Post-streptococcal acute glomerulonephritis in children: Association between proteinuria levels and renal outcomes

Abstract

BACKGROUND

Post-streptococcal acute glomerular nephritis (PSAGN) is mostly a benign condition. The usual sequelae of PSAGN include hypertension, its complications, and acute kidney injury. Severe PSAGN is associated with significant long-term morbidity, and histological abnormalities such as crescentic glomerulonephritis are infrequently reported. PSAGN has also been linked to late-onset chronic kidney disease in some populations due to high levels of proteinuria.

AIM

To evaluate the association between proteinuria levels and renal outcomes in children with PSAGN.

METHODS

This prospective observational study was conducted at Lady Ridgeway Hospital (Colombo, Sri Lanka) over 15 months. Children with PSAGN were enrolled based on clinical and laboratory criteria. Persistent proteinuria ≥ 2+ for 2 weeks and serum creatinine > 100 μmol/L warranted renal biopsy, assessed via light microscopy and immunofluorescence. Normalization of complement 3 (C3) within 6 to 8 weeks was required for inclusion. Data on clinical features, urine protein levels, and renal function were collected from patient records, and potential associations were analysed using Statistical Package for the Social Sciences and R language for statistical computing. Ethical approval was obtained from the Ethical Review Committee, Lady Ridgeway Hospital for Children (Ref No: LRH/ERC/2021/60).

RESULTS

Forty-four patients were recruited. There were 27 (61.4%) male patients and 17 (38.6%) female patients. Thirty-seven (84%) of them were above 5 years of age. Twenty (45%) patients had a history of skin sepsis, and eighteen (41%) had a history of throat infection. Among patients with proteinuria ≥ 2+, 53% had serum creatinine > 100 µmol/L, while among those with proteinuria < 2+, 7% had serum creatinine > 100 µmol/L. The association of high-degree proteinuria with elevated serum creatinine was significant (χ² = 7.8, P = 0.005) in PSAGN. The odds ratio of the logistic regression model was 1.049 (95% confidence interval: 1.003-1.098), indicating a positive direction with statistically significant association (P = 0.037). There was no significant association between proteinuria and the degree of hypertension or estimated creatinine clearance. Ten children underwent renal biopsy. Crescents (less than 50%) were demonstrated in five children, while three children had typical diffuse proliferative glomerulonephritis. One child had severe acute tubular necrosis, and another had crescentic glomerulonephritis (crescents > 50%). The immunofluorescence studies revealed deposition of immunoglobulin G and C3 in all biopsy specimens.

CONCLUSION

High-degree proteinuria was significantly associated with elevated serum creatinine (> 100 μmol/L) in children with PSAGN. The majority of children with persistent proteinuria ≥ 2+ for more than 2 weeks and the highest recorded serum creatinine > 100 μmol/L had atypical renal histological findings.

Key Words: Proteinuria; Post-streptococcal acute glomerulonephritis; Renal outcome; Renal biopsy; Crescents; Serum creatinine

Core Tip: This study highlights the significant association between high-grade proteinuria and elevated serum creatinine in children with post-streptococcal acute glomerular nephritis. Children with high-grade proteinuria and elevated serum creatinine require follow-up to monitor long-term kidney function, given the association with the histological presence of crescents and a potential increased risk for chronic kidney disease. Although no association was found between high blood pressure and the degree of proteinuria during the acute phase, further evaluation of this cohort is needed to monitor for late-onset renal hypertension.

INTRODUCTION

Post-streptococcal acute glomerulonephritis (PSAGN) in children is characterised by the appearance of haematuria (macroscopic or microscopic), oedema, oliguria, hypertension, and mild proteinuria[1] following either group A β haemolytic Streptococcal pharyngitis or skin infection. The disease is immune-mediated, and the underlying mechanism includes type 3 hypersensitivity reaction that leads to deposition of immune complexes in glomerular blood vessels. Immune-mediated damage leads to features of AGN. The severity of PSAGN is variable and can be anywhere in the spectrum from asymptomatic illness[2,3] to severe acute kidney injury (AKI) secondary to rapidly progressive glomerulonephritis (RPGN)[4,5]. Most children with PSAGN have mild proteinuria, however; massive proteinuria can be seen in about 5% of affected children[6].

