Paroxysmal cold haemoglobinuria leading to acute renal failure: A case report

K Das and D Hughes

Renal Unit, Royal Hospital for Sick Children, Dalnair Street, Glasgow G3 8SJ.

SMJ 2008 53(1): 60

Abstract

Paroxysmal cold haemoglobinuria (PCH) is a form of autoimmune haemolytic anaemia (AIHA) characterised by sudden onset of haemoglobinuria either spontaneously or following exposure to cold. It was previously believed to be a rare form of AIHA, but is now thought to be responsible for up to 40% of cases of acute transient haemolytic anaemia in children¹ following viral infections. Although the anaemia can be severe, illness is characteristically of short duration and does not recur. Published reports of PCH in children leading to renal impairment are rare, with supportive management only being used in affected cases.^{2, 3} We describe the case of a 2-year-old boy who developed acute renal failure (ARF) secondary to PCH, requiring peritoneal dialysis, where the early institution of plasma exchange was associated with rapid recovery of renal function.

Case Report

A previously well 2-year-old boy presented with a short history of fever, pallor and lethargy. There had been red-brown urine staining in his nappy. There was no history of diarrhoea. He received Co-amoxiclav for presumed urinary tract infection. No urine was available for examination. He presented to the local emergency department at 24 hours with persisting fever, lethargy and anuria. He was pale and jaundiced. Initial investigations demonstrated renal impairment and a haemolytic anaemia (see Table 1). The platelet count was normal. A blood film was not examined. A provisional diagnosis of haemolytic uraemic syndrome was reached and he was transferred to our institution after 6 hours.

On arrival, the patient was pale and jaundiced. He was tachycardic, but normotensive, well perfused and hydrated, and with no fluid retention. Investigations confirmed a progressive haemolytic anaemia with a normal platelet count (Table 1). Blood film showed microcytic red cells with mild polychromasia and no red cell fragments. The direct antiglobulin test (DAT) was positive and there was evidence of agglutination and complement (C3) coating with no immunoglobulin at 37ºC. The direct Donath-Landsteiner (DL) test was positive with anti-P antigen specificity confirming a diagnosis of paroxysmal cold haemoglobinuria. His renal function had deteriorated further. He remained anuric. A renal ultrasound scan was not performed at the time.

Peritoneal dialysis (PD) and central venous access was established. He initially received two blood transfusions via a blood warmer (37ºC) to maintain his Hb above 7.0g/dL. He was matched for packed red cells that were P1 antibody negative, CMV negative, irradiated and from a <10-hour pool. Continuous cycling peritoneal dialysis was initiated. Despite transfusions there was continuing haemolysis with a falling haemoglobin level over the next 48 hours. The plasma was heavily discoloured by free haemoglobin. He remained anuric. Plasma exchange was commenced, with 100% volume exchange with 4.5% human albumin solution as replacement. After two plasma exchanges in 24 hours laboratory markers of continuing haemolysis resolved and no further blood transfusions were needed.

Renal biochemical indices and urine output began to improve from day 3. A renal biopsy was not performed. Peritoneal dialysis was discontinued after 2 days and the patient returned to his local hospital 11 days after initial presentation, by which time his renal function had returned to normal.

Other investigations documented blood and stools to be negative for verotoxin producing E. coli and other pathogens; serology for EBV, CMV and mycoplasma were also negative. A standard viral screen for influenza virus, parainfluenza virus, respiratory syncytial virus and adenovirus were negative. ASOT was less than 200; ANA and ANCA negative; C3 and C4 reduced.

Table 1. Laboratory results during the acute phase of illness

^a Initial presentation at local hospital.

^b On presentation to ourselves.

Discussion

We present a case of a child with PCH and ARF where early recognition of the diagnosis and introduction of plasma exchange was associated with a rapid recovery from dialysis dependence. Two recently published cases describe acute renal impairment in children with PCH.^{2, 3} The first published case report was of a 5-year-old with a history of diarrhoea, vomiting, dehydration and ‘cola’ coloured urine. The development of a haemolytic anaemia was associated with oliguria, raised creatinine, normal platelets and a positive direct Coombs test. Direct and indirect DL test confirmed the diagnosis of PCH. Fresh frozen plasma had been administered on presentation and forced diuresis with urinary alkalinisation after diagnosis was made. Haemolysis settled over 48 hours. Dialysis was not required and renal function had returned to normal after 11 days. The second child was 2-year-old with coryzal symptoms and “black stools”. There were features of a haemolytic anaemia with DAT strongly positive, normal platelets, biochemical changes of acute renal impairment and haemoglobinuria. Peritoneal dialysis was required. Renal biopsy revealed pigmentary changes of haemoglobinuria within tubules without tubulo-interstitial nephritis or acute tubular necrosis. Glomeruli were normal. Further management was with IV methylprednisolone, for possible auto-immune haemolytic anaemia, and blood transfusion support. Direct DL test was negative but the indirect test was positive confirming the diagnosis of PCH. Dialysis was required for 11 days.

