Atypical hemolytic-uremic syndrome: A comparison with postdiarrheal disease

Atypical hemolytic-uremic syndrome: A comparison with postdiarrheal disease

Atypical hemolytic-uremic syndrome: A comparison with postdiarrheal disease Richard L. Siegler, MD, A n d r e w T. Pavia, Mb, Floyd L. Hansen, RN, Rya...

632KB Sizes 0 Downloads 81 Views

Atypical hemolytic-uremic syndrome: A comparison with postdiarrheal disease Richard L. Siegler, MD, A n d r e w T. Pavia, Mb, Floyd L. Hansen, RN, Ryan D. Christofferson, a n d Joshua B. C o o k From the Depaffment of Pediatrics,DMsionsof Nephrology and Hypertensionand [nfectious Diseases, Universityof Utah School of Medicine, Salt Lake City, Utah

Objectives: To compare the epidemiologic, laboratory, clinical, and outcome variables of atypical (nondiarrheal) hemolytic-uremic syndrome with those of classic postdiarrheal disease. Methods: A 24-year retrospective review of 28 episodes of atypical HUS that occurred in 22 children compared with 266 episodes of typical postdiarrheal disease in 265 children treated during the same period. Results: Of the 294 episodes of HUS, 9.5% were atypical (nondiarrheal), and 18% of the patients in the atypical disease group had recurrences. Prodromal features (other than the presence or absence of diarrhea) were similar between the groups. White blood cell count and serum creatinine concentration on admission to the hospital and most abnormal blood urea nitrogen values during hospitalization were significantly Iower (p = 0.02) in the patients with atypical HUS. Oliguria, anuria, and the need for dialysis were also less common (p = 0.02) in the atypical disease group. There were no deaths in the subset of patients with atypical disease; 3.4% of the patients in the typical disease group died. Although there were no statistically significant differences in the incidence of end-stage renal disease between the atypical and typical disease groups, two of the four patients with atypical disease who had recurrences also had end-stage renal disease. There were no significant differences in chronic renal sequelae between the groups one or more years after HUS. Conclusions: In contrast to reports from most other regions, patients with atypical disease in our area of the western United States have milder acute nephropathy and, with the exception of those with recurrence, do not experience worse outcomes. (J PEDIATR1996; 128:505-1 I) Approximately 90% of all cases of childhood hemolyticuremic syndrome in the United States follow an episode of diarrhea that is usually bloody) Although there are numerous reports describing postdiarrheal HUS, there is limited information on nondiarrheal (atypical) cases. With the exception of a recent report from London, England, that focused on nondiarrheal HUS,2 details of the acute phase and follow-up information have usually been scanty. Even so, the Submitted for publication March 16, 1995; accepted Dec. 13, 1995. Reprint requests: Richard L. Siegler, MD, Department of Pediatrics, Division of Nephrology and Hypertension, 50 North Medical Dr., Salt Lake City, UT 84132. Copyright © 1996 by Mosby-Year Book, Inc. 0022-3476/96/$5.00 +0 9/20171612

conclusion of most reports 211 has been that nondiarrheal HUS is associated with greater morbidity and a higher mortality rate than typical HUS. The intermountain region of Utah and surrounding states is an endemic region for HUS; Utah has one of the highest ESRD GFR HUS

End-stagerenal disease Glomerularfiltration rate Hemolytic-uremicsyndrome

reported incidences in the United States (1.4 cases/100,000 people <18 years of age). 1 Because there is a paucity of reported experience from North America, we conducted a 24year (1970 to 1993) retrospective review of our experience with 28 episodes of atypical HUS in 22 children. These ep-

505

50 6

Siegler et al.

The Journal of Pediatrics April 1996

J D- HUS Patients J 22

D- non-recurrent

I D- recurrent 4

I Eligible for Follow-up Study 17 (geR)* I

J Eligible for Follow-up Study

18

I I Lost to Follow-up

(75g)*

I

I

I

5 (29x)

I Abnœmal ** L 6 (»0x) I

I Hyperten=ion ] on~

2 (17x)

l,

I Iow Ccr. Proteinurlo. hypeH:ensionI 1 (~»x)

-

I Proteinuria

& hypertension l, 1 (s~.)