Severe PSAGN is associated with significant long-term morbidity, and histological abnormalities such as crescentic glomerulonephritis are infrequently reported[4,5]. Although, crescentic glomerulonephritis is more commonly seen following vasculitic illnesses such as systemic lupus erythematosis (SLE) and antineutrophil cytoplasmic antibody-associated vasculitis, in one series, crescentic glomerulonephritis was frequently observed following initial diagnosis of acute PSAGN[7]. Furthermore, PSAGN is related to late-onset chronic kidney disease in some populations due to high levels of albuminuria[8]. Persistent macroscopic haematuria and high-grade proteinuria in PSAGN usually have a less favourable prognosis[9].

The aim of this prospective study was to assess the association between high-grade proteinuria and disease outcomes (e.g., AKI, renal histological findings, hypertension) in children with PSAGN. Therefore, the investigators hypothesised that children with PSAGN presenting with persistent proteinuria ≥ 2+ for more than 2 weeks and serum creatinine value > 100 μmol/L are likely to have atypical renal histological findings.

MATERIALS AND METHODS

Study design

This observational, cross-sectional, prospective study was conducted at the University Paediatric Unit of Lady Ridgeway Hospital (Colombo, Sri Lanka) over a duration of 15 months (from January 2022 to March 2023). All children for whom the diagnosis of PSAGN was confirmed based on clinical criteria (history of sore throat/skin sepsis, haematuria, oliguria, hypertension) and laboratory criteria (high antistreptolysin O titer, red blood cells in urine, low complement 3 [C3]) and who were followed up at the unit were recruited prospectively. Biochemical studies of urine and blood were performed in the biochemistry laboratory of Lady Ridgeway Hospital. Only first morning urine samples were considered for the evaluation of proteinuria using urine protein strips.

Data were collected using a structured data collection sheet, and sources of data included patients’ medical records and investigation reports. Data on clinical features at presentation, serial urine protein values, blood pressure records, and renal function test results were collected.

Study population

All children who had persistent proteinuria ≥ 2+ for more than 2 weeks and the highest recorded serum creatinine > 100 μmol/L were subjected to renal biopsy. The highest recorded serum creatinine and urine protein values were taken for the analyses. Light microscopy and immunofluorescence were used to assess the renal biopsies.

Only children in whom the initial low complement level (C3) was normalised 6 to 8 weeks following the onset of nephritis were included in the analysis; others were excluded from the study. Subsequent normalization of the complement level was used as inclusion criterion to exclude children who likely had conditions causing persistent low complement levels (e.g., SLE, C3 glomerulopathy).

Statistical analysis

All data were collected by the same clinical research associate to minimise record retrieval bias. Data were analysed using Statistical Package for the Social Sciences version 19.0 (IBM Corp., Armonk, NY, United States). Association of serum creatinine level, hypertension, and estimated creatinine clearance (eCCl) on dichotomised proteinuria was assessed using the χ² test. Binary logistic regression analysis was conducted to evaluate the level of proteinuria based on independent factors such as serum creatinine level, blood pressure, and estimated creatinine clearance. P < 0.05 was considered statistically significant.

Ethical standards

Ethical approval for the study was granted by the Ethics Review Committee of the Lady Ridgeway Hospital for Children (Ref No: LRH/ERC/2021/60). Informed written consent was obtained from the parents of all participating children in the study.

RESULTS

Data of 44 children who met the inclusion criteria were used for the analysis. Twenty-seven were male (61.4%), while seventeen were female (38.6%). Thirty-seven children were over 5 years of age (84%). Table 1 shows the age and sex distribution of the study population.

Table 1 Age and sex distribution of the study population.

Age in years	Male	Female
Birth-5	5 (11.4)	2 (4.5)
5-10	9 (20.4)	9 (20.4)
> 10	13 (29.5)	6 (13.6)

Data are n (%).

Twenty children (45%) had clinical and microbiological evidence of preceding skin infection, whereas eighteen children had evidence of preceding pharyngitis (41%). The mean C3 complement level at the time of initial PSAGN diagnosis was 57 mg/dL (range: 41-69 mg/dL, normal range: 80-160 mg/dL).

Thirty-four children had sequential blood pressure measurements recorded together with degree of proteinuria. There was no significant association between proteinuria (≥ 2+) and the degree of hypertension (> 95^th percentile for age, sex, and height) or its complications (χ² = 0.854, P > 0.05) (Table 2). The odds ratio (OR) of the logistic regression model was 1.902 (95% confidence interval [CI]: 0.116-11.258) and P = 0.657, indicating that the association was not statistically significant.

Table 2 Association between the degree of proteinuria and hypertension in children with post-streptococcal acute glomerular nephritis.

