Minimal Change Nephrotic Syndrome in Children: Deaths During the First 5 to 15 Years' Observation

(1)

Minimal

Change

Nephrotic

Syndrome

in

Children:

Deaths

During

the First

5 to 15 Years’

Observation

Report

of the International

Study

of Kidney

Disease

in Children*

ABSTRACT. Of 521 children with a previously untreated nephrotic syndrome, as defined by proteinuria 40 mgi h/rn2 and serum albumin 2.5 g/dL, entering the Inter-national Study of Kidney Disease in Children between January 1967 and April 1976, 389 showed minimal changes on renal biopsy. Of these, seven boys and three girls died, all before July 1972. Infection was the cause of

death in six patients. One child died of dural sinus

thrombosis, one died as a result of cardiorespiratory

failure following salt-poor albumin infusion, and another died from chronic renal failure due to focal and segniental glomerutosclerosis not apparent on initial biopsy. The

mode of death in the remaining child was uncertain.

Analysis according to histopathologic subgroups of mm-imal change disease showed no statistically significant differences in the incidence of deaths, although mesangial abnormalities and tubular atrophy were associated with higher mortality than nil disease or focal glomerular

obsolescence. Nine ofthe ten children who died had either failed to respond to initial prednisone therapy (initial nonresponders, n = 5), or responded but relapsed during the initial 8 weeks of treatment (early relapser, n = 4),

even though only one quarter of the total sample were nonresponders or early relapsers (P < .0005). Nearly one fifth of all initial nonresponders with minimal change

nephrotic syndrome died. Thus the pattern of response

to initial steroid therapy in patients with minimal change

nephrotic syndrome may have prognostic significance.

Pediatrics 1984;73:497-501; nephrotic syndrome, renal

disease, sepsis, dural sinus thrombosis.

Reports appearing some 15 years after the

intro-duction of corticostenoids in the treatment of chit-dren with the nephrotic syndrome have docu-mented a substantial improvement in survival rates.”2 Statements made by these authors

regard-ing the prognostic significance of “nephnitic”

fea-tures, eg, hypertension, hematunia, and azotemia, when considered in the light of present knowledge regarding the response to steroid therapy of

pa-tients with different forms of glomerular disease,

indicate that the increased survival occurred pni-manly in children. with minimal change nephrotic syndrome (MCNS).

However, there is little factual information on the incidence and causes of mortality in children with .MCNS, treated with contemporary regimens, in whom a histopathologic diagnosis was made by means of renal biopsy at the time of diagnosis. White et al3 reported four deaths among 95 patients (4.2%) followed up for 4 to 7 years, and Habib and Kleinknecht4 reported 14 deaths among 209 chil-dren (6.7%) observed for a period of 1 to 10 years. Only four of the 18 deaths in these two series were due to chronic renal failure. Likewise, Habib et a15 found only three cases of MCNS among 2,052 chil-dren suffering from chronic renal failure.

Thus progression to chronic renal failure is a rare event among children whose condition was initially diagnosed as MCNS, and this implies either varia-tion of the underlying pathologic process on an

Received for publication Feb 14, 1983; accepted May 20, 1983.

* Participants in the ISKDC who contributed to the present study: Central Office (New York)-H. L. Barnett and C. M. Edelmann, Jr (Directors); I. Greifer (Associate Director); D. I. Goldsmith and A. Spitzer (Directors of Coordinating Center); G. Laddomada (Project Administrator); Regional Coordinators: I. B. Houston, R. H. Kuijten, and L. B. Travis; Directors of Participating Centers: B. S. Arant, S. Roy III (Memphis), G. C. Arneil (Glasgow), J.C.M. Chan (Washington, DC), A. Fanconi (Winterthur), G. Gordillo-P. (Mexico City I), A. B. Gruskin (Philadelphia), I. B. Houston (Manchester), 0. Kobayashi (Nh-gata), R. H. Kuijten and H.A.W.M. Tiddens (Utrecht, Amster-dam), E. Leumann (Zurich), J. F. Lewy, M. Kaplan (New York-Cornell), M. I. McVicar (New York-Downstate), J.-G. Mongeau

(Montreal), M. A. Nash, M. J. Schoeneman, R. Weiss (New York-Albert Einstein), 0. Oetliker (Bern), D. Santos (Mexico City II), K. S. Sch#{228}rer(Heidelberg), H. Stark (Petach Tikva, Israel), J. Strauss (Miami), L. B. Travis (Galveston), Y. C. Tsao (Hong Kong), N. Haliman and J. Vilska (Helsinki), C. D. West (Cincinnati), and R.H.R. White and M. Winterborn (Binning-ham); Consultants: J. Bernstein, J. Churg, R. Habib, and R.H.R. White (Pathology); J. S. Cameron, R. H. Kuijten, E. Lewis, and

C.D. West (Immunology); J. Fertig, K. Freeman, and K.Suffivan (Biostatistics).