I

Proteinuria 2

only 07g)

* > 1 yeor post HUS ** 0ne addltional patient < 1 yeor »ost HUS had Ccr of 28 tal/min Fig. 1. Outcome for 22 patients with atypical HUS who experienced a total of 28 episodes of HUS. Ccr, Creatinine clearance rate; D-, atypical HUS. *Follow-up more than I year after HUS. **One additional patient with follow-up less than 1 year after HUS had a creatirtine clearance rate of 28 ml/min. isodes were compared with 266 episodes of typical HUS in 265 children treated during the same period. METHODS A case of HUS was defined as disease in a child younger than 18 years of age who had an acute onset of microangiopathic hemolytic anemia and acute nephropathy. Cases were not excluded because of a normal platelet count (i.e., 150 × 109/L) because thrombocytopenia cannot be documented in about 6% of cases. 1 If these feamres were preceded by diarrhea within the previous 2 weeks (as recalled by the parents), the case was defined as typical (postdiarrheal); those without a diarrheal prodrome were ¢lassified as atypical (nondiarrheal) cases. Two previously described patients whose atypical HUS was the result of preexisting glomemlar disease were excluded. 12 Patients were residents of Utah or surrounding stares and had been cared for at the University of Utah Health Sciences Center, or its close affiliate, Primary Children's Medical Center. Follow-up clinic visits were scheduled at 3 months after discharge from the hospital. Thereafter return visits were recommended yearly or more offen depending on the severity of the child's acute illness, and the status at the 3-month evaluation.

Selected epidemiologic (sex, age, season, recurrence), prodromal (preceding infections, fever, drugs, vomiting, oliguria, anuria, seizures), laboratory (hematocrit, leukocyte count, platelet count, blood urea nitrogen concentration, serum creatinine concentration), clinical (hypertension, seizures, transfusions, oliguria, anuria, dialysis), and outcome variables (death, end-stage renal disease, chronic hypertension, proteinuria, low glomemlar filtration rate) were compared between the atypical and typical cases. Prodromal and acute-phase variables relative to the second recurrence of atypical disease that occurred in one child shortly after the child received a renal graft were excluded from analysis. Oliguria and anuria dufing the acute phase were defined as <240 ml/m 2 body surface area per day, and <15 ml/day, respectively. Hypertension at follow-up was defined as three or more serial clinic visit blood pressure measurements or the two most recent clinic visit measurements above the 95th percentile for age, 12a or a regimen of blood pressure medication. Low GFR was defined as an estimated creatinine clearance, according to the Schwartz formula, 13 of less than 90 tal/min per 1.73 m 2. Proteinuria was considered to be present if random urine samples showed 1+ or more protein by dipsdck, or if random urine samples had albumin/creatinine ratios ->0.1, or total urinary protein/creatinine ratios

The Joumal of Pediatrics Volume 128, Number 4

Siegler et al.

507

Onset of HUS by Month 70 mDB D+

60

z/// ù //.t z/// ,1"// / / / / ù./// «//.,

50

"/'/,4, i / / I /il1 / I / / c/z/ i / / / ù/// / / / / zile.

co 40 03

0

30

///¢ ig.z/

7/½ 1/.,'1

i / / i //,,./ z///

¢"/A t./// / / / / ~.///

N

1

JGn Feb Mar Apr Moy Jun

I

I

I

Jul Aug Sep Oct Nov Dec

Fig. 2. Seasonal distribution of episodes of atypical (D-) and typical (D+) HUS. Episodes of atypical disease occuned throughout the year; episodes of typical HUS were more frequent during the warmer months.

Clinical Features of HUS Patients

Hypertension

p=0. 19

Seizures TrGnsfusions Oligurie

Death

p=O.13

~"//////////////////////////////////////////////J ////////////////////////////////////A

p=O.02

p=O.02

Anuria Dielysis

p=0.84

2"/////////////////////////////////////////,t

"/////////////////////////////A ~;~

p=O.O1

p=l.0

p=0.07

ESRD

0

I

1

I

I

20

40

60

80

100

Hus Patients (%) Fig. 3. Clinical features in the atypical (D-) and typical (D+) HUS groups. Acute nephropathy was milder in the atypical disease group. End-stage renal disease in the atypicaJ disease group has been limited to the subset with recurrent atypical disease.