Previously published cases have been reported of ARF with paroxysmal nocturnal haemoglobinuria (PNH).^{4, 5} This is a rare condition characterised by chronic intravascular haemolysis. It is believed to be secondary to a membrane abnormality of erythrocytes susceptible to complement-mediated haemolysis. The histology in these cases showed features of acute tubular necrosis (ATN), with extensive mild dilatation of proximal tubules and focal tubular cell necrosis. Heavy haemosiderin deposition was seen mostly within the proximal tubular cells. In PNH it is believed that this does not lead to an acute deterioration of renal function on its own this but rather causes progressive renal insufficiency.⁶

In PCH the Donath-Landsteiner (DL) antibody responsible for haemolysis is a cold biphasic IgG antibody specific for the glycosphingolipid globoside P antigen in the red blood cell membrane.⁷ The antibody binds to red cells at lower temperatures in the peripheral circulation. When the red cells return to the warmer core temperature, complement is activated leading to cell membrane lysis, intravascular haemolysis and a massive release of haemoglobin to the circulation.

A combination of mechanisms is thought to be responsible for the development of ARF with this massive, acute haemolysis. Free haemoglobin is directly toxic to renal tubules⁸ causing proximal tubular epithelial necrosis. This toxicity is dose dependent, and therefore occurs when haemolysis is sudden and massive. Haem proteins can also cause intrarenal vasoconstriction by scavenging the potent vasodilator, nitric oxide, in the renal microcirculation.^{9, 10} Haemoglobin casts plug tubules, and the toxic damage to tubular epithelium may lead to accumulation of necrotic cells in the lumina, further exacerbating tubular occlusion and ARF.

Our patient presented with clinical and laboratory features of acute, severe and rapidly progressive haemolysis. The cause of his acute renal impairment was felt likely to be the massive haemolysis leading to haemoglobinuria obstructing tubular lumina.³ There was no history of drug ingestion and no evidence of other causes of acute glomerulonephritis. A renal biopsy was deferred, initially while the severe haemolytic anaemia was corrected, and subsequently when rapid renal recovery ensued.

With the initiation of plasma exchange after early diagnosis of PCH, the plasma removed was noted to be extremely dark due to the haemoglobin release from the massive intravascular haemolysis. With the second exchange the plasma was much clearer. His haemoglobin stabilised, LDH levels fell and bilirubin cleared. The use of plasma exchange has been reported in post-infectious AIHA due to the Donath-Landsteiner (DL) antibody, resulting in resolution of haemolysis.¹¹ The DL antibody is of moderately low titre and is produced only transiently in post viral illness. It magnifies its effect by repeatedly detaching and rebinding to other red blood cells, and can therefore sensitise a large amount of red cells to complement. Institution of plasma exchange in the initial stage may therefore help to remove these antibodies, decreasing the duration and magnitude of haemolysis and reducing exposure to repeated blood transfusions. We speculate that the plasma exchange hastened the recovery of renal function in our patient by reducing exposure of renal tubules to the large free haemoglobin load.

In summary, we report a case of paroxysmal cold haemoglobinuria associated with the Donath Landsteiner antibody resulting in massive haemolysis and acute renal failure in a two-year-old boy. Early introduction of plasma exchange was associated with rapid resolution of anuric renal failure. The use of plasma exchange should be considered in children with PCH associated acute renal failure.

References

1.   Sokol RJ, Hewitt S, Stamps BK, et al. Autoimmune haemolysis in childhood and adolescence. Acta Haematologica. 1984; 72:245-257.

2. Vergara LH, Mota MCC, Sarmento AG, et al. Insuficiencia renal aguda secundaria a hemoglobinuria paroxística al frío. An Pediatr (Barc) 2006; 64:267-269.  

3. Hothi DK, Bass P, Morgan M, et al. Acute renal failure in a patient with paroxysmal cold hemoglobinuria. Pediatr Nephrol. 2007; 22(4):593-6. 

4. Santamaria R, Espinosa M, Ortega R, et al. Acute renal failure in a patient  with myelodysplastic syndrome and paroxysmal nocturnal haemoglobinuria phenotype. Nefrologia. 2004; 24 Suppl 3:56-60.

5. Chow KM, Lai FM, Wang AY, et al. Reversible renal failure in paroxysmal nocturnal hemoglobinuria. Am J Kidney Dis. 2001 Feb; 37(2):E17.

6. Zachee P, Henckens M, Van Damme B, et al. Chronic renal failure due to renal hemosiderosis in a patient with paroxysmal nocturnal hemoglobinuria. Clin Nephrol. 1993; 39:28-31.

7. Wickramasinghe SN, McCullough J. Blood and bone marrow pathology: Chapter 10- Acquired hemolytic anemia. New York, Churchill Livingstone, 2002.

8. Chan WL, Tang NL, Yim CC, et al. New features of renal lesion induced by stroma free hemoglobin. Toxicol Pathol. 2000; 28:635-642.

9. Braun SR, Weiss FR, Keller AI, et al. Evaluation of the renal toxicity of heme proteins and their derivatives: A role in genesis of acute tubular necrosis. J Exp Med. 1970; 131:443-460. 

10. Zager RA, Gamelin LM. Pathogenetic mechanisms in experimental  hemoglobinuric acute renal failure. Am J Physiol. 1989; 256:F446-455. 

11. Roy-Burman A, Glader BE. Resolution of severe Donath-Landsteiner autoimmune hemolytic anemia temporally associated with institution of plasmapheresis. Crit Care Med. 2002; 30(4):931-934.

Back to February Contents