(2)

unrepresentative initial renal biopsy. Deaths from

causes other than renal failure are of greater

con-cern to the practicing pediatrician, especially

be-cause they may be preventable. The present study

examines the incidence and causes of death occur-ring during the first 5 to 15 years after diagnosis in a large series of children with histologically proven

MCNS in a prospective survey conducted by the

International Study of Kidney Disease in Children (ISKDC).

PATIENTS AND METHODS

Selection

of Patients

Between January 1967 and April 1976, 521 chit-dren with the nephrotic syndrome were admitted to the study from 24 participating clinics. Patients had to satisfy the following criteria: (1) heavy pro-teinunia (40 mg/h/m2) determined quantitatively

on an overnight urine collection; (2)

hypoalbumin-emia (2.5 g/dL); (3) age 12 weeks and 16 years at the time of diagnosis; (4) no prior treatment with

steroids or other cytotoxic or immunosuppressive

agents; and (5) no evidence of underlying systemic

disease or exposure to agents known to be

associ-ated with the nephrotic syndrome. Informed

con-sent concerning participation in the study was

ob-tamed.

Clinical and Histologic Methods

Methods used for data collection, coding and reporting, clinical and laboratory examinations, standardization of age and sex-dependent variables, and histologic preparation and interpretation of renal biopsy material have been described

previ-ously.

Clinical Definitions

Initial prednisone treatment was 60 mg/24 h/m2 (maximum dosage 80 mg/24 h) in divided doses for

4 weeks, followed by 40 mg/24 h/rn2 in divided

doses, three consecutive days out of seven for 4

weeks. Response was defined as a reduction in rate of urinary excretion of protein to <4 mg/h/m2 (Albustix 0 to trace) for three consecutive days. Relapse was defined as a reappearance of protein-unia 540 mg/h/m2 (Albustix ++ or greater) for

three consecutive days. “Initial responder” was any patient who responded during the 8 weeks of initial

prednisone treatment. “Early relapser” was any

initial responder who relapsed during the 8 weeks of the initial prednisone treatment. “Initial nonre-sponder” was any patient who failed to respond during the initial 8 weeks of predisone therapy.

RESULTS

Of

the 521 children with primary nephrotic

syn-drome, 389 (75%) had MCNS. (The small differ-ences between the number of patients with MCNS in this report and the numbers in references 9 and

10 resulted from reclassification of some patients.)

There were ten deaths among the 389 children

(2.6%). All ten patients who died had entered the

study between August 1967 and July 1972, during which period more than 85% of the entire group

had been enrolled. The patients were

proportion-ately distributed throughout the geographic regions

represented by the 24 participating centers.

The pertinent clinical features of each of the ten

children who died are shown in Table 1. The ratio of boys to girls was 7:3. The age at entry ranged from 1 year 6 months to 14 years 9 months, with a

mean of 4 years 5 months and a median of 3 years

1 month. The estimated duration of. illness before treatment ranged from six days to 6 weeks (mean 1 month). None of these characteristics of the ten

children who died differed significantly from those

in the 379 patients who survived 5 years or longer.

The period of illness from initial treatment to death ranged from 8 days to 20.7 months, with a

mean of 10.6 months. Excluding one child who died

on the ninth day of treatment, the mean interval to death was 11.7 months.

In Table 2, the initial course of the ten children

who died is compared with that of the total of 373 patients for whom adequate data on the initial response to steroids are available. Two of the non-responders who died did not have complete urinary

data at the end of the initial prednisone treatment but were included in the analysis because one died on the ninth day of treatment and the other was determined to be a nonresponder by the pediatni-cian in charge and accordingly was allocated to a group given a trial of cyclophosphamide therapy.

None of the 16 patients excluded because of incom-plete urinary data died. It can be seen in Table 2 that nine of the ten children who died were

nonre-sponders or early relapsers.

