Clinical Review Lesson of the week

Depot corticosteroid treatment for hay fever causing avascular necrosis of both hips

BMJ 2001; 322 doi: http://dx.doi.org/10.1136/bmj.322.7302.1589 (Published 30 June 2001) Cite this as: BMJ 2001;322:1589
  1. S M S Nasser, senior registrara,
  2. P W Ewan, consultantb (Pamela.ewan{at}addenbrookes.nhs.uk)
  1. a Department of Medicine, University of Cambridge Clinical School, Addenbrooke's Hospital, Cambridge CB2 2QQ
  2. b Department of Allergy and Clinical Immunology, Addenbrooke's Hospital
  1. Correspondence to: P W Ewan
  • Accepted 8 March 2001

Consider immunotherapy rather than depot corticosteroid treatment for hay fever

The prevalence of hay fever (seasonal allergic rhinitis) has increased dramatically over the past four decades in the United Kingdom, with a doubling from 10 per 1000 people in the 1970s to 20 per 1000 in the 1980s.1 The British Society for Allergy and Clinical Immunology has published guidelines on the management of rhinitis that recommend that, when measures for avoiding pollen are insufficient, treatment should be either oral or topical antihistamine, to be taken as required, or topical sodium cromoglycate, to be taken regularly.2 If these treatments are ineffective, intranasal corticosteroids should be taken regularly, with oral or topical antihistamines added if required. If this stepped programme is adhered to the vast majority of patients will respond, but if patients fail to respond systemic corticosteroids may be used.

Over the last few decades the practice of treating severe symptoms with depot corticosteroids has become common, mainly because the duration of treatment of 4–6 weeks is often sufficient to cover the whole season. If the season is exceptionally severe or the patient experiences prolonged symptoms as a result of multiple allergies (for example, a combination of tree and grass pollen allergy), more than one depot corticosteroid injection may be required. Oral corticosteroids are less commonly used, perhaps because of the inconvenience of taking daily medication for 6–8 weeks every year.

The United Kingdom is unusual in the developed world in that it has minimal provision of allergy services, hence only small numbers of patients with severe seasonal allergic rhinitis have access to immunotherapy despite the growing evidence for the efficacy of this treatment. 3 4 We report a case of a man who developed avascular necrosis of both hip joints after depot corticosteroids were used to treat his severe hay fever.

Case report

A 42 year old man of Sri Lankan origin developed hay fever at 21 years of age, but his symptoms became severe and disabling only at the age of 27. Every year in mid-April he developed uvular swelling, palatal irritation, profuse rhinorrhoea, incessant sneezing, nasal obstruction, and conjunctivitis. Symptoms continued until the end of July, peaking in June, consistent with tree and grass pollen allergy. His symptoms were so severe that he was unable to work, and he had to sleep with a towel over his pillow. For the past eight years he had also developed seasonal asthma that started in early July and lasted for 2–3 weeks. He had used antihistamines and a number of topical nasal corticosteroids, including beclometasone, fluticasone, and budesonide, but without relief of symptoms; therefore his general practitioner resorted to depot corticosteroids. From 1987 to 1997 he was given 16 injections of depot corticosteroid: nine of Kenalog (triamcinolone; Bristol-Myers Squibb, Hounslow) 40 mg and seven of Depo-Medrone (methylprednisolone; Pharmacia & Upjohn, Milton Keynes) 80 mg. After each injection his symptoms were substantially relieved for up to six weeks. He had no history of atopy or asthma outside the grass pollen season. He was a non-smoker, did not drink alcohol, and had no history of substance abuse or previous musculoskeletal problems.

Fig 1
Fig 1

Plain radiograph of the hip joints, showing severe avascular necrosis on the right, with collapse of the femoral head associated with underlying lysis and sclerosis. The acetabular joint surface is normal, confirming that this is not a degenerative (osteoarthritic) process. Mild avascular necrosis is seen on the left, with an area of lysis surrounded by a diffuse area of sclerosis

In August 1997 he developed a limp, with pain in his right hip, and a few months later developed similar symptoms in his left hip. Plain radiographs showed abnormalities in both hip joints (figure 1). A magnetic resonance scan arranged by an orthopaedic surgeon showed avascular necrosis of the superior segment of the right femoral head, with early avascular necrosis of the left femoral head (figure 2). It was thought that he would require bilateral total hip replacement at some stage in the future.