->0.2, or 24-hour urine protein excretion rates exceeding 150 mg/1.73 m 2. The significance of differences was tested with nonparametric tests for continuous variables (Mann-Whitney);Mantel-Haenszel and Fisher Exact tests were used for categorical variables. Values ofp <0.05 were considered significant; all p values are two-sided.

RESULTS

Epidemiologic informaüon. Of the 294 episodes of HUS that occurred during the 24 years, 9.5% (28) were atypical. The 28 atypieal episodes occurred in 22 children (Fig. 1). Eighteen (82%) of these 22 children have experienced a single episode dufing a median observation interval of 5 years (range 1 to 16 years).

508

Siegler et al.

Four (18%) subjects (all female patients) have had one or more recurrences of atypical disease (Fig. 1), with a median interval between the first and second episode of 1 year 7 months (range 7 months to 5~ years). In two patients, recurrent atypical HUS followed an initial episode of classic postdiarrheal disease. One of these patients had severe chronic hypertension and progressive azotemia after her initial recurrence of atypical disease that occurred 3 years after an episode of typical HUS. Several years later, but before the beginning of the follow-up component of our smdy, she had a second recurrence that involved her renal graft. The other patient initially experienced had atypical HUS 5 years after an episode of typical HUS. She had a second episode of atypical disease 2 years later. Two female siblings had atypical HUS that was not preceded by typical disease. Both children, as well as their symptom-free mother, had low levels of the third component of complement. Although those with atypical HUS tended to be older (median ages 3.5 years vs 2.0 years), the difference was not statistically significant. There was no clear seasonal variation in the atypical disease cases; the postdiarrheal cases occurred more frequently during the warmer months (Fig. 2). Interestingly, about one third of the atypical episodes occurred during the month of March. Prodromal features. None of the cases of atypical disease followed immunizations, viral exanthems, pneumococcal infections, or exposure to drugs (e.g., cyclosporine). Onset of HUS was preceded by upper respiratory tract infection in 23%, fever in 46%, and vomiting in 81%. There were no statistically significant differences between atypical and typical cases with regard to prevalence of upper respiratory tract infection, fever, vomiting, oliguria, anuria, or seizures during the prodrome (i.e., before diagnosis of HUS). L a b o r a t o r y values. Selected mean laboratory values (on admission and the most abnormal values during hospitalization) for the atypical and typical disease groups were compared. Hematocrit values and platelet counts were similar, but leukocyte counts on admission to the hospital were lower in those with atypical HUS (12.8 x 109/L vs 17.0 X 109]L; p = 0.02). The serum creatinine concentration on admission to the hospital (106.1 prnol/L [1.2 mg/dl] vs 221 pmol/L [2.5 mg/dl]) and the most abn0rmal blood urea nitrogen concentration during hospitalization (27.1 mmol/L [76 mg/dl] vs 42.5 mmol/L [119 mg/dl]) were also significantly lower in the atypical disease group (p = 0.02). Clinieal eourse. The incidence of hypertension and the need for blood transfusions during the acute phase of the illness were similar between the groups (Fig. 3). Patients in the atypical disease group, however, were less than half as likely to experience seizures (7.7% vs 19%), but the difference was not statistically significant (p = 0.19). Oliguria was less frequent in the atypical disease group (38% vs 58%; p = 0.02),