Degree of proteinuria	Blood pressure centile			Total
	> 95th	90th-95th	< 90th
< 2+	7 (20.6)	1 (2.9)	3 (8.8)	11 (32.3)
≥ 2+	18 (52.9)	1 (2.9)	4 (11.8)	23 (67.6)
Total	25 (73.5)	2 (5.9)	7 (20.6)	34 (100)

Data are n (%).

Thirty-nine children had sequential serum creatinine measurements recorded together with degree of proteinuria. Fifty-four percent of children with proteinuria ≥ 2+ had a maximal serum creatinine ≥ 100 µmol/L. Only 7.6% of children with proteinuria less than 2+ had elevated serum creatinine > 100 µmol/L (Table 3). The association of high-degree proteinuria with elevated serum creatinine was significant (χ² = 7.8, P = 0.005) in PSAGN. The OR of the logistic regression model was 1.049 (95%CI: 1.003-1.098) indicating a positive direction with statistically significant association (P = 0.037).

Table 3 Association between the degree of proteinuria and serum creatinine in children with post-streptococcal acute glomerular nephritis.

Degree of proteinuria	Serum creatinine in µmol/L		Total
	< 100	≥ 100
< 2+	12 (30.8)	1 (2.6)	13 (33.3)
≥ 2+	12 (30.8)	14 (35.9)	26 (66.7)
Total	24 (61.5)	15 (38.5)	39 (100)

Data are n (%).

The AKI following PSAGN was assessed in 39 children using paediatric Risk, Injury, Failure, Loss of Kidney Function, and End-stage Kidney Disease (pRIFLE) criteria. Twenty-seven percent of children with proteinuria ≥ 2+ had eCCl less than 50% resembling injury to the kidney, while only 7.7% of children with proteinuria less than 2+ had eCCl less than 50% (Table 4). The association of high-degree proteinuria with injury to the kidney (eCCl < 50%-pRIFLE) was not significant by the χ² test (P = 0.16). The OR of the logistic regression model was 0.993 (95%CI: 0.978-1.090) and P = 0.374, indicating that the association was not statistically significant.

Table 4 Association between the degree of proteinuria and injury to kidney according to pRIFLE criteria in children with post-streptococcal acute glomerular nephritis.

Degree of proteinuria		Estimated creatinine clearance		Total
		< 50%	> 50%
< 2+		1 (2.5)	12 (30.8)	13 (33.3)
≥ 2+		7 (18)	19 (48.7)	26 (66.7)
Total		8 (20.5)	31 (79.5)	39 (100)

Data are n (%).

Ten children who had persistent proteinuria ≥ 2+ for 2 weeks and highest recorded serum creatinine > 100 μmol/L were underwent renal biopsy. Crescents (less than 50%) were demonstrated in five children, whereas three children had typical diffuse proliferative glomerulonephritis. One child had severe acute tubular necrosis. Another child had crescentic glomerulonephritis (crescents > 50%). The immunofluorescence studies revealed deposition of IgG and C3 in all biopsy specimens. The renal biopsy histological findings are presented and summarised in Table 5.

Table 5 Renal biopsy findings in children with persistent proteinuria ≥ 2+ for 2 weeks and highest recorded serum creatinine > 100 μmol/L.

