Abstract
Hemolytic uremic syndrome (HUS), a cause of pediatric acute kidney injury (AKI), has a spectrum of extra-renal manifestations. While neurological and gastrointestinal system involvement is common, cardiac involvement is rare. This is more so with pericardial involvement, though it has been reported in a handful of HUS cases associated with shiga toxin-producing Escherichia coli (STEC HUS). However, this complication has scarcely been reported in atypical HUS (aHUS) where there is alternate complement abnormality or DKGE (diacylglycerol kinase epsilon) mutation. We describe two children diagnosed with anti-complement factor H (CFH) antibody-associated aHUS who had pericardial involvement. Two boys, one 10-year-old and another 8-year-old, presented with pallor, oliguria and hypertension. They both had microangiopathic haemolytic anemia, thrombocytopenia and AKI suggestive of HUS. Complement workup revealed elevated anti-CFH antibody titres. With a diagnosis of anti-CFH antibody aHUS, they were started on plasmapheresis, pulse methylprednisolone and cyclophosphamide. The first case developed cardiac tamponade during the second week of hospital stay for which he needed pigtail drainage and further immunosuppression with rituximab. He gradually improved and pigtail was removed. The second case presented with pericardial effusion which subsequently resolved during the course of treatment. Thus, our patients developed pericardial effusion, with one of them progressing to life-threatening cardiac tamponade. Therefore, it is prudent that we are aware of this complication while treating children with aHUS.
Keywords: Case report, Acute kidney injury, Hemolytic uremic syndrome, Pericardial effusion, Plasmapheresis
Introduction
Hemolytic uremic syndrome (HUS) is a triad of microangiopathic hemolytic anemia, thrombocytopenia and acute kidney injury (AKI) [1]. It is a leading cause of AKI in children. Although HUS is predominantly a disease of the kidneys, extra-renal involvement is commonly seen in central nervous system, liver and pancreas. Cardiac involvement, especially of pericardium is rarely documented in the literature. There are reported cases of shiga toxin-producing Escherichia coli (STEC HUS) with pericardial involvement (Table 1). However, this clinical feature is rarely documented in atypical HUS (aHUS). Here, we report 2 children with aHUS due to anti-complement factor H (CFH) antibodies who developed pericardial effusion.
Table 1.
Profile of reported cases of hemolytic uremic syndrome with pericardial involvement and our two cases
| Author | Birk et al. [7] | Mohammed [6] | Case 2 | Case 1 | |
|---|---|---|---|---|---|
| Age (years) | 6 | 2 | 8 | 10 | |
| Sex | Female | Male | Male | Male | |
| Presentation | Diarrhea | Diarrhea | No diarrhoea | Diarrhea | |
| Cardiac tamponade | Cardiac tamponade | Pericardial effusion | Cardiac tamponade | ||
| Se Cr (µmol/L) | 324 | 381 | 600 | 880 | |
| Urine protein | – | 1 g/L | – | 4 + | |
| Hematuria | – | Present | – | Microscopic | |
| Stool examination |
During epidemic of E. coli; Not tested |
Culture E. coli O157: H7 | Stool culture sterile | ||
| Anti CFH Ab | – | – | – | Elevated | Elevated |
| Troponin | - | High | - | - | |
| Dialysis | Yes | Yes | Yes | Yes | |
| Treatment | Supportive | Supportive |
PLEX Immunosuppressant |
PLEX Immunosuppressant |
|
| Outcome | Died | Alive | Alive | Alive |
Case 1
10-year-old previously normal boy presented with progressive pallor for 1 week and oliguria for 2 days. He had history of loose stools 15 days back which lasted for 1 week. However, there was no history of skin or throat infection, cola-colored or frothy urine. On examination, he had normal weight and height for age, but had stage II hypertension, tachypnea, pallor and anasarca.
Investigations revealed haemolytic anemia with hemoglobin of 6.9 g/dL, peripheral smear showing > 2% schistocytes, elevated lactate dehydrogenase (LDH) 1190 U/L and platelet count of 85 × 109/L. His serum creatinine was 880 µmol/L. Urine analysis was suggestive of nephrotic range proteinuria (54 mg/m2/h) with no RBCs. HUS was considered. Further work-up revealed ANA (anti-nuclear antibody) was negative, HIV serology was non-reactive and homocysteine level was normal (11.9 µmol/L). C3 (76 mg/dL) and C4 (35 mg/dL) were normal; however, anti-CFH antibody levels done by VIDITEST ELISA (enzyme-linked immunosorbent assay) kit were elevated to the tune of 382AU/ml (normal < 27AU/ml). Therefore, a diagnosis of anti-CFH antibody aHUS was proferred.
He received 2 sessions of hemodialysis and 10 cycles of plasmapheresis (PLEX). Further, 5 pulses of methylprednisolone (30 mg/kg/dose) and one pulse cyclophosphamide (500 mg/m2), followed by oral steroids (1 mg/kg/day) were given with which he had remission, in terms of increasing urine output, no hemolysis and normal renal function (serum creatinine 70.9 µmol/L) by day 12. His hypertension was well controlled with six anti-hypertensive drugs.