The incidence of death among the five histologic

variants of MCNS is shown in Table 3. Although

the numbers are small, there is a trend toward increased fatalities in patients having mesangial and tubular changes, compared with those showing nil disease or focal glomerular obsolescence.

The causes of death are shown in Table 1. Serious infection occurred in seven patients and led to death in six. In the seventh child (No. 3105), pneu-mococcal peritonitis complicated a late relapse but

(3)

TABLE 1. Clinical Features of the Ten Children with Minimal Change Nephrotic Syndrome Who Died* Series No. Age at Entry (yr/mo) Sex Histologic Subgroup Initial Re-sponse to Prednisone Subsequent Course and Treatment Interval, Treatment to Death

Cause of Death and Necropsy Diagnosis

1206 3/11 F Nil NR LRes

Rel

Pred

7.3 mo Septicemia, active NS; necropsy: septicemia

(paracoli bacteria)

1622 4/2 M MMH ERe1 Rel

Pred

9.2 mo Chronic renal failure; necropsy: FSGS

2412 2/8 M FGO ERel Rel

Pred

3.6 mo Active NS; autopsy: multiple dural sinus thromboses

3105 1/10 F MMT Res Rel

Prod

18.8 mo Treated peritonitis. Vascular overload from IV albumin; cardiorespiratory arrest; autopsy: no active peritonitis

0434 2/11 M FGO ERel Rel

Pred

20.7 mo Cause uncertain; active NS; congestive cardiac

failure; no autopsy

2426t 6/9 M Nil NR NR

Pred

8 d Peritonitis and Escherichia coli septicemia; ac-tive NS; no autopsy

2510 14/9 M

FTC

NR Progression

to uremia

Pred Cyclo

16.5 mo Gram negative septicemia and disseminated intravascular coagulation; autopsy: FSGS

0628 2/4 M Nil ERel LNR

Pred; Aza

Cyclo

12.8 mo Active NS; sepsis; disseminated intravascular coagulation; autopsy: FSGS

0832 1/6 F FTC NR LRes

Rel

Pred

14.0 mo E coli peritonitis and septicemia; shock, car-diac arrest; no autopsy

3413 3/3 M Nil NR NR

Pred Cyclo

2.6 mo Pneumococcal peritonitis; pancreatitis on

au-topsy

* Abbreviations used are: Nil, nil disease; MMH, mild mesangial hypercellularity; FGO, focal glomerular obsolescence;

MMT, mild mesangial thickening; FTC, focal tubular changes; NS, nephrotic syndrome; FSGS, focal segmental glomerulosclerosis; NR, nonresponder; ERel, early relapser; Res, responder, LRes, late responder; Rel, relapser; Pred, prednisone; LNR, late nonresponder; Cyclo, cyclophosphamide; Aza, azathioprine.

t

Patient No. 2426 died after nine days of initial prednisone treatment. Classification of this patient is somewhat arbitrary. If he had lived, he might have been a responder. However, because he did not respond during initial therapy, we have chosen to classify him as a nonresponder.

TABLE 2. Distribution

Response

of Deaths According to Initial

Response to Initial Prednisone Treatment Total No. of Patients Deaths No. % P Initial responders

Not early relapser Early relapser Initial nonresponsers 283 63 27 1 0.4 4 6.3 5 18.5 <.005 <.0005

Total 373t 10 2.7

*P values for comparison of early relapsers or initial

nonresponders with initial responders who were not early

relapsers. Terminology defined in text under “Clinical

Definitions.”

t In 16 children, data concerning initial response to pred-nisone were inadequate to characterize the initial re-sponse.

chronic renal failure and had advanced focal and segmental glomerulosclerosis (FSGS) at necropsy. Details of the mode of death in patient No. 0434 unfortunately are lacking; this child, a frequent relapser, was admitted to a hospital remote from the participating clinic and was reported to have

TABLE 3. Mortality Amo

Minimal Change Nephrotic

ng Histologic Subgroups of Syndrome

Biopsy Diagnosis No. of Deaths

at Onset Patients

No. %

Nil 219 4 1.8

Mild mesangial hypercel- 27 1 3.7

lulanity

Mild mesangial thicken- 16 1 6.3 ing

Focal glomerular obso- 98 2 2.0

lescence

Focal tubular changes 29 2 6.9

Total 389 10 2.6

died with massive edema, cardiac failure.