The Journal of Pediatrics April 1996

as was anuria (16% vs 41%; p = 0.02). Therefore patients in the atypical disease group were less likely to need dialysis (21% vs 47%; p ---0.02). This correlated with a lower maximum blood urea nitrogen concentration (27.1 mmol/L [76 mg/dl] vs 42.5 mmol/L [119 mg/dl]; p = 0.02). Duration of oliguria and anuria was similar between the atypical and typical disease groups (5 vs 7 days, respectively; p = 0.91), as was duration of dialysis support (4.5 vs 4.0 days; p = 0.98). Outeome. There were no deaths in the subset of patients with atypical disease; 3.4% of patients with typical disease died. End-stage renal disease occurred in two (9%) of the patients with atypical disease (Figs. 1 and 3), compared with four (1.5%) patients in the group with typical disease. This difference was not statistically significant (p = 0.07). Two of the four patients with recurrent atypical disease had ESRD, and they accounted for all cases of ESRD in the atypical disease group. In both patients, ESRD developed after recurrences of atypical disease that followed an initial episode of typical postdiarrheal HUS. One of these patients received a cadaveric renal transplant. The transplanted kidney soon failed as a result of the recurrent HUS. In addition, a 15-year-old female adolescent with nonrecurrent atypical HUS required dialysis for 8 months; 9 months after the onset of HUS she has an estimated GFR of only 28 tal/min per 1.73 m 2. Follow-up information 1 or more years after HUS was available for 68% of patients in the atypical disease group and 47% of patients with typical HUS. A boy with typical HUS who also had reflux nephropathy and those who had ESRD before the follow-up component of out study were excluded from the analysis. The mean and median time from onset of HUS to most recent evaluation was 5 years in the atypical disease group (fange 1 to 16 years) compared with a mean of 6.3 years (median, 4 years, range 1 to 17 years) in the group with typical disease. At most recent evaluation 47% of the atypical disease group had one or more chronic sequelae (hypertension, proteinuria, low GFR) compared with 43% of the typical disease subset. Hypertension was present in 27% of the atypical disease group, compared with 11% of the typical disease group (p = 0.12). Twenty-seven percent of the atypical disease group had proteinuria, compared with 17% of the typical disease group (p = 0.5). Although a GFR <90 tal/min was noted in 13.0% of the atypical disease group and 25.9% of the typical disease group (p = 0.35), only a single individual (6.7%) in the atypical disease group and 11 (4.2%) of the typical disease group had a GFR <60 ml/min per 1.73 m z. Howerer, it is important to note that a second patient wirft atypical HUS, a 15-year-old girl with only 8 months since her HUS episode and therefore not yet eligible for the follow-up component of the study, has a GFR of only 28 ml/min. There

The Journal of Pediatrics Volume 128, Number 4

were no significant differences between the two groups with regard to the prevalence of isolated proteinuria, low GFR, hypertension, or a combination of hypertension, proteinuria, and low GFR. Follow-up information was available for a greater proportion of patients in the atypical disease group (68%) than the typical disease group (47%), primarily because of more complete follow-up of patients with mild HUS in the atypical disease group (i.e., those without oliguria during the acute phase). There were no other differences between patients who were seen in follow-up and those who were not. To control for this potentialbias, we compared risk factors among those with renal sequelae at follow-up with those without sequelae, after stratifying for potential confounders. After adjustment for presence of oliguria, there was no association between the type of HUS and the risk of proteinuria, low GFR, or one or more chronic sequelae. Atypical HUS was weakly associated with the risk of hypertension, but was not statistically significant (Maentel-Haenszel summary odds ratio, 0.36; 95% confidence interval, 0.12 to 1.05; p=0.11). DISCUSSION

Childhood HUS has traditionally been divided into two major types, postdiarrheal and atypical or nondiarrheal HUS. This difference may become blurred if the provocative findings of the Italian HUS Study Group can be corroborated.14, 15 They found serologic evidence of verotoxin-producing Escherichia coli infection (i.e., E. coli O157:H7 lipopolysaccharide antibodies) in the majority of their cases of atypical disease. Eren so, it is widely believed that atypical (nondiarrheal) HUS is a distinct entity that is associated with greater morbidity and higher mortality rates than typical (Postdiarrheal) disease. 2-n Moreover, as noted in a recent editorial by Kaplan, 16 the onset is usually insidious, the histopathologic findings are largely arteriolar (rather than glomerular), and chronic malignant hypertension and subsequent heart failure are common. The majority of pubfished reports support this view. Kibel and Bamard, 3 in a 1968 report from Rhodesia, stated that 8 (57%) of 14 children with atypical HUS died, compared with 35 deaths (37%) in the group of 95 patients with postdiarrheal HUS. The majority of the European experience has been similar. In 1982 Habib et al. 5 in Paris noted a strong correlation between atypical cases of HUS and progression to ESRD. The following year in Great Britain, Trompeter et al. 6 concluded that the likelihood of complete recovery of renal function was less likely, at least in boys, in atypical disease. In a 1989 study from England, Levin at al. 7 reported that only two of nine children (22%) with atypical HUS recovered. The remaining seven had relapses; two patients died and the five survivors had chronic renal damage. This contrasted with an 81% complete

Siegler et al.