On day 14, he had chest pain and muffled heart sounds. Chest X-ray was suggestive of cardiomegaly and echocardiogram (ECHO) revealed pericardial effusion. In the next 2 days, he developed poor perfusion with elevated JVP. Possibility of cardiac tamponade was kept and pigtail insertion was done which drained 350 ml of exudative pericardial fluid on the first day. Initially, infections, hypothyroidism, vasculitis were considered as differentials of the cause for pericardial effusion. However, on analysis, pericardial fluid showed 10 cells (predominantly polymorphs), normal sugar (97 mg/dL) and elevated protein levels (500 mg/dL). Gram stain and culture were not suggestive of any organism. Workup for tuberculosis included gastric lavage and pericardial fluid for acid-fast bacilli (AFB) smear and culture, which were negative. Epstein–Barr virus (EBV) and Cytomegalovirus (CMV) serology was non-reactive. Thyroid profile was normal. ANA and ANCA (anti-neutrophil cytoplasmic antibody) were negative which ruled out vasculitis-associated and autoimmunity-related serositis. At this time, he also had features of hemolysis. Therefore, possibility of disease activity with extra-renal involvement of pericardium was considered and rituximab (375 mg/m2/day) was given following which he improved and pigtail was removed.
At discharge, he had normal urine output, no hemolysis and normal renal function. He had a 3-month follow-up during which he had normal renal function (serum creatinine 44.2 µmol/L) but hypertension was persistent needing anti-hypertensive drugs.
Case 2
An 8-year-old, previously normal boy presented with acute-onset oliguria and anasarca for 8 days. He had no history of throat or skin infection, cola-colored or frothy urine, dyspnea or previous episodes or any preceding diarrhea. On examination, his weight and height were normal for his age. He had tachycardia (132/ min), tachypnea (42/min), stage II hypertension (125/96 mmHg), pallor and anasarca. He also had muffled heart sounds and pericardial rub.
Investigations revealed hemolytic anemia with hemoglobin of 7 g/dL, thrombocytopenia (92,000 cells/ mm3), peripheral smear showing > 2% schistocytes and an elevated LDH (907 U/L). Renal parameters were suggestive of stage 3 AKI. Urine analysis revealed 5–10 RBCs and nephrotic range proteinuria (60 mg/m2/hr). He had low C3 (60 mg/l) and normal C4 levels (24 mg/l). ANA and HIV serology was negative. Renal biopsy was suggestive of thrombotic microangiopathy (TMA) (Fig. 1). Anti-CFH antibody titres (done by VIDITEST anti-CFH antibody ELISA kit) were elevated at 260 AU/ml (normal < 27AU/ml). Chest X-ray revealed cardiomegaly and ECHO confirmed pericardial effusion of 12–14 mm (Figs. 2, 3, 4).
Fig. 1.
Photomicrographs of kidney biopsy showing swollen endothelial cells (stars, a–c), re-duplicated membranes (thin arrows, a–c) and small arterioles showing swollen endothelial cell (broad arrow, a) with subendothelial fibrinoid material (a-PAS, b-silver methenamine, c-uranyl acetate and lead citrate; a-X20, b-X40, c-X75000, original magnification)
Fig. 2.
Echocardiography of second case: parasternal long axis and apical 4-chamber views showing moderate pericardial effusion (red arrow head)
Fig. 3.

Chest radiography: cardiomegaly
Fig. 4.
Electrocardiogram showing low-voltage complexes
With a diagnosis of anti-CFH antibody HUS, he was initiated on PLEX (11cycles). He also received immunosuppressant therapy (5 pulses of methylprednisolone initially followed by oral steroids and cyclophosphamidepulse). He developed hypertensive encephalopathy (MRI brain suggestive of posterior reversible encephalopathy syndrome) for which he was managed with parenteral anti-hypertensives. During the course, repeat ECHO showed resolution of pericardial effusion.
He improved subsequently with normal urine output, normal renal function and was discharged. He received another 5 doses of cyclophosphamide with tapering doses of steroids, thereafter was given mycophenolate mofetil (1000 mg/m2/day) for 18 months. On follow-up of 2 years, he remains to be asymptomatic except for hypertension, with normal urinalysis (no proteinuria or hematuria) and renal function (serum creatinine 53.4 µmol/L).
Discussion
The above two cases highlight that patients with anti-CFH antibody aHUS may have pericardial effusion, which can even lead to cardiac tamponade. The strengths of our approach were that we had two previously well children who developed almost similar symptoms after being diagnosed with anti-CFH antibody aHUS and improved with PLEX along with immunosuppressants. We were able to rule out the common causes of pericardial effusion including infection, vasculitis and hypothyroidism in the first case. Further, our 10-year-old boy had developed cardiac tamponade when renal function had decreased and urine output had increased ruling out complication of uremia and anasarca. At the same time, he was showing hemolysis on peripheral blood smear suggesting that disease may have relapsed. Finally, he improved with the addition of rituximab. The limitations, however, were that we were unable to see the trend of anti-CFH antibody titres, Troponin I levels and also did not do genetic testing or pericardial tissue biopsy.
