Progress
SpO2 was 97% on 2L/min oxygen through nasal cannula. On day 1 of admission, blood tests showed a slightly increased total white blood cell count, due mainly to eosinophilia (1.4 x 109/mm3). Renal and liver functions were impaired. HbsAg was positive.

 

Table 1. Investigation results

	D1	D3	D4	D7	D14	D21
Hb, g/dL	12.7	12.9	12.7	13.6	13.1	8.5
WBC	11.9	12	15.8	17.8	22	21.6
N	8.7	9.8	13.5	15		20.9
L	1.0	0.8	0.8	0.6		0.6
E	1.4	0.7	0.3	0		0
M	0.8	0.7		1.3		0
Platelet	125	131	142	145	84	49
Na, mmol/L	140	131	132	128	127	127
K, mmol/L	3.9	3.9	4.2	4.7	4.5	4.6
Ur, mmol/L	19.5	16.3	17.4	16	27	27.3
Cr, umol/L	266	268	282	201	257	366
Alb	23	21		23	16
Globulin	33	34		30	32
Bilirubin	15	22		12	13
Alkaline phosphatase	120	107		106	275
SGOT
SGPT	26	27		29	62
Urate		0.52
Oxygen					62%
pH	7.376	7.341	7.295		7.304
PCO2	4.48	4.43	4.18		4.83
PO2	11.8	8.39	11.07		11.96
HCO3	19.7	17.9	15.2		18
BE	-4.1	-6.4	-9.8		-7.3
O2 sat	96.6	91.1	95.3		96.1

 

He was on the following medications: allopurinol, indapamide 2.5mg om, glibenclamide 5mg om, metformin 500mg bd, and atenolol 50mg om. He had been started on allopurinol 100mg three times daily nine months before this admission by a general practitioner for tophaceous gout. Dosage was reduced to 100mg daily after development of the skin lesion a month ago. After admission, allopurinol was stopped on the suspicion that the rash was allopurinol-induced and the skin rash managed conservatively.

On day 3 after admission, the patient progressively deteriorated with increasing oxygen and respiratory distress to 5L/min O2 via nasal cannula. Bronchoscopic lavage of the right upper lobe showed the following differential leucocyte count: 20% eosinophils, 25% polymorphs, 10% lymphocytes, and 55% macrophages. No bacteria, fungal element or malignant cells were found. Transbronchial lung biopsy was not done because of the oxygen desaturation. Stool for parasites was negative. This patient was therefore treated as allopurinol hypersensitivity syndrome, with features of exfoliative dermatitis and erythema multiforme, peripheral and pulmonary eosinophilia, with underlying renal impairment. Because of the oxygen desaturation with the findings of pulmonary eosinophilia, intravenous methylprednisolone 30mg Q12H.

High resoulution CT thorax on Day 8 confirmed the presence of peripherally situated lung infiltrates in the upper lung fields. Significant amount of artifacts were present because of the unstable respiratory status. Skin biopsy taken on Day 6 showed interface dermatitis: The epidermis showed parakeratosis, scattered apoptosis and vacuolar degeneration of the basal layer. Mild inflammatory infiltrate and pigmentory incontinence was noted in the superficial dermis. There was no well-formed bullous lesion or malignancy. The features were consistent with fixed drug eruption or erythema multiforme. Anti-nuclear factor was negative. Anti-DNA was 1:32 IU/ml, C3 0.67, C4 0.23, C-reactive protein 44, ANCA negative, Anti-HIV antibodies negative.

The skin rash showed gradual improvement after allopurinol was withheld. The peripheral count also showed a decreasing trend, even before steroid was started. However, the patient continued to run a downhill course. Sputum culture grew methicillin-sensitive Staphylococcus aureus. Piperacillin-tazobactam was given. On Day 8, he developed further respiratory distress and required intubation and mechanical ventilation. Cheat X-ray showed worsening consolidation. Oxygen requirement increased to 95%. Serial tracheal aspirate for culture showed 000. Renal function continued to deteriorate and continuous renal replacement therapy was started on D20. Despite aggressive treatment, he developed sudden cardiac arrest on Day 21 and died.

Paramortem lung, liver and renal biopsies were taken. The lung biopsy showed patchy mild septal fibrosis and focal type II pneumocyte hyperplasia. Occasional foamy alveolar macrophages were noted. Scanty interstitial mononuclear cellular infiltration were seen. There was no increased alveolar eosinophils, vasculitis, pulmonary haemorrhage, granuloma or malignancy. These changes were non-specific. Liver biopsy showed hepatitis B associated liver cirrhosis. Renal biopsy showed diabetic glomrulosclerosis together with hypertensive changes. Immunofluorescence study showed non-specific staining for immunoglobulins and C3 only.