509

recovery rate in their typical disease group. In 1992 Milford 8 reported the results of a 3-year (1985 to 1988) collaborative study in the United Kingdom. Twenty-seven percent of the patients in the atypical disease group died, and almost one half had ESRD, compared with a mortality rate of 7.9% and a chronic morbidity rate of 10% in the postdiarrheal HUS group. Fitzpatrick et al. 2 described 20 children with atypical disease who were treated in London, England, from 1968 to 1991. Although no comparative data for typical HUS were mentioned, five of the children with atypical disease died, four had ESRD, and only two of the survivors had no chronic renal damage. Eleven of the 20 patients had orte or more relapses. A similar experience was reported by Loirat et al., 9 who noted that 46% of their patients with atypical disease in Paris, France, between 1975 and 1991 had ESRD, compared with only 3% of the population with typical HUS. In 1994 Tönshoff et al. 1° reported the outcome of 59 patients with typical HUS and 21 patients with atypical disease. Of these German children with atypical HUS, 33% had chronic renal insufficiency, ESRD, or died compared with 17% of those with postdiarrheal disease. Renaud et a1.11 recently provided an npdate of their previously cited French experience 5 and focused on 42 children older than 3 years of age; 21 of these had atypical HUS. The onset of disease was insidious, ofigufia was usually absent, but nephrotic syndrome and uncontrollable hypertension developed in 50% and 100% of patients, respectively. Outcome was good for 18 of 21 (86%) patients with typical HUS, but 17 (81%) of the children with atypical HUS had ESRD. Moreover, persistent malignant hypertension in 17 children (81%) necessitated bilateral nephrectomy in six (29%); nine (43%) had one or more recurrences, and eight (38%) died. Two 1994 reports from Belgium provide different results. Kelles et al. 17 described their experience with 95 patients who had HUS between 1970 and 1982. The cohort included 12 patients with atypical disease. They found no differences in death rates, chronic sequelae, or the prevalence of ESRD between the groups. Proesmans et al. 18 reported on their experience since 1969; 70% of their 20 patients with atypical disease had mild disease, 15% died, 25% had recurrence of disease, but only 10% had ESRD. Reports from North America have been very limited. In 1978 Dolislager and Tune 4 commented on their experience with three cases of atypical disease, as well as the collective experience with five additional cases of atypical disease from two other California centers. Three (37.5%) of the eight Califomia patients died, compared with an overall mortality rate of 11% among those with postdiarrheal HUS. A more recent report by Martin et al. 19 included 16 cases of atypical HUS among 117 patients seen in Minnesota between 1979 and 1988. Although limited details were provided for their subset of patients with atypical disease, the mortality rate

5 10

Siegler et aI.