HUS predominantly affects the kidney. However, involvement of other organ systems can occur as pathogenesis involves complement activation. Hepatitis, seizures and gastrointestinal involvement are common extra-renal manifestations. On the other hand, cardiac involvement in HUS is rare. Among the cardiac manifestations, a varied spectrum of myocardial involvement like myocarditis, myocardial infarction, cardiac failure or dilated cardiomyopathy has been reported. [2–4]
Pericardial involvement in the form of effusion or tamponade has been previously reported in diarrhea-associated HUS, though not in aHUS. Brandt et al., Birk et al. and Mohammad et al. have described pediatric STEC associated HUS presenting with pericardial effusion. In the retrospective study of 37 children of STEC HUS, Brandt observed that 5 children had pericardial effusion. The case reported by Birk reported that cardiac tamponade was the terminal event in the 6-year-old girl. [5–7] However, occurrence of pericardial effusion in these cases has not been sufficiently explained.
It has been opined that pericardial effusion could be due to direct cardiac involvement by active disease or a secondary manifestation due to over-hydration, electrolyte imbalances and hypertension. It has also been proposed to be a reactive process associated with myocardial involvement [4–7]. It is difficult to retrospect whether our patients also had myocardial injury as we did not do any Troponin I level or tissue biopsy. Nonetheless, it could not have been explained by uremia, fluid overload or hypertension as these were taken care of. Isolated pericardial involvement due to primary damage remains a possibility. Further, the literature suggests that aHUS with anti-CHF antibody HUS, those with mutations in CFB, CHI and C3 have more chance of developing cardiac complication [4].
Both our patients had hypertension, which is a known clinical feature of aHUS. Hypertension in these cases is usually due to vascular disease and volume expansion [3].
Conclusion
Cardiac involvement in the form of pericardial effusion may be seen in aHUS. Thus, we emphasize that cardiac involvement as an extra- renal manifestation in pediatric HUS needs a special consideration. Routine screening to evaluate cardiac involvement is mandatory as early detection and prompt management can be crucial in reducing morbidity and mortality.
Author contributions
SG was involved in the collection of patient details, framing the rough draft of the manuscript and approved the final draft. RR was involved in care of patient, added intellectual content and approved the final draft. AR helped extensively in analyzing the complement changes in both the children, added intellectual content and approved the final draft. RN supported with a good pathological correlation to the disease process, added intellectual content and approved the final draft. SHN confirmed the pericardial effusion by ECHO and management of the patients, added intellectual content and approved the final draft. LD was involved in care of patient, added intellectual content and approved the final draft. AK supported with a good pathological correlation to the disease process, added intellectual content and approved the final draft. KT managed these cases, supervised in collection of data, added intellectual content and approved the final draft.
Funding
None.
Compliance with ethical standards
Conflict of interest
All authors declare that they have no conflict of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Due permission from the Departmental Review Board was taken.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Fakhouri F, Zuber J, Frémeaux-Bacchi V, Loirat C. Haemolytic uraemic syndrome. Lancet. 2017;390(10095):681–696. doi: 10.1016/S0140-6736(17)30062-4. [DOI] [PubMed] [Google Scholar]
- 2.Khalid M, Andreoli S. Extrarenal manifestations of the hemolytic uremic syndrome associated with Shiga toxin-producing Escherichia coli (STEC HUS) Pediatr Nephrol. 2019;34(12):2495–2507. doi: 10.1007/s00467-018-4105-1. [DOI] [PubMed] [Google Scholar]
- 3.Raina R, Krishnappa V, Blaha T, et al. Atypical hemolytic-uremic syndrome: an update on pathophysiology, diagnosis, and treatment. TherApher Dial. 2019;23(1):4–21. doi: 10.1111/1744-9987.12763. [DOI] [PubMed] [Google Scholar]
- 4.Noris M, Remuzzi G. Cardiovascular complications in atypical haemolytic uraemic syndrome. Nat Rev Nephrol. 2014;10(3):174–180. doi: 10.1038/nrneph.2013.280. [DOI] [PubMed] [Google Scholar]
- 5.Brandt JR, Fouser LS, Watkins SL, Zelikovic I, Tarr PI, NazarStewart V, Avner ED. Escherichia coli O 157:H7-associated hemolytic-uremic syndrome after ingestion of contaminated hamburgers. J Pediatr. 1994;125:519–526. doi: 10.1016/S0022-3476(94)70002-8. [DOI] [PubMed] [Google Scholar]
- 6.Mohammed J, Filler G, Price A, Sharma AP. Cardiac tamponade in diarrhoea-positive haemolytic uraemic syndrome. Nephrol Dial Transplant. 2009;24(2):679–681. doi: 10.1093/ndt/gfn649. [DOI] [PubMed] [Google Scholar]
- 7.Birk PE, Chakrabarti S, Lacson AG, Ogborn MR. Cardiac tamponade as a terminal event in the hemolytic uremic syndrome in childhood. Pediatr Nephrol. 1994;8(6):754–755. doi: 10.1007/BF00869112. [DOI] [PubMed] [Google Scholar]