 

Diagnosis
1. Allopurinol Hypersensitivity Syndrome - allopurinol-induced exfoliate dermatitis with pulmonary eosinophilia, complicated with MSSA chest infection and multi-organ failure
2. Underlying renal impairment from glomerulosclerosis due to diabetes and hypertension
3. Underlying hepatitis B-associated liver cirrhosis

 
Discussion
Allopurinol (4-hydroxypyrazole (3,4,-d)pyrimidine), is an analog of xanthine. Its metabolite, oxypurinol, binds and inhibits xanthine oxidase. Xanthine oxidase is the enzyme responsible for converting hypothanine to xanthine and to uric acid. In patients with normal renal function, oxypurinol is cleared by the kidneys, with a half-life of 14 to 26 hours (1,2). Clearance of oxypurinol is decreased, and chance of severe adverse reactions is increased, if there is concomitant renal insufficiency and with the use of thiazide diuretics (3). Our patient happened to be taking high dose of allopurinol (300mg per day), together with indapamide (natrilix), and having impaired renal function on presentation. Indapamide is not a thiazide diuretic, though the action is similar. It is a sulfonamide diuretic, and differs structurally from the thiazide diuretics by the presence of a methylindoline ring rather than a thiazide ring (4).

Our patient fulfilled the criteria of Allopurinol Hypersensivity Syndrome (AHS) proposed by Singer and Wallace in 1986 (5).

Table. Criteria for the Allopurinol Hypersensitivity Syndrome

1. A clear history of exposure to allopurinol
2. A clinical picture including
a. At least 2 of the following major criteria:
i. worsening renal function
ii. acute hepatocellular injury
iii. a rash including either toxic epidermal necrosis, erythema multiforme, or a diffuse exanthematour or exfoliative dermatitis
b. or 1 of the major criteria plus at least 1 of the following minor criteria:
i. fever
ii. eosinophilia
iii. leukocytosis
3. Lack of exposure to another drug that may have caused a similar picture

 

Although peripheral eosinophilia has been well reported in the literature as one of the features of Allopurinol Hypersensitivity Syndrome, pulmonary eosinophilia has not been reported to our knowledge. Post-mortem could not reveal eosinophilic infiltration in our case, possibly due to withholding of the drug and the administration of corticosteroid. The time course, with associated features like biopsy-proven drug-induced skin rash, and the exclusion of parasitic infection and pulmonary fungal infection by bronchoalveolar lavage, all point to the conclusion that allopurinol was the culprit of the pulmonary eosinophilia.

 
The exact mechanism of allopurinol hypersensitivity is not clear. A proposed mechanism is that renal insufficiency predisposes to oxypurinol accumulation, which results in an immunologic reaction against the metabolite, cross-reacting with purines, ribonucleoproteins, and nucleic acids. This reaction causes tissue injury with release of cellular antigens, and leads to further immunological reaction and a viscous circle of further tissue injury (5).

  
Cell-mediated immunity has been suspected because most patients developed hypersensitivity only after 4 - 6 weeks after drug initiation, and tissue biopsies of affected organs show significant lymphocyte and granulomatous infiltraton. Cell-mediated immunity directed toward allopurinol and more importantly to its oxypurinol metabolite is involved in the pathogenesis of allopurinol-induced hypersensitivity, on the finding that oxypurinol-stimulated lymphocytes from the hypersensitive patient demonstrated enhanced expression of activation antigens compared to unstimulated lymphocytes (6).

 
Causes of pulmonary eosinophilia include parasitic infection (roundworms, Toxocara larvae, filariae), drugs (e.g., penicillin, aminosalicylic acid, hydralazine, nitrofurantoin, chlorpropamide, sulfonamides), chemical sensitizers (e.g., nickel carbonyl inhaled as a vapor), and fungi (e.g., Aspergillus fumigatus, which causes allergic bronchopulmonary aspergillosis). However, most eosinophilic pneumonia is of unknown etiology.

 
The mechanism of drug-induced eosinophilic pneumonia has not been fully worked out. Eosinophilia may suggest a type I hypersensitivity reaction. However, for allopurinol hypersensitivity syndrome, the mechanism is probably type III and possibly type IV. In review of previous cases of allopurinol hypersensitivity, there has not been a single case of immediate shock or bronchospasm, but rather, symptoms mainly involved skin rash which developed subacutely, with a mean time interval between initiation of drug to appearance of about 21 days to 25 days (7,8). 
 

References

1. Elion GB, Kovensky A, Hitching GH, Metz E, Rundles RW: Metabolic studies of allopurinol, an inhibitor of xanthine oxidase. Biochem Pharmacol 1966; 15: 863 - 880
2. Hande KR, Reed E, Chabner BA: Allopurinol kinetics. Clin Pharmacol Ther 1978; 23: 598 - 605
3. Young JL, Boswell RB, Nies AS: Severe allopurinol hypersensitivity: association with thazides and prior renal compromise. Arch Intern Med 1974; 134: 553 - 558
4. Chaffman M, Heel RC, Brogden TM et al. Indapamide, a review of its pharmacodynamic properties and therapeutic efficacy in hypertension. Drugs. 1984; 28:189-35.
5. Arellano F, Sacristán JA. Allopurinol hypersensitivity syndrome: a review. Ann Pharmacother. 1993 Mar;27(3):337-43. Review.
6. Braden GL, Warzynski MJ, Golightly M, Ballow M.Cell-mediated immunity in allopurinol-induced hypersensitivity. Clin Immunol Immunopathol. 1994 Feb;70(2):145-51.
7. Singer JZ, Wallace SL. The allopurinol hypersensitivity syndrome: unnecessary morbidity and mortality. Arthritis and Rheum 1986; 29: 82 - 7
8. Khoo BP, Leow YH. A review of inpatients adverse drug reactions to allopurinol. Sing Med J 2000; 41(4): 156 - 160