Age	Macroscopic findings	Microscopic findings	Conclusion
12 years	Two pieces of tissue measuring 5 mm and 6 mm	Glomeruli-Twelve glomeruli were seen. All glomeruli showed mesangial proliferation and increased matrix. Mesangial hyperplasia was seen with increased intimal proliferation and obliteration of lumina. Three glomeruli showed cellular crescents. Basement membrane was not thickened. Tubules-No tubular distortion was noted. Interstitium-No inflammation was seen. Vessels-Vessels were not seen	Acute diffuse proliferative glomerulonephritis with cellular crescent formation
8 years 3 months	Two cores measuring 4 mm × 1 mm × 1 mm, 3 mm × 1 mm × 1 mm	Glomeruli-Ten glomeruli were seen. Increased mesangial cell hyperplasia and matrix were noted. No wire loops/necrosis was seen. Basement membrane was not thickened. Neutrophil infiltration was present. Crescents were seen in two glomeruli. No chronic changes were noted. Tubules-Tubules showed mild dilatation and flattening. There was no evidence of tubulitis. Amorphous material was seen in the lumen of tubules. Some degree of tubular damage was evident. Calcification was not seen. Interstitium-Mild inflammation was present. Vessels-Vessels were normal	Acute diffuse proliferative glomerulonephritis with cellular crescent formation. Mild degree of tubular damage was seen
6 years 3 months	Four cores measuring 15 mm, 10 mm, 4 mm, and 3 mm in length and 1 mm in diameter	Glomeruli-Sixteen viable glomeruli were seen. They showed diffuse mesangial proliferation. There was inflammation with neutrophil infiltration. Three glomeruli showed endocapillary proliferation with early crescent formation. Cellular crescents were seen in three glomeruli. No necrosis or sclerosis was present. Basement membrane was not thickened. Tubules-The tubules were unremarkable. No tubulitis or evidence of acute tubular necrosis was present. Interstitium: The interstitium was unremarkable. Vessels: Vessels were not seen	Acute diffuse proliferative glomerulonephritis with crescent formation
8 years	Three cores measuring 15 mm × 1 mm × 1 mm and 5 mm × 1 mm × 1 mm	Glomeruli-Twelve viable glomeruli were seen. All showed varying degree of intimal proliferation and obliteration of lumina, mesangial hyperplasia with increased of intimal proliferation, and obliteration of lumina. Early crescents formation seen in four glomeruli. Tubules: Morphology was unremarkable. Interstitium: Focal infiltrate of lymphocytes, neutrophils and oedema were seen	Proliferative glomerulonephritis with crescents formation
5 years 6 months	Two cores measuring 18 mm and 14 mm in length and 2 mm in diameter	Glomeruli-Eighteen viable glomeruli were seen in two cores. Varying degree of mesangial hypercellularity and increased matrix were evident. Three glomeruli showed endothelial hyperplasia with obliteration of lumen and infiltration by neutrophils. No sclerosis or crescents was seen. Tubules-No Tubular distortion was seen. Interstitium-No inflammation was present. Vessels-Unremarkable	Proliferative glomerulonephritis suggestive of postinfectious glomerulonephritis
12 years 4 months	Two cores measuring 8 mm and 4 mm	Glomeruli-Sixteen viable glomeruli were seen. They showed diffuse mesangial proliferation. There was inflammation with neutrophil infiltration. Three glomeruli showed endocapillary proliferation. No crescents or sclerosis was seen. Tubules-The tubule showed dilatation, vacuolation and flattening. There was evidence of tubulitis. Amorphous material was seen in the lumen of tubules. Severe degree of tubular damage was present. Interstitium-Focal infiltrate of lymphocytes was seen. Blood Vessels-Unremarkable	Acute diffuse glomerulonephritis with severe tubular damage
11 years 10 months	Two cores measuring 8 mm × 1 mm × 1 mm and 4 mm × 1 mm × 1 mm	Glomeruli-Fourteen viable glomeruli were seen. They showed mild diffuse mesangial proliferation with neutrophil infiltrates. No evidence of crescent formation, hypercellularity, or karyorrexhis was seen. Tubules-Tubules were histologically unremarkable. Interstitium-No evidence of inflammation or fibrosis was seen. Blood vessels-unremarkable	Acute diffuse proliferative glomerulonephritis
7 years 8 months	Two pieces of tissue measuring 6 mm and 7 mm	Glomeruli-Total of 10 Glomeruli were seen. All glomeruli showed mesangial proliferation, and increased matrix infiltrated with neutrophils. Mesangial hyperplasia with increased intimal proliferation and obliteration of lumina were seen. Three glomeruli showed endocapillary proliferation with early crescents formation. Three glomeruli showed cellular crescents. Basement membrane was not thickened. Tubules- Mild tubular distortion was seen. Interstitium-Focal infiltrate of lymphocytes was seen. Vessels-Vessels were not seen	Acute diffuse proliferative glomerulonephritis with > 50% cellular crescent formation (rapidly progressive glomerulonephritis)
5 years 3 months	Two cores measuring 14 mm and 10 mm in length and 1 mm in diameter	Glomeruli-Fourteen viable glomeruli were seen. They showed diffuse mesangial proliferation. There was neutrophil infiltration. Three glomeruli showed endocapillary proliferation with early crescent formation. No necrosis or sclerosis was seen. Basement membrane was not thickened. Tubules-The tubules were unremarkable. Interstitium-The interstitium was unremarkable. Vessels-Unremarkable	Acute diffuse proliferative glomerulonephritis with early crescent formation
7 years 6 months	Two cores measuring 12 mm and 10 mm in length and 2 mm in diameter	Glomeruli-Twenty viable glomeruli were seen in two cores. Varying degree of mesangial hypercellularity and increased matrix were seen. No sclerosis or crescents was seen. Mild neutrophil infiltrates were present. Basement membrane was normal. Tubules-No Tubular distortion was seen. Interstitium-No inflammation was present. Vessels-unremarkable	Proliferative glomerular nephritis. The findings were in favour of postinfectious glomerulonephritis

DISCUSSION

PSAGN is the most common type of glomerulonephritis reported in 3-year-old to 12-year-old children, and hypertensive encephalopathy and RPGN are known complications[10]. Although only mild proteinuria is seen in most children with PSAGN, proteinuria can be variable ranging from no proteinuria[11,12] to gross proteinuria[6].