was only 6.3%, not significantly different from the 3% mortality rate noted in the group with typical HUS. Our experience with atypical HUS in the intermountain region of the western United States has likewise been different from that reportedin Affica, 3 Great Britain, 2, 6-8France,», 9 Germany, 1° and California. 4 0 u r patients with atypical disease had milder renal involvement, fewer seizures, no deaths, and otherwise had an acute-phase clinical course similar to our patients with typical HUS, As in all published long-term studies of HUS, the lack of follow-up data on all patients is a potential limitation. However, with the probable exception of the four patients who had recurrent disease, out patients with atypical disease did not appear to have a greater ilsk of long-term sequelae. A survey of the reports that include information on atypical disease recurrence 2-9, 11, 17q9 (182 patients) revealed an average recurrenee rate of 24% (range 0 to 78%), similar to out recurrence rate of 18%. Recurrent HUS m a y represent an important subset of atypical HUS. Two of our atypical patients, both with recurrent atypical disease, have ESRD. This represents an ESRD incidence of 50% for this subset; however, none have died. Kaplan 2O recently surveyed the international experience with 71 cases of recurrent HUS. He noted an incidence of ESRD of only 8%, but a mortality rate of 19%. It is interesting that our two patients with ESRD both had recurrent atypical HUS after an initial episode of classic postdiarrheal HUS. A renal biopsy was performed in orte of these female patients several months after her initial occurrence of atypical HUS; findings showed severe arterial involvement with "onionskinning." Her subsequent course was characterized by severe and persistent hypertension, congestive heart failure, and progressive azotemia. She had her second occurrence of atypical HUS (third episode of HUS) after a cadaveric renal transplant; cyclosporine, a drug implicated in causing HUS, was not used in her posttransplant management. In 1982 Habib et al.5 were the first to recognize the association between atypical HUS and arterial (rather than glomerular) renal angiopathy involving arterioles and medium-sized atteiles. Our other two patients with recurrent atypical HUS, who do not have ESRD, are siblings with familial hypocomplementemic HUS, a rare recurrent variant of HUS.21 One sibling has had four episodes, and the other has had HUS on two occasions. All episodes in these sisters were mild; none required dialysis or were associated with extrarenal (e.g., central nervous system) involvement. At follow-uP 1~ and 7 years after their most recent epis0des of HUS, respectively, they are both in good health, with no hypertension, proteinuria, or impaired renal function. In summary, atypical HUS represents a number of disease subsets that differ in etiology and natural history. In most patts of the world atypical HUS is characterized by a high

The Journal of Pediatrics April 1996

incidence of recurrence, death, and ESRD. Moreover, in contrast to typical disease, its onset is often insidious, and persistent malignant hypertension is common. In the intermountain region of the western United States, however, those with atypical disease have milder acute nephropathy than those with typical HUS. Moreover, with the exception of those with recurrent disease, they do not expeilence a worse outcome. This suggests that patients with atypical HUS compilse a very heterogeneous population and do not uniformly have poor outcomes. Studies to further characterize the pathophysiology, epidemiologic features, and prognosis are needed, and different geographic regions should define their own experience with the nondiarrheal vailety of this intriguing syndrome. ADDENDUM

Since completion of this manuscilpt, the second of two female patients with recurrent atypical HUS and ESRD received a cadaveric kidney transplant after 2 years of peritoneal dialysis. Although she had no recurrences during dialysis, 3 weeks after transplantation, while receiving cyclosporine, she expeilenced a mild episode of biopsy-proven HUS. The cyclospoilne therapy was discontinued and she recovered rapidly without specific therapy. Ten days later, when she was still not receiving cyclosporine, she was readmitted with severe encephalopathy associated with multiple brain infarcts (basal ganglia, brain stem, and thalamus). Multiple retinal infarcts also developed. Thrombocytopenia was severe, but anemia and renal involvement were mild. Becanse her features were classic for thrombotic thrombocytopenic purpura, she was treated for several weeks with plasmapheresis. Her condition gradually improved, and 3 months later she had normal hematologic values and kidney function, but still had mild neurologic sequelae and moderately severe visual impairment. REFERENCES

1. Siegler RL, Pavia AT, Christofferson RD, Milligan MK. A 20 year population-based study of postdiarrheal hemolytic uremic syndrome in Utah. Pedia~cs 1994;94:35-40. 2. Fitzpatfick MM, Walters MDS, Trompeter RS, Dillon MJ, Barratt TM. Atypical (non-diarrhea-associated) hemolyticuremic syndrome in childhood. J PEDIATR1993;122:532-7. 3. Kthel MA, Barnard PJ. The haemolytic-~aemic syndrome: a survey in southem Africa. S Afr Med J 1968;42:692-8. 4. Dolislager D, Tune B. The hemolytic-uremic syndrome. Arch Pediatr Adolesc Med [Am J Dis Child] 1978;132:55-8. 5. HabibR, Levy M, Gagnad0ux M-F, Broyer M. Prognosis of the hemolytic uremic syndrome in children. Adv Nephrol 1982; 11:99-128. 6. Trompeter RS, Schwartz R, Chantler C, et al. Haemolyticuraemic syndrome: an analysis of prognostic features. Arch Dis Chi|d 1983;58:101-5.

