, BRUIJN JA, JENNETTE JC: Pathologic classi-fication of focal segmental glomerulosclerosis: A working proposal.Am J Kidney Dis 43:368–382, 2004

How can we be sure that renaldysfunction after coronaryangiography is just explainedby contrast nephropathy?

To the Editor: A review has been performed on thevery hot topic of the role of N-acetylcysteine (NAC) oncontrast nephropathy (CN) [1], of great clinical impact asa result of the CN-linked role in worsening prognosis andincreasing costs. The authors wrote that they “assess theefficacy of NAC for preventing CN after . . . intravenouscontrast media,” and concluded that “NAC may reducethe incidence of increased creatinine after administrationof intravenous contrast, but this was of borderline statis-tical significance.” However, both sentences are wrongand misleading, as are similar conclusions reached byanother meta-analysis [2], because all 15 [1] or 16 [2]reviewed papers regarded coronary angiography (CA),except one [3]. First of all, to perform CA, contrast me-dia are introduced into the arterial vascular bed, andnot intravenously, as when performing computed to-mography (CT). Second, mechanisms of renal dysfunc-tion after CA are not only caused by CN, but also byother causes such as, for instance, cholesterol crystalembolization.

We suggest that: (1) further studies be analyzed byseparating prevention strategies for CT from those forCA; (2) for CA, attempts will be made to dissect othercauses of renal damage by looking for the blue toessyndrome or eosinophilia, in order to exclude choles-terol embolization; (3) even urea increase was consid-ered as end point, to avoid the possibility that creatininechanges might be resulting simply from a direct effect ofNAC [4].

The only quoted paper regarding the use of NAC be-fore performing CT did demonstrate a protection, by alsousing urea values as end point [3].

CATERINA CANAVESE, FABIO MORRA, VERONICA MORELLINI,ELISA LAZZARICH, MADDALENA BRUSTIA,

MARIO BO, and PIERO STRATTA

Novara and Torino, Italy

Correspondence to Caterina Canavese, Transplantation and Nephrol-ogy, Department of Nephro-Urology, Amedeo Avogadro University, No-vara, Ospedale Maggiore della Carita, Corso Mazzini 18, 28100 Novara,Italy.E-mail: [email protected]

REFERENCES

1. PANNU N, MANNS B, LEE H, TONELLI M: Systematic review of theimpact of N-acetylcysteine on contrast nephropathy. Kidney Int65:1366–1374, 2004

2. KSHIRSAGAR AV, POOLE C, MOTTL A, et al: N-acetylcysteine for theprevention of radiocontrast induced nephropathy: A meta-analysisof prospective controlled trials. J Am Soc Nephrol 15:761–769,2004

3. TEPEL M, VAN DER GIET M, SCHWARZFELD C, et al: Prevention ofradiographic-contrast-agent-induced reductions in renal function byacetylcysteine. N Engl J Med 343:180–184, 2000

4. HOFFMANN U, FISCHEREDER M, KRUGER B, et al: The value ofN-acetylcysteine in the prevention of radiocontrast agent-inducednephropathy seems questionable. J Am Soc Nephrol 15:407–410,2004


We thank Dr. Canavese et al for their letter. Weagree that the term “intravenous contrast administra-tion” may have been misleading. Perhaps the term“parenteral contrast administration” would have beenpreferable.

As we mentioned in our article, atheroemboli mightexplain why NAC seemed to be less efficacious thancomputed tomography in the context of coronaryangiography. For this reason we performed subgroupanalysis including only trials of patients undergoing coro-nary angiography. Although “looking for blue toes oreosinophilia” has theoretical appeal, we are uncertainhow helpful this would be, because such findings maytake weeks to appear [1], and clinically silent cholesterolembolization after invasive procedures appears to becommon [2].

The suggestion to use serum urea rather than creati-nine as an outcome measure seems to miss one of themain points of our article—that data on costs or clini-cally relevant outcomes such as death or hospitalization(and not surrogates such as estimated kidney function)are needed.

After reading their letter carefully, we are uncertainwhether Dr. Canavese et al feel that our conclusions are


“wrong and misleading” because they believe NAC is ef-ficacious for prevention of contrast nephropathy, or thecontrary. To clarify our position, like Ksihirsagar et al [3],we believe that available data are insufficient to recom-mend NAC as the standard of care for prevention of con-trast nephropathy, and that additional randomized trialsare required.

MARCELLO TONELLI, BRADEN MANNS,and NEESH PANNU

Edmonton and Calgary, Alberta, Canada

Correspondence to Marcello Tonelli, Division of Nephrology &Transplantation Immunology, CSB 11-107, Edmonton, Alberta, CanadaT6G 2G3.E-mail: [email protected]

REFERENCES

1. MODI KS, RAO VK: Atheroembolic renal disease. J Am Soc Nephrol12:1781–1787, 2001

2. RAMIREZ G, O’NEILL WM, JR., LAMBERT R, et al: Cholesterolembolization: A complication of angiography. Arch Intern Med138:1430–1432, 1978

3. KSHIRSAGAR AV, POOLE C, MOTTL A, et al: N-acetylcysteine for theprevention of radiocontrast induced nephropathy: A meta-analysisof prospective controlled trials. J Am Soc Nephrol 15:761–769, 2004

Heterozygous NPHS1 or NPHS2 mutations in responsive nephrotic syndrome and the multifactorial origin...

Documents

Transcript of Heterozygous NPHS1 or NPHS2 mutations in responsive nephrotic syndrome and the multifactorial origin...