Fig 2
Fig 2

Magnetic resonance image (T1 weighted) of the hips. The entire femoral head is reduced in signal on the right, indicating diffuse oedema around the necrotic segment. On the left, a line of low signal surrounds an area of normal signal, indicating a previous avascular insult. The normal signal within the lesion corresponds to normal fat and indicates that the marrow at this site is still viable

Comment

A man with severe hay fever was given at least one depot corticosteroid injection each year for 11 years, leading to avascular necrosis of both femoral heads. Depot corticosteroids are widely used in general practice for severe cases and at the patient's request. Avascular necrosis, an uncommon though serious complication of corticosteroid treatment, has not previously been described in hay fever or with the use of depot corticosteroids.

Avascular necrosis of the hip is associated with corticosteroid therapy, but there is no direct relation with the total corticosteroid dosage.5 Avascular necrosis has been reported to develop after as little as two weeks of corticosteroid treatment, 6 7 but susceptibility factors remain elusive. In our case it is difficult to know whether the pharmocokinetics of a depot preparation contributed to avascular necrosis of the femoral head. Avascular necrosis affects patients in their 30s or early 40s, corticosteroids being the leading cause, although alcohol and substance misuse have also been implicated. Patients who develop advanced stages of avascular necrosis of the femoral head at a young age must undergo total hip replacement, which carries a poor long term prognosis.

Specialist advice for hay fever

Hay fever is common and can be debilitating, as in this case, and we suggest that patients whose hay fever is poorly controlled despite treatment with topical corticosteroids and antihistamines should be referred at an early stage to an allergy centre for specialist advice. This patient was eventually referred at age 41 years, when allergy to grass and tree pollens was confirmed by skin tests, and an extremely high grass pollen specific IgE concentration (325 kU/l (normal: <0.35 kU/l)) was found. He is currently undergoing grass pollen immunotherapy, which is showing clear benefit.

A recent article advised against the use of depot corticosteroids in hay fever and advised that corticosteroids could be given orally but only in special circumstances and only for short periods at the lowest effective dose.8 However, if conventional drug therapy fails, patients with hay fever should be referred to an allergist and considered for pollen immunotherapy. In seasonal allergic rhinitis (grass and birch pollen) the efficacy of immunotherapy has been confirmed in a number of carefully controlled studies. 3 4 915 Immunotherapy consists of a series of injections with incremental doses of allergen extract and is a specific treatment for allergic disease that can alter the course of the allergy. A recent study showed that three years of grass pollen immunotherapy in carefully selected patients had benefit for at least three years after discontinuation.4

Allergy services need improving

In 1986 the Committee on the Safety of Medicines expressed concerns about 25 deaths over 30 years from severe bronchospasm and anaphylaxis in patients receiving immunotherapy (desensitisation). These deaths were almost exclusively in patients with asthma. Furthermore, the injections were given in general practice by people without experience of immunotherapy or an understanding of the potential side effects and without monitoring of the patient. It is now recommended that allergen immunotherapy should be given only by experienced doctors and where facilities are available for resuscitation and monitoring of patients for at least one hour after the injection. In view of the safety issues, patients with multiple allergies, chronic asthma, or other serious medical diseases are excluded from this treatment. 2 16 Seasonal asthma is not a contraindication, provided that the patient is asymptomatic outside the pollen season.

It was a failure in management that this patient was not referred to an allergist sooner. The lack of consultant allergists and good quality allergy services has led to the poor availability of immunotherapy in the United Kingdom (about 1000 patients are currently being treated). This is in sharp contrast to the rest of Europe, especially Scandinavia, and the United States, where immunotherapy is used widely. The newer vaccines, which are more potent (a higher content of the major allergen(s)) and biologically standardised, are particularly effective. Immunotherapy is never without risk, but the risks are low if treatment is undertaken in specialised allergy clinics by staff experienced in treating large numbers of patients. More allergists are urgently needed in the United Kingdom. Better provision of allergy services will be cost effective and save NHS funds, as exemplified by this patient.

Acknowledgments

We thank George Bentley of the Royal National Orthopaedic Hospital, Stanmore, London, for kind permission to use the plain radiographs and magnetic resonance images. We are grateful to Philip Bearcroft of the radiology department at Addenbrooke's Hospital for his expert interpretation of these images.

Contributors: PWE and SMSN both came up with the idea of writing the paper. SMSN undertook the literature review. The paper was written jointly by both authors. PWE will act as guarantor for the paper.

Footnotes

  • Competing interests None declared.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.