DISCUSSION

ascites, and congestive

(4)

complications of treatment continue to occur. The mortality of 2.6% in the present series is signifi-cantly lower (P < .05) than those reported by White

et al3 and Habib and Kleinknecht4, although the range of causes is similar.

Of

great relevance is the high incidence of infec-tion as a cause of death. Infection was an important cause of death in the presteroid era,” especially during the first 2 years after diagnosis,’2 and the advent of antibiotics evidently yielded little im-provement in the overall mortality from the ne-phrotic syndrome.”2 Children with the nephrotic syndrome have been found to have T- and B-cell abnormalities as well as abnormalities of the corn-plement system.’3 The conventional assumption would be that increased susceptibility to infection

is secondary to active disease, resistant to prolonged

steroid therapy, and perhaps associated with irn-munosuppression. Only 27 (7.2%) of the 373 chit-dren were initial nonnesponders, and yet five of them (18.5%) died of infection, accounting for half the deaths in this series (Table 2). Four of the ten deaths occurred among the 63 children who were early relapsers and who constituted less than 20% of the total group with MCNS. Relapses in these patients were not more prolonged than in other relapsing patients, with the exception of patient

No. 0628, who became a late nonresponder. The

only initial responder (No. 3105) who died and was not an early relapser did not succumb to hen infec-tion but died as a complication of therapy.

Although it might be expected that those patients

who respond poorly to prednisone initially will con-tinue to do so and thus be at higher risk of toxicity and even death, the strength of the association is striking. It may be relevant that in a previous study of the ISKDC,8 children having mesangial and tu-bular changes on renal biopsy had an increased rate of nonresponse to initial treatment with predni-sone, a finding that may be reflected in the trend noted in the current report of increased fatalities in the same categories of patients. It could be spec-ulated that the increased vulnerability to infection and the resistance to steroid therapy may have a common biologic origin.

The enhanced coagulability of the blood of adults with MCNS has been fully discussed by Cameron et al.’4 In addition to increased blood levels of fibninogen and factor VIII and increased platelet aggregation, there is a decrease of fibninolytic

activ-ity, while the blood viscosity is increased,

presum-ably as a result of hemoconcentration. Although peripheral venous thrombosis, with or without

put-monary embolus, is a well-recognized complication

of the nephrotic syndrome, dural sinus thrombosis occurs rarely.’5

The intravenous infusion of salt-poor albumin is

a useful form of supportive treatment in nephrotic

patients who have an inadequate response to steroid

therapy and who are extremely hypovolemic as a result of persistent, heavy proteinuria. Its value has

been discussed by Chamberlain et al’6 and by

Day-ison et al,’7 but its potential dangers appear to have

received less attention. Although pneumococcal

peritonitis in patient No. 3105 was successfully

treated, the child’s death was precipitated by

infu-sion of albumin, which presumably caused too rapid

expansion of plasma volume, with cardiac decom-pensation, pulmonary edema, and, finally, cardi-orespiratory arrest.

The one patient who died from chronic renal failure had extensive FSGS at necropsy; careful review of the initial biopsy disclosed no segmental

lesions. It is not pertinent in this report to resurrect

previous debates over whether FSGS evolves out of genuine MCNS’8 or is a primary disorder in which an unrepresentative biopsy specimen might miss

early segmental lesions.’9 It is an uncommon se-quence of events in MCNS, one that is unlikely to be influenced by treatment. It is of interest, how-ever, that two other patients who did not have renal failure also had lesions of FSGS at necropsy. The

possible significance of this is not known.

IMPLICATIONS

These results have clinical and, perhaps, biologic implications. They suggest that the prognosis in early relapsers and in early nonnesponders is less good than generally believed and indicate the need for special care and increased vigilance for signs of infection and other complications in these small groups of patients.

SUMMARY

Death occurred in ten of 389 children with mm-imal change nephrotic syndrome followed for 5 to

15 years from onset of disease. All deaths occurred

during the initial 2 years of illness. The cause of death in six children was overwhelming infection. One child died from each of the following: dural

sinus thrombosis, cardiorespiratory failure, and

chronic renal insufficiency. The cause of death in the tenth child was uncertain.

Although only one fourth of the total group were initial nonresponders or early relapsers, nine of the ten children who died fell into these categories,

suggesting that the pattern of response to initial

steroid therapy in patients with MCNS may have prognostic significance.

