Children with nephrotic syndrome (NS) are exposed to multiple infectious complications resulting in significant mortality and morbidity [1,2]. There is paucity of literature regarding the clinical spectrum of major infections in childhood NS from developing countries in recent years. Knowledge of these parameters has therapeutic and preventive relevance [2]. We, therefore, conducted the present study.

This prospective study was conducted from November 2010 to July 2012 at a tertiary hospital, and was approved by the Institutional ethics committee. The primary objective was to study the incidence of major infections in NS, while secondary objectives were to evaluate (a) the risk factors for major infections and (b) their etiological spectrum. Children below 13 years of age, fulfilling ISKDC diagnostic criteria for NS [3-5] were included if hospitalization was required for: major infections, severe anasarca, shock, seizures, thrombosis, unconsciousness, respiratory distress, tetany, or for renal biopsy.

The incidence of major infections in NS was estimated as 35% [2,6,7]. Assuming 10% variation with 95% confidence, sample size for estimating incidence of major infections was 88. Results were analyzed with SPSS version16, using Student t-test, Chi square test or Fisher exact test. Logistic regression was used to analyse predictors of major infections and peritonitis.

Major infections were defined as disseminated, affecting deep organs, requiring hospitalization (e.g. cellulitis, disseminated varicella) or potentially life-threatening [8]. Specific major infections were defined as follows:

Severe ascites was defined as tense ascites or ascites with dyspnea. Generalized edema (including scrotal edema, vulval edema or severe ascites) was considered as severe anasarca. Frequently relapsing (FRNS), steroid dependent (SDNS), steroid resistant (SRNS) or infrequently relapsing NS (IFRNS) were considered as ‘more severe clinical types of NS’.

86 children with NS, requiring 101 hospitalizations, were recruited. Peritonitis and pneumonia accounted for 27 out of the 37 children with major infections (72.9%) (Table I). Urine cultures were sent in 10 cases (3 had UTI); while ascitic tap was performed in 20 (14 had peritonitis). The infectious agents included E.coli (2 cases) and Klebsiella (1 case) in UTI, Pneumococcal peritonitis (1 case), Methicillin Resistant Staphylococcus aureus (MRSA) in cellulitis and osteomyelitis (1 each), and Pseudomonas in cellulitis (1 case). One child with SRNS (having chronic glomerulonephritis) died of pneumonia. Children aged >4 years constituted 78.6% of peritonitis, while duration of NS >2 years constituted 64.3% of peritonitis. The mean age of children with peritonitis was 7.9 ± 3.3 years, whereas mean age of children with pneumonia was 6.4 ± 3.1 years.

TABLE I Clinical and Biochemical Profile of Study Subjects (among 101 hospital admissions * )

Web Table I depicts predictors of peritonitis on univariate analysis. On logistic regression, hypercholesterolemia >400 mg/dL independently predicted peritonitis (OR = 6.89 (1.30 – 36.43), P = 0.023).

Predictors for major infections on logistic regression were ‘more severe clinical types of NS’ when compared to first episode of NS, i.e. IFRNS [OR 9.94 (2.56 -38. 67), P< 0.01], FRNS [OR 28.19 (3.73- 212.28) , P< 0.01], SDNS [OR 25.84 (2.89-231.1), P< 0.01], SRNS [OR 4.9 (2.56 -38. 67), P=0.033]; and severe ascites [OR 0.15(0.05-0.49), P< 0.01].

In our study, major infections constituted 36.6% of the subjects. Other studies reported incidence ranging from 32% to 38% [2,7]. The commonest major infection was peritonitis (13.8% incidence), as compared to 1.4% to 16% in other studies [1,2,7,11,12]. Ascitic fluid culture was positive in only 1 patient. Eleven of the 14 cases with peritonitis were initially seen by general practitioners who prescribed antibiotics. These were ineffective but probably contributed to low ascitic fluid culture positivity [2]. Pneumonia was the second commonest infection in the subjects (12.9%), as compared to 3.9% to14% in Indian studies [1,2]. UTI was relatively uncommon in our subjects. In consonance with our observations, Srivastava, et al. [7] reported no episodes of UTI in their study. In other studies, UTI incidence ranged from 13.7% to 46% [1]. These variations probably reflect geographical or socioeconomic heterogeneity of patient populations.

Hypercholesterolemia >400 mg/dL was a risk factor for peritonitis. Hypercholesterolemia may have a direct pathophysiological role as hypercholesterolemic serum inhibits lymphocyte proliferation in response to antigen stimulation [15] Secondly, hypercholesterolemia is an indirect consequence of hypoalbuminemia which has been mentioned as a predictor of peritonitis in NS [2,13].

The merits of our study include prospective design, and a large sample size. Our study also has limitations. Due to resource constraints, detailed pathophysiological investigations were not undertaken. Secondly, although the predictors for major infections and peritonitis were analyzed, this study is not powered enough for the same.

Our hospital caters to patients of low socioeconomic status, which explains poor coverage of pneumococcal vaccination (6%). Improving pneumococcal vaccination coverage in NS could be a potentially important strategy to decrease the incidence of major infections. In the presence of hypercholesterolemia >400 mg/dL, an aggressive search for serious infections is essential along with prompt and aggressive antibiotic therapy.

Contributors: PA, SK and NB: were involved in management of the patients; PA: collected the data, reviewed the literature and drafted the manuscript; SK: conceptualized the study, reviewed the literature and critically reviewed the manuscript; NB: also critically reviewed the manuscript; JM: supervised the laboratory tests. All authors contributed to writing the paper and approved the final version of the manuscript and SK: shall act as guarantor of the paper.

Funding: None; Competing interests: None stated.