The current study observed no relationship between degree of proteinuria and severity of hypertension during the acute stage. In PSAGN, the likely mechanisms of hypertension include nephritis-induced reduction in glomerular filtration rate, plasma volume expansion, and activation of renin-angiotensinogen-mediated aldosterone secretion. However, studies have reported that patients with high-grade and persistent proteinuria in PSAGN can develop persistent or later onset renal hypertension[13]. Therefore, it is important to follow up the current cohort of children to monitor blood pressure and evaluate long-term impact of high-grade proteinuria on blood pressure compared to mild proteinuria. Although a significant association of heavy persistent proteinuria with serum creatinine was observed in the current study, the association of injury to kidney resembling > 50% reduction of eCCl (pRIFLE criteria) with proteinuria was not significant. The reduction of eCCl > 50% indicates that the moderate to severe degree of AKI and PSAGN is mostly associated with a mild degree of AKI. This may explain the insignificant association of the latter.

All children in the current study had their complement levels (C3) within normal range 6 weeks following clinical recovery of PSAGN. Hypocomplementenemia, which presents during the acute illness but corrects 6 to 8 weeks after clinical recovery, is highly supportive of PSAGN and enables differentiation from vasculitic nephritis such as SLE and C3 glomerulopathy[14]. Furthermore in our cohort, only one child had clinical and histological (crescents present in more than 50% glomerulus) findings compatible with RPGN, while five children had crescents (less than 50%) in renal histology.

Crescents, particularly when large or present in a higher percentage, indicate poor long-term prognosis and high risk for end-stage renal disease[15]. Further, the presence of fibrocellular crescents and increased frequency of gaps in Bowman’s space are other histological risk factors for a poor outcome[4]. It is postulated that these gaps promote albuminuria and facilitate progression to chronic kidney disease. The presence of crescents in renal histology among children with high-grade proteinuria in the current study indicates that these children may still be at risk for chronic kidney disease later in life despite having no criteria for RPGN.

Histological progression and long-term clinical outcomes of this entity are yet to be ascertained. Hyperfiltration injury during acute nephritis[16] and late-onset albuminuria[17] are suggested other mechanisms for chronic kidney disease following PSAGN. Thus, the authors would like to consider this specific group, with crescents less than 50%, as having atypical PSAGN and highlight the importance of long-term follow-up. It will be interesting to determine whether this group of children would benefit from some degree of immunosuppression in future studies.

Previous studies have reported an association of high-grade proteinuria with low birth weight and high body mass index among children with PSAGN[17]. Furthermore, in children with PSAGN, high levels of antinuclear antibodies are associated with high grades of renal dysfunction[18]. The current study effectively ruled out differential diagnoses for PSAGN by demonstrating normal complement titres during follow-up. One limitation of the current study was that the authors had to rely on early morning spot checks of urinary albumin rather than 24-hour estimates due to practical difficulties. However, since all children in both groups were subjected to the same investigations by the same investigator, the overall effect of this limitation could be minimised. The authors also propose that the association of proteinuria with kidney injury can be more comprehensively assessed with a larger sample size.

As per the observations and findings of the current study, the renal histology revealed important prognostic findings in patients with more severe and atypical presentations of PSAGN. Therefore, the authors recommend that a kidney biopsy be considered at a lower threshold for children with PSAGN presenting with persistent high-degree proteinuria and high serum creatinine. The authors also recommend the long-term follow-up of children with persistent proteinuria ≥ 2+ for more than 2 weeks and the highest recorded serum creatinine value > 100 μmol/L for early detection of chronic kidney disease.

CONCLUSION

This study brings to light the significant association between high-grade proteinuria and high serum creatinine value in children with PSAGN. Children with high-grade proteinuria for more than 2 weeks and the highest recorded serum creatinine value > 100 μmol/L need follow-up to monitor long-term kidney function, given its association with the histological presence of crescents and the potential increased risk for chronic kidney disease. Although there was no association of high blood pressure with the degree of proteinuria during the acute phase, further evaluation of this cohort is required to monitor for late-onset renal hypertension.

ACKNOWLEDGEMENTS

All authors sincerely thank Dr. Supun Manathunga of the University of Peradeniya for his valuable input in completing the revised manuscript.