ACKNOWLEDGMENTS

(5)

Health Research Grant 1 ROl AM18234, National Kid-ney Foundation of New York, Kidney Disease Institute of the State of New York, the John Rath Foundation, National Kidney Research Foundation (United

King-dom), and the Kidney Foundation of the Netherlands.

REFERENCES

1. Cornfeld D, Schwartz MW: Nephrosis: A long-term study of children treated with corticosteroids. J Pediatr 1966;68:

507-515

2. Arneil GC, Lam CN: Long-term assessment of steroid ther-apy in childhood nephrosis. Lancet 1966;2:819-821

3. White RHR, Glasgow EF, Mills R,J: Clinicopathological study of nephrotic syndrome in childhood. Lancet 1970; 1:1353-1359

4. Habib R, Kleinknecht C: The primary nephrotic syndrome of childhood: Classification and clinicopathologic study of 406 cases. Pathol Annu 1971;6:417-474

5. Habib R, Broyer M, Benmaiz H: Chronic renal failure in children: Causes, rate of deterioration, and survival data.

Nephron 1973;11:209-220

6. International Study of Kidney Disease in Children: Pro-spective, controlled trial of cyclophosphaniide therapy in children with the nephrotic syndrome. Lancet 1974;2: 423-427

7. Churg J, Habib R, White RHR: Pathology of the nephrotic syndrome in children: A report for the International Study of Kidney Disease in Children. Lancet 1970;1:1299-1302 8. International Study of Kidney Disease in Children: Primary

nephrotic syndrome in children: Clinical significance of histopathologic variants of minimal change and of diffuse

mesangial hypercellularity. Kidney

mt

1981;20:765-771 9. International Study of Kidney Disease in Children:

Predic-tion of histopathology from clinical and laboratory charac-teristics at time of diagnosis. Kidney

mt

1978;13:159-165 10. International Study of Kidney Disease in Children: The

primary nephrotic syndrome in children: Identification of patients with minimal change nephrotic syndrome from initial response to prednisone. J Pediatr 1981;98:561-564 11. Lawson D, Moncrieff A, Payne WW: Forty years of

ne-phrosis in childhood. Arch Dis Child 1960;35:115-126 12. Arneil GC: One hundred sixty-four children with nephrosis.

Lancet 1961;2:1103-1110

13. Bellow M, Kennedy TL III, Gaudio KM, et al: Serum hemolytic factor D values in children with steroid-respon-sive idiopathic nephrotic syndrome. J Pediatr 1982;100: 192-196

14. Cameron JS, Turner DR, Ogg CS, et al: The nephrotic syndrome in adults with ‘minimal change’ glomerular le-sions. QJ Med 1974;43:461-488

15. Lau SO, Bock GH, Edson JR, et al: Sagittal sinus thrombosis in the nephrotic syndrome. J Pediatr 1980;97:948-950 16. Chamberlain MJ, Pringle A, Wrong OM: Oliguric renal

failure in the nephrotic syndrome. Q J Med 1966;35: 215-235

17. Davison AM, Lambie AT, Verth AH, et al: Salt-poor human albumin in management of nephrotic syndrome. Br Med J

1974;1:481-484

18. Hayslett JP, Krassner LS, Bensch KG, et al: Progression of “lipoid nephrosis” to renal insufficiency. N Engi J Med 1969;281:181-187

19. White RHR, Glasgow EF, Mills RI: Focal glomerulosclerosis in childhood, in Kincaid-Smith P, Mathew TH, Beck EL (eds): Glomerulonephritis. New York, John Wiley, 1973, pp 231-248

A PUZZLE

FOR

PEKING:

SAME-NAME

CHILDREN

The authorities in northeast China say they are overwhelmed with keeping track of people with the same name, including 4,800 women with the same name, and as a result they have written a guide to naming babies.

In one work unit alone, there are 10 men named Li Wei (“Wei” meaning “great” according to a news report today from the industrial city of Shenyang, 400 miles northeast of Peking. “The result is that people must call them “Big Li Wei,” “Li Wei No. 2,” “Big Eyes Li Wei,” “Long-haired Li Wei,” to avoid confusion,” said a report in Guangming Daily, a newspaper intended for intel-lectuals.

The biggest problem is the popularity of “Shuzhen” (fair and precious), especially with families with the common surname of Li. At last count, the report said, there were more than 4,800 “Li Shuzhens” in Shenyang.

(6)

1984;73;497

Pediatrics

Years' Observation