The Journal of Pediatrics Volume 128, Number 4

Siegler et al.

7, Levin M, Walters MDS, Barratt TM. Hemolytic aremic syndrome. Adv Pediatr Infect Dis 1989;4:51-82. 8. Milford D. The hemolytic uremic syndromes in the United Kingdom. In: Kaplan BS, Trompeter RS, Moake JD, eds. Hemolytic uremic syndrome and thrombotic thrombocytopenicpurpura, vo110. New York: Marcel Dekker, 1992: 559. 9. Loirat C, Baudouin V, Sonsino E, Mariani-Kurdjian P, Elion J. Hemolytic-uremicsyndromein the child. Adv Nephrol 1993; 22:141-68. 10. Tönshoff B, Sammet A, Sanden I, Mehls O, Waldherr R, Schärer K. Outcome and prognostic determinants in the hemolytic uremic syndrome of children. Nephron 1994;68:6370. 11. Renaud C, Niaudet P, Gagnadoux MF, Broyer M, Habib R. Haemolytic uraemic syndrome: prognostic factors in children over 3 years of age. Pediatr Nephrol 1995;9:24-9. 12. Siegler RL, Brewer ED, Pysher TJ. Hemolytic uremic syndrome associated with glomerular disease. Am J Kidney Dis 1989;13:144-7. 12a. Report of the Second Task Force on Blood Pressure Control in Children--1987. Pediatfics 1987;79:1-25. 13. Schwartz GJ, Gauthier B. A simple estimate of glomerular filtration rate in adolescent boys. J PEDtATR 1985;106: 522-6. 14. Caprioli A, Luzzi I, Rosmini F, et al. Hemolytic-uremic syn-

15.

16.

17.

18.

19.

20.

21.

511

drome and vero cytotoxin-producingEscherichia coli infection in Italy. J Infect Dis 1992;166:154-8. Gianviti A, Rosmini F, Caprioli A, et al. Haemolytic-uraemic syndromein childhood: surveillanceand case-controlstudiesin Italy. Pediatr Nephrol 1994;8:705-9. Kaplan BS. Another step forward in our understanding of the hemolytic uremic syndromes: tying up some loose ends. Pediatr Nephrol 1995;9:30-2. KellesA, Van Dyck M, Proesmans W. Childhood haemolytic uraemic syndrome:long-termoutcome and prognosticfeatures. Eur J Pediatr 1994;153:38-42. Proesmans W, Sercu E, Goos L. Hernolytic uremic syndrome not associated with diarrhea: D(-) HUS. Single center experience with 20 patients [Abstract]. Pediatr Nephrol 1994;8: C68. Martin DL, MacDonald KL, White KE, Soler JT, Osterholm MT. The epidemiology and clinical aspects of the hemolytic uremic syndrome in Minnesota. N Engl J Med 1990;323: 1161-7. Kaplan BS. Recurrent hemolytic uremic syndrome. In: Kaplan BS, Trompeter RS, Moake JL, eds. Hemolytic uremic syndrome and thromboticthr0mbocytopenicpurpura, vo110. New York: Marcel Dekker, 1992:151-62. Carreras L, Romero R, Requesens C, et al. Familial hypocomplementemic hemolytic uremic syndrome with HLA-A3,B7 haplotype. JAMA 1981;245:602-4.

FELLOWSHIPS Fellowships available in pediatric subspecialfies and those for general academic pediatric training are listed once a year, in January, in THE JOURNALOF PEDIATRICS.Each June, forms for listing fellowships available for the academic year beginning 18 months after publication are sent to the Chairman of the Department of Pediatrics at major hospitals in the United States and Canada. In addition, a copy of the application form appears in the July, August, and September issues of THE JOUe,NnL (please use the current form). Should you desire to list fellowships, a separate application taust be made each year for each position. All applications taust be returned to Mosby-Year Book, Inc., by October 15 preceding the listing year to ensure publication. Additional forms will be supplied on request from the Journal Editing Department, Mosby-Year Book, Inc., 11830 Westline Industrial Drive, St. Louis, MO 63146-3318/800-325-4177, ext. 4317, or